Renewable-Energy-Driven Future: Technologies, Modelling, Applications, Sustainability and Policies 9780128205396

Table of contents :
Front Cover
Renewable-Energy-Driven Future
Copyright Page
Contents
List of Contributors
I. Technologies
1 Solar energy technologies: principles and applications
1.1 Introduction
1.2 Photovoltaic technologies
1.2.1 Solar photovoltaic principles
1.2.1.1 Power of a solar cell
1.2.1.2 Fill factor
1.2.1.3 Conversion efficiency
1.2.2 Recent advancements in solar photovoltaic technologies
1.2.2.1 Perovskite solar cells
1.2.2.2 Other emerging photovoltaic technologies
1.2.2.3 Cadmium telluride
1.2.2.4 Copper indium gallium selenide
1.2.2.5 Dye-sensitized solar cells
1.2.2.6 Quantum dot solar cells
1.2.3 Applications of solar cells
1.3 Solar thermal collectors
1.3.1 Stationary collectors
1.3.2 Tracking concentrating collectors
1.4 Solar cooling technologies
1.4.1 Solar photovoltaic powered cooling system
1.4.1.1 Solar vapour compression cooling system
1.4.1.2 Solar thermoelectric cooling system
1.4.1.3 Solar ground source heat pump system
1.4.2 Solar thermal powered cooling system
1.4.2.1 Solar sorption cooling system
1.4.2.2 Solar desiccant cooling system
1.4.2.3 Solar ejector cooling system
1.5 Solar pond
1.6 Solar cooking
1.7 Solar desalination
1.7.1 Indirect type desalination
1.7.1.1 Humidification and dehumidification desalination
1.7.1.2 Multistage flash desalination
1.7.1.3 Vapour compression desalination
1.7.1.4 Osmotic desalination driven by solar energy
1.7.2 Direct type desalination
Nomenclature
References
2 Bioenergy for better sustainability: technologies, challenges and prospect
2.1 Introduction
2.2 Technologies
2.2.1 Microorganisms
2.2.2 Feedstocks
2.2.3 Fermentation technologies
2.3 Challenges
2.4 Future prospects
References
3 Organic Rankine cycle driven by geothermal heat source: life cycle techno-economic–environmental analysis
3.1 Introduction
3.2 Organic Rankine cycle system description and working fluid selection
3.3 Methods and models
3.3.1 Thermodynamic and technical analysis
3.3.2 Heat exchanger model
3.3.3 Economic and exergoeconomc analysis
3.3.4 Life-cycle environmental analysis
3.3.4.1 Life-cycle boundary
3.3.4.2 Carbon footprint analysis
3.3.4.3 Data sources
3.3.5 Multicriteria integrated assessment and decision-making
3.4 Thermodynamic and economic results
3.4.1 Effects of design parameters on thermodynamic performance
3.4.2 Effects of design parameters on economic performance
3.4.3 Effects of design parameters on exergoeconomic performance
3.4.4 Sensitivity analysis on the economic performance and inlet temperature of geothermal source
3.5 Life-cycle and carbon footprint analysis of the organic Rankine cycle
3.5.1 Environmental evaluation of life cycle
3.5.2 Environmental evaluation of components
3.5.3 Environmental evaluation of working fluids
3.5.4 Analysis of emission reductions
3.5.5 Sensitivity analysis
3.6 Comparison between different layouts of organic Rankine cycle systems
3.7 Results of multifactor evaluation
3.8 Conclusions
Appendix A
References
4 Renewable energy based trigeneration systems—technologies, challenges and opportunities
4.1 Introduction
4.2 Cogeneration and trigeneration
4.2.1 Trigeneration systems classification
4.2.1.1 Classification by size
4.2.1.2 Classification by applications
4.2.1.3 Classification by type of prime-mover
4.2.1.4 Classification by sequence of energy
4.2.2 Microgeneration
4.2.3 Polygeneration
4.2.4 Distributed/decentralized energy system
4.2.5 District energy systems and polygeneration microgrids
4.2.6 Combined cooling, heating and power operation strategies (modes)
4.2.7 Energy tools/software used in energy systems
4.3 Heat-recovery units
4.3.1 Types of heat-recovery units
4.3.1.1 Unfired units
4.3.1.2 Fired units
4.3.2 Heat pumps
4.4 Cooling technologies
4.4.1 Types of cooling technologies
4.4.1.1 Sorption technology
4.4.1.2 Desiccant technology
4.4.2 Cooling applications in trigeneration systems
4.5 Thermal energy storage
4.5.1 Storage concept
4.5.1.1 Active system
4.5.1.2 Passive system
4.5.2 Storage mechanisms/types of thermal energy storage
4.5.2.1 Sensible heat storage
4.5.2.2 Latent heat storage
4.5.2.3 Chemical storage
4.5.3 Combined heat storage
4.5.4 Packed bed systems
4.5.5 Solar thermal energy storage
4.6 Renewable energy
4.6.1 Hybrid energy systems
4.6.1.1 Zero energy building
4.6.2 Wind energy
4.6.2.1 Wind power meteorology and wind modelling
4.6.2.2 Turbine technology
4.6.2.3 Wind hybrid systems and applications
4.6.2.3.1 Wind–diesel system
4.6.2.3.2 Wind–photovoltaic-hydrogen system
4.6.2.3.3 Seawater desalination
4.6.2.4 Wind power development
4.6.3 Geothermal energy technologies
4.6.4 Biomass energy
4.6.4.1 Biomass energy technologies
4.6.4.2 Biofuels
4.6.4.2.1 Straight vegetable oils
4.6.4.2.2 Biodiesel
4.6.4.2.3 Bioethanol
4.6.4.2.4 Biomethanol
4.6.4.2.5 Biogas
4.6.4.3 Biomass-fuelled combined cooling, heating and power systems
4.6.5 Solar energy
4.6.5.1 Solar collectors
4.6.5.1.1 Nonconcentrating solar collectors
4.6.5.1.2 Concentrating solar collectors
4.6.5.2 Solar photovoltaic systems
4.6.5.3 Hybrid photovoltaic-thermal systems
4.6.5.4 Solar thermal applications
4.6.5.5 Solar-renewable hybrids
4.6.5.5.1 High-renewable hybrids
Concetrating solar plant-biomass hybrids
Concetrating solar plant-geothermal hybrids
Concetrating solar plant-wind hybrids
4.6.5.5.2 Medium-renewable hybrids
4.6.5.5.3 Low renewable hybrids
Solar-Brayton cycles
Solar-aided coal power plants (Rankine cycle)
Integrated solar combined cycles
4.6.6 Other renewable sources
4.7 Research trends in renewable energy integrated trigeneration technologies
4.8 Challenges and opportunities in renewable energy-based trigeneration systems
4.8.1 Challenges and barriers
4.8.2 Opportunities and prospects
4.9 Conclusions
Abbreviations
References
Further reading
5 Integrated power transmission and distribution systems
5.1 Introduction
5.2 Mathematical model
5.2.1 First-stage unit commitment model
5.2.2 Second-stage economic dispatch model
5.2.3 Distributed energy resource management problem
5.2.4 Tighter formulations
5.3 Numerical results
5.3.1 Isolated unit commitment problem
5.3.2 Isolated distributed energy resource management problem
5.3.3 Integrated transmission and distribution systems
5.3.3.1 Sensitivity analyses: types of the integrated distribution systems
5.3.4 IEEE 118-bus network results
5.4 Conclusions
References
II. Modelling
6 Integrated inexact optimization for hybrid renewable energy systems
6.1 Introduction
6.2 Deterministic optimization techniques
6.2.1 Classical techniques
6.2.2 Metaheuristic algorithm
6.2.3 Commercial software
6.3 Inexact mathematical programming methods
6.3.1 Stochastic mathematical programming
6.3.1.1 Chance-constrained programming
6.3.1.2 Stochastic programming with recourse
6.3.2 Robust optimization
6.3.3 Fuzzy mathematical programming
6.3.3.1 Fuzzy flexible programming
6.3.3.2 Fuzzy possibilistic programming
6.3.3.3 Fuzzy robust programming
6.3.4 Interval mathematical programming
6.3.5 Hybrid inexact mathematical programming
6.4 Integrated inexact optimization framework
6.5 Conclusions
References
7 Large-scale integration of variable renewable resources
7.1 Introduction
7.2 Climate change and greenhouse gas emissions trends
7.3 Global renewable power deployment
7.4 High penetration of renewable sources in the power sector
7.4.1 Optimal development of nondispatchable resources (solar and wind)
7.4.2 Surplus and backup powers—curtailment
7.4.3 Energy storage
7.4.3.1 Pumped-storage hydropower
7.4.3.2 Batteries
7.4.3.3 Hydrogen
7.5 Main strategies for the 2030 European energy transition
7.5.1 Coal phase-out
7.5.2 Decrease in renewable energy costs
7.5.2.1 Evolution of levelized cost of energy on renewable sources
7.5.3 International interconnections
7.5.4 Digitalization and smart grids
7.5.5 Demand response
Acknowledgements
References
8 The climate and economic benefits of developing renewable energy in China
8.1 Introduction
8.2 Methods and scenarios
8.2.1 Integrated model of energy, environment and economy for sustainable development/computable general equilibrium model
8.2.2 Economic assessment of renewable energy
8.2.3 Investment in nonfossil power generation
8.2.4 Data sources
8.2.5 Scenarios
8.2.5.1 Reference scenario
8.2.5.2 REmax scenario
8.3 Results
8.3.1 Macroeconomic trends towards 2050
8.3.2 Impacts on the energy system
8.3.2.1 Primary energy
8.3.2.2 Power structure
8.3.3 Benefits of developing renewable energy in carbon and air pollutant emissions reduction
8.3.4 Economic impacts of renewable energy development
8.3.4.1 Investment
8.3.4.2 Impacts on industrial output, value-added and employment
8.4 Discussion
8.4.1 Policy implications
8.4.2 Comparison with other studies
8.4.3 Sensitivity analysis
8.4.4 Limitations and next step
8.5 Conclusions
References
III. Applications
9 The utilization of renewable energy for low-carbon buildings
9.1 Building and energy and environmental challenges
9.2 Net-zero energy building and low-carbon building
9.3 Building life-cycle systems and greenhouse gas emissions
9.4 Renewable energy technologies for low-carbon buildings
9.4.1 Building material extraction and transportation
9.4.2 Building construction
9.4.3 Building operation
9.4.3.1 Solar photovoltaics
9.4.3.2 Solar thermal
9.4.3.3 Photovoltaic–thermal
9.5 Path forward for advancing low-carbon buildings
References
10 Towards a renewable-energy-driven district heating system: key technology, system design and integrated planning
10.1 Introduction
10.2 Key technologies and system design for renewable-energy-driven district heating
10.2.1 Indicators and design principle for enhancement of district heating systems
10.2.1.1 Energy efficiency and exergy efficiency
10.2.1.2 Cascade and upgrade use of heat energy
10.2.2 System design and key technologies of renewable-energy-driven district heating system
10.2.2.1 System composition of a renewable-energy-driven district heating system
10.2.2.2 Key technologies for a renewable-energy-driven district heating system
10.2.2.2.1 Energy conversion
10.2.2.2.2 Heat distribution
10.2.2.2.3 Heat storage
10.2.3 Optimization for a renewable-energy-driven district heating system
10.2.3.1 Supply side optimization
10.2.3.2 Demand-side management
10.2.3.2.1 Demand response
10.2.3.2.2 Building mix
10.2.3.2.3 Land use change
10.3 Integrated urban planning for renewable-energy-based district heating
10.3.1 Urban and industrial symbiosis
10.3.2 Modelling the strategic urban renewal for promoting district heating
10.4 Conclusions
Acknowledgements
References
11 Renewable energy-driven desalination for more water and less carbon
11.1 Introduction
11.2 Desalination technology
11.2.1 Thermal desalination techniques
11.2.1.1 Multieffect distillation
11.2.1.2 Multistage flash desalination
11.2.1.3 Vapour compression desalination
11.2.1.4 Adsorption desalination
11.2.2 Membrane desalination techniques
11.2.2.1 Reverse osmosis
11.2.2.2 Electrodialysis
11.2.2.3 Forward osmosis
11.2.3 Desalination installed capacity and trends
11.2.3.1 Global status of desalination
11.2.3.2 Research trends in desalination
11.3 Energy and desalination
11.3.1 Renewable energy resources for desalination
11.4 Renewable energy integrated desalination: technical, economic and social development aspects
11.4.1 Solar desalination
11.4.1.1 Solar photovoltaic desalination
11.4.1.2 Solar thermal desalination
11.4.2 Nuclear energy-driven desalination
11.4.3 Wind energy-driven desalination
11.4.4 Geothermal energy-driven desalination
11.4.5 Ocean/wave energy-driven desalination
11.5 Barriers, issues and opportunities in desalination technology development
11.5.1 Brine production
11.5.2 Desalination cost and CO2 emissions
11.6 Outlook
Abbreviations
References
IV. Sustainability
12 The environmental performance of hydrogen production pathways based on renewable sources
12.1 Introduction
12.2 H2 production pathways and applications
12.2.1 Water electrolysis
12.2.2 Biomass to H2
12.2.2.1 Thermal gasification
12.2.2.2 Supercritical water gasification of biomass
12.2.2.3 Bio-oil reforming
12.3 Method
12.3.1 Life cycle assessment
12.3.2 Goal and scope definition
12.3.3 Inventory analysis of wind-based water electrolysis
12.3.4 Inventory analysis of solar-based water electrolysis
12.3.5 Inventory analysis of the thermal gasification of biomass
12.3.5.1 Feedstock production
12.3.5.2 Biomass transportation
12.3.5.3 Gasification process
12.3.6 Inventory analysis of bio-oil reforming
12.3.7 Inventory analysis of supercritical water gasification of algae
12.3.7.1 Algae cultivation
12.3.7.2 Process conversion
12.3.8 Sensitivity and uncertainty analyses
12.4 Greenhouse gas footprints of H2 production pathways
12.4.1 Greenhouse gas footprint of water electrolysis
12.4.2 Greenhouse gas footprint of gasification
12.4.3 Greenhouse gas footprint of bio-oil reforming
12.4.4 Greenhouse gas footprint of supercritical water gasification
12.4.5 Comparative assessment incorporating sensitivity and uncertainty analyses
12.5 Conclusions
Acknowledgements
References
13 Integrated economic–environmental–social assessment of straw for bioenergy production
13.1 Introduction
13.2 Methods
13.2.1 Estimation of straw available for energy production
13.2.1.1 Influential factors of grain yield
13.2.1.2 Energy potential of straw
13.2.2 Cost and profit of straw utilization for energy production
13.2.3 Environmental impacts of straw utilization for energy production
13.2.4 Selection of evaluation indicators
13.3 Case study
13.3.1 Estimation of the quantity of straw
13.3.1.1 Regional grain yield
13.3.1.2 Conversion coefficients of straw
13.3.2 Parameters of energy conversion technologies
13.4 Results and discussion
13.4.1 Energy potential of straw
13.4.2 Energy, environmental and socioeconomic benefits of straw utilization
13.4.3 Analysis of major factors affecting the results
13.4.3.1 Changes in collection radius
13.4.3.2 Changes in purchase price of straw
13.4.3.3 Changes in utilization proportion of straw
13.5 Discussion
13.6 Conclusions
Subscripts and superscripts
References
14 Sustainability assessment of renewable energy-based hydrogen and ammonia pathways
14.1 Introduction
14.1.1 Importance of energy storage
14.1.2 Chemical energy storage
14.1.2.1 Renewable hydrogen (H2)
14.1.2.2 Renewable ammonia (NH3)
14.2 Hydrogen and ammonia production pathways
14.2.1 Hydrogen production
14.2.1.1 Steam methane reforming
14.2.1.2 Wind power-based electrolysis
14.2.1.3 Hydropower-based electrolysis
14.2.1.4 Photoelectrochemical water splitting
14.2.2 Ammonia production
14.2.2.1 Steam methane reforming and Haber–Bosch ammonia synthesis method
14.2.2.2 Wind power-based electrolysis and Haber–Bosch ammonia synthesis process
14.2.2.3 Hydropower-based electrolysis and the Haber–Bosch ammonia synthesis
14.2.2.4 Photoelectrochemical water splitting and electrochemical ammonia synthesis
14.3 Methodology
14.3.1 Efficiency index
14.3.1.1 Energy efficiency
14.3.1.2 Exergy efficiency
14.3.2 Cost
14.3.3 Environmental impact
14.3.4 Weighting scheme
14.4 Results and discussion
14.5 Conclusions
Acknowledgements
Nomenclature
Abbreviations
Greek letters
Subscripts
References
15 An extended fuzzy divergence measure-based technique for order preference by similarity to ideal solution method for ren...
15.1 Introduction
15.2 Prerequisites
15.3 Divergence measures for fuzzy sets
15.3.1 An example for developed fuzzy divergence measures
15.4 Divergence measures-based fuzzy TOPSIS method
15.4.1 Case study of renewable energy investment
15.5 Conclusions
Appendix: Proof of the properties
References
16 Multicriteria decision making for the selection of the best renewable energy scenario based on fuzzy inference system
16.1 Introduction
16.2 Method
16.3 Application
16.4 Conclusions
References
V. Policy
17 How much is possible? An integrative study of intermittent and renewables sources deployment. A case study in Brazil
17.1 Introduction – understanding of the question
17.2 Irresistible expansion
17.2.1 Wind
17.2.2 Solar
17.3 Undesirable effects of the intermittent renewable resources expansion
17.3.1 Complexity
17.3.2 The operation problem with the increasing insertion of intermittent renewable resources
17.3.3 Economic effects
17.3.4 Externalities and the merit order effect
17.4 Rebound effect – social acceptance of intermittent renewable sources – the opponents
17.5 Conclusions
17.6 Acknowledgments
References
18 Renewable energy technologies: barriers and policy implications
18.1 Introduction
18.2 Literature on barriers to renewable energy
18.3 Barriers identification and policy frameworks
18.3.1 Economic barriers
18.3.2 Technical barriers
18.3.3 Awareness and information barriers
18.3.4 Financial barriers
18.3.5 Regulatory and policy barriers
18.3.6 Institutional and administrative barriers
18.3.7 Social and environmental barriers
18.3.8 End-use/demand-side barriers
18.4 Barriers identification framework
18.4.1 Selection of renewable energy technologies for the study of barriers
18.4.2 Identification of barriers for the study
18.5 Measures to overcome barriers
18.5.1 Renewable energy targets
18.5.2 Renewable energy promotion measures
18.5.2.1 Support mechanisms
18.5.2.1.1 A feed-in tariff
18.5.2.1.2 Auctions or tendering schemes
18.5.2.1.3 Renewable energy certificates
18.5.2.1.4 Renewable portfolio standard
18.5.3 Net metering/net billing
18.5.3.1 Fiscal incentives
18.5.3.2 Public financing of renewable energy
18.6 Current challenges
References
19 Policies for a sustainable energy future: how do renewable energy subsidies work and how can they be improved?
19.1 Introduction
19.2 Renewable energy development and renewable energy subsidies
19.2.1 The development of renewable energy varies across countries
19.2.2 A brief review of the renewable energy subsidy policies in United States
19.2.3 A brief review of the renewable energy subsidy policies in European Union
19.2.3.1 Germany
19.2.3.2 Spain
19.2.3.3 Denmark
19.2.4 A brief review of renewable energy subsidies in China
19.3 The mechanism of how renewable energy subsidy works
19.3.1 A model of renewable energy generation
19.3.1.1 Government
19.3.1.2 Electricity generation enterprises
19.3.2 Discussion and policy implications
19.4 Conclusions
References
20 Renewable energy-based power generation and the contribution to economic growth: the case of Portugal
20.1 Introduction
20.2 Methodology
20.2.1 Econometric model and data
20.2.2 Testing for unit roots and detecting outliers
20.2.3 Testing for cointegration and estimating parameters
20.3 Empirical results
20.4 Conclusions
Appendix 1
References
Index
Back Cover

Citation preview

Renewable-Energy-Driven Future Technologies, Modelling, Applications, Sustainability and Policies

This page intentionally left blank

Renewable-EnergyDriven Future Technologies, Modelling, Applications, Sustainability and Policies

Edited by

Jingzheng Ren Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, China

Academic Press is an imprint of Elsevier 125 London Wall, London EC2Y 5AS, United Kingdom 525 B Street, Suite 1650, San Diego, CA 92101, United States 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom Copyright © 2021 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress

ISBN: 978-0-12-820539-6 For Information on all Academic Press publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Brian Romer Acquisitions Editor: Lisa Reading Editorial Project Manager: Chiara Giglio Production Project Manager: Sojan P. Pazhayattil Cover Designer: Victoria Pearson Typeset by MPS Limited, Chennai, India

Contents List of Contributors

xvii

Part I Technologies 1.

Solar energy technologies: principles and applications D. Sakthivadivel, K. Balaji, D. Dsilva Winfred Rufuss, S. Iniyan and L. Suganthi 1.1 Introduction 1.2 Photovoltaic technologies 1.2.1 Solar photovoltaic principles 1.2.2 Recent advancements in solar photovoltaic technologies 1.2.3 Applications of solar cells 1.3 Solar thermal collectors 1.3.1 Stationary collectors 1.3.2 Tracking concentrating collectors 1.4 Solar cooling technologies 1.4.1 Solar photovoltaic powered cooling system 1.4.2 Solar thermal powered cooling system 1.5 Solar pond 1.6 Solar cooking 1.7 Solar desalination 1.7.1 Indirect type desalination 1.7.2 Direct type desalination Nomenclature References

2.

3 5 6 11 18 18 20 21 23 25 28 32 33 34 34 37 38 39

Bioenergy for better sustainability: technologies, challenges and prospect Senqing Fan, Jingyun Liu, Xiaoyu Tang and Zeyi Xiao 2.1 Introduction 2.2 Technologies 2.2.1 Microorganisms 2.2.2 Feedstocks

43 45 45 48

v

vi

Contents

2.2.3 Fermentation technologies 2.3 Challenges 2.4 Future prospects References

3.

54 61 62 63

Organic Rankine cycle driven by geothermal heat source: life cycle techno-economic
environmental analysis Chao Liu, Shukun Wang and Jingzheng Ren 3.1 Introduction 3.2 Organic Rankine cycle system description and working fluid selection 3.3 Methods and models 3.3.1 Thermodynamic and technical analysis 3.3.2 Heat exchanger model 3.3.3 Economic and exergoeconomc analysis 3.3.4 Life-cycle environmental analysis 3.3.5 Multicriteria integrated assessment and decision-making 3.4 Thermodynamic and economic results 3.4.1 Effects of design parameters on thermodynamic performance 3.4.2 Effects of design parameters on economic performance 3.4.3 Effects of design parameters on exergoeconomic performance 3.4.4 Sensitivity analysis on the economic performance and inlet temperature of geothermal source 3.5 Life-cycle and carbon footprint analysis of the organic Rankine cycle 3.5.1 Environmental evaluation of life cycle 3.5.2 Environmental evaluation of components 3.5.3 Environmental evaluation of working fluids 3.5.4 Analysis of emission reductions 3.5.5 Sensitivity analysis 3.6 Comparison between different layouts of organic Rankine cycle systems 3.7 Results of multifactor evaluation 3.8 Conclusions Appendix A References

4.

68 70 73 73 76 76 81 84 85 85 89 92 95 99 99 103 104 105 106 108 112 118 118 122

Renewable energy based trigeneration systems— technologies, challenges and opportunities Deepesh Sonar 4.1 Introduction

126

Contents

4.2 Cogeneration and trigeneration 4.2.1 Trigeneration systems classification 4.2.2 Microgeneration 4.2.3 Polygeneration 4.2.4 Distributed/decentralized energy system 4.2.5 District energy systems and polygeneration microgrids 4.2.6 Combined cooling, heating and power operation strategies (modes) 4.2.7 Energy tools/software used in energy systems 4.3 Heat-recovery units 4.3.1 Types of heat-recovery units 4.3.2 Heat pumps 4.4 Cooling technologies 4.4.1 Types of cooling technologies 4.4.2 Cooling applications in trigeneration systems 4.5 Thermal energy storage 4.5.1 Storage concept 4.5.2 Storage mechanisms/types of thermal energy storage 4.5.3 Combined heat storage 4.5.4 Packed bed systems 4.5.5 Solar thermal energy storage 4.6 Renewable energy 4.6.1 Hybrid energy systems 4.6.2 Wind energy 4.6.3 Geothermal energy technologies 4.6.4 Biomass energy 4.6.5 Solar energy 4.6.6 Other renewable sources 4.7 Research trends in renewable energy integrated trigeneration technologies 4.8 Challenges and opportunities in renewable energy-based trigeneration systems 4.8.1 Challenges and barriers 4.8.2 Opportunities and prospects 4.9 Conclusions Abbreviations References Further reading

5.

vii 127 128 130 130 130 132 133 133 134 134 135 135 135 135 136 136 136 137 137 138 138 139 139 141 142 144 148 149 154 154 156 158 160 161 168

Integrated power transmission and distribution systems Abolhassan Mohammadi Fathabad, Jianqiang Cheng and Kai Pan 5.1 Introduction 5.2 Mathematical model 5.2.1 First-stage unit commitment model 5.2.2 Second-stage economic dispatch model

169 174 175 177

viii

Contents

5.2.3 Distributed energy resource management problem 5.2.4 Tighter formulations 5.3 Numerical results 5.3.1 Isolated unit commitment problem 5.3.2 Isolated distributed energy resource management problem 5.3.3 Integrated transmission and distribution systems 5.3.4 IEEE 118-bus network results 5.4 Conclusions References

178 183 184 185 190 192 195 197 198

Part II Modelling 6.

Integrated inexact optimization for hybrid renewable energy systems Y. Zhou and Z.X. Zhou 6.1 Introduction 6.2 Deterministic optimization techniques 6.2.1 Classical techniques 6.2.2 Metaheuristic algorithm 6.2.3 Commercial software 6.3 Inexact mathematical programming methods 6.3.1 Stochastic mathematical programming 6.3.2 Robust optimization 6.3.3 Fuzzy mathematical programming 6.3.4 Interval mathematical programming 6.3.5 Hybrid inexact mathematical programming 6.4 Integrated inexact optimization framework 6.5 Conclusions References

7.

203 204 205 206 207 207 208 212 214 218 219 220 222 223

Large-scale integration of variable renewable resources R Go´mez-Calvet, A.R. Go´mez-Calvet and J.M. Mart´ınez-Duart 7.1 7.2 7.3 7.4

Introduction Climate change and greenhouse gas emissions trends Global renewable power deployment High penetration of renewable sources in the power sector 7.4.1 Optimal development of nondispatchable resources (solar and wind) 7.4.2 Surplus and backup powers—curtailment 7.4.3 Energy storage

233 234 238 239 241 243 244

Contents

7.5 Main strategies for the 2030 European energy transition 7.5.1 Coal phase-out 7.5.2 Decrease in renewable energy costs 7.5.3 International interconnections 7.5.4 Digitalization and smart grids 7.5.5 Demand response Acknowledgements References

8.

ix 247 247 248 251 253 253 254 254

The climate and economic benefits of developing renewable energy in China Hancheng Dai, MD. Shouquat Hossain and Xiaorui Liu 8.1 Introduction 8.2 Methods and scenarios 8.2.1 Integrated model of energy, environment and economy for sustainable development/computable general equilibrium model 8.2.2 Economic assessment of renewable energy 8.2.3 Investment in nonfossil power generation 8.2.4 Data sources 8.2.5 Scenarios 8.3 Results 8.3.1 Macroeconomic trends towards 2050 8.3.2 Impacts on the energy system 8.3.3 Benefits of developing renewable energy in carbon and air pollutant emissions reduction 8.3.4 Economic impacts of renewable energy development 8.4 Discussion 8.4.1 Policy implications 8.4.2 Comparison with other studies 8.4.3 Sensitivity analysis 8.4.4 Limitations and next step 8.5 Conclusions References

257 259

259 260 261 264 264 265 265 267 271 273 277 277 278 279 279 282 283

Part III Applications 9.

The utilization of renewable energy for low-carbon buildings Yuan Chang and Yayin Wei 9.1 Building and energy and environmental challenges 9.2 Net-zero energy building and low-carbon building

289 290

x

Contents

9.3 Building life-cycle systems and greenhouse gas emissions 9.4 Renewable energy technologies for low-carbon buildings 9.4.1 Building material extraction and transportation 9.4.2 Building construction 9.4.3 Building operation 9.5 Path forward for advancing low-carbon buildings References

292 292 292 295 296 305 307

10. Towards a renewable-energy-driven district heating system: key technology, system design and integrated planning Yi Dou, Lu Sun, Minoru Fujii, Yasunori Kikuchi, Yuichiro Kanematsu and Jingzheng Ren 10.1 Introduction 10.2 Key technologies and system design for renewable-energydriven district heating 10.2.1 Indicators and design principle for enhancement of district heating systems 10.2.2 System design and key technologies of renewableenergy-driven district heating system 10.2.3 Optimization for a renewable-energy-driven district heating system 10.3 Integrated urban planning for renewable-energy-based district heating 10.3.1 Urban and industrial symbiosis 10.3.2 Modelling the strategic urban renewal for promoting district heating 10.4 Conclusions Acknowledgements References

311 314 314 318 321 324 324 326 328 329 329

11. Renewable energy-driven desalination for more water and less carbon Aamir Mehmood and Jingzheng Ren 11.1 Introduction 11.2 Desalination technology 11.2.1 Thermal desalination techniques 11.2.2 Membrane desalination techniques 11.2.3 Desalination installed capacity and trends 11.3 Energy and desalination 11.3.1 Renewable energy resources for desalination

333 336 337 339 340 345 345

xi

Contents

11.4 Renewable energy integrated desalination: technical, economic and social development aspects 347 11.4.1 Solar desalination 347 11.4.2 Nuclear energy-driven desalination 350 11.4.3 Wind energy-driven desalination 353 11.4.4 Geothermal energy-driven desalination 354 11.4.5 Ocean/wave energy-driven desalination 357 11.5 Barriers, issues and opportunities in desalination technology development 357 11.5.1 Brine production 359 11.5.2 Desalination cost and CO2 emissions 359 11.6 Outlook 361 Abbreviations 361 References 362

Part IV Sustainability 12. The environmental performance of hydrogen production pathways based on renewable sources Eskinder Demisse Gemechu and Amit Kumar 12.1 Introduction 12.2 H2 production pathways and applications 12.2.1 Water electrolysis 12.2.2 Biomass to H2 12.3 Method 12.3.1 Life cycle assessment 12.3.2 Goal and scope definition 12.3.3 Inventory analysis of wind-based water electrolysis 12.3.4 Inventory analysis of solar-based water electrolysis 12.3.5 Inventory analysis of the thermal gasification of biomass 12.3.6 Inventory analysis of bio-oil reforming 12.3.7 Inventory analysis of supercritical water gasification of algae 12.3.8 Sensitivity and uncertainty analyses 12.4 Greenhouse gas footprints of H2 production pathways 12.4.1 Greenhouse gas footprint of water electrolysis 12.4.2 Greenhouse gas footprint of gasification 12.4.3 Greenhouse gas footprint of bio-oil reforming 12.4.4 Greenhouse gas footprint of supercritical water gasification 12.4.5 Comparative assessment incorporating sensitivity and uncertainty analyses 12.5 Conclusions Acknowledgements References

376 377 379 380 384 384 384 385 386 386 391 392 394 395 395 395 397 398 399 401 401 401

xii

Contents

13. Integrated economic
environmental
social assessment of straw for bioenergy production Junnian Song, Kexin Li and Wei Yang 13.1 Introduction 13.2 Methods 13.2.1 Estimation of straw available for energy production 13.2.2 Cost and profit of straw utilization for energy production 13.2.3 Environmental impacts of straw utilization for energy production 13.2.4 Selection of evaluation indicators 13.3 Case study 13.3.1 Estimation of the quantity of straw 13.3.2 Parameters of energy conversion technologies 13.4 Results and discussion 13.4.1 Energy potential of straw 13.4.2 Energy, environmental and socioeconomic benefits of straw utilization 13.4.3 Analysis of major factors affecting the results 13.5 Discussion 13.6 Conclusions Subscripts and superscripts References

407 410 410 411 413 414 414 416 419 421 421 421 425 428 429 430 431

14. Sustainability assessment of renewable energy-based hydrogen and ammonia pathways Yusuf Bicer and Farrukh Khalid 14.1 Introduction 14.1.1 Importance of energy storage 14.1.2 Chemical energy storage 14.2 Hydrogen and ammonia production pathways 14.2.1 Hydrogen production 14.2.2 Ammonia production 14.3 Methodology 14.3.1 Efficiency index 14.3.2 Cost 14.3.3 Environmental impact 14.3.4 Weighting scheme 14.4 Results and discussion 14.5 Conclusions Acknowledgements Nomenclature Abbreviations Greek letters

435 435 436 439 439 440 448 448 449 450 452 453 466 466 466 467 467

Contents

Subscripts References

xiii 467 467

15. An extended fuzzy divergence measure-based technique for order preference by similarity to ideal solution method for renewable energy investments Pratibha Rani, Arunodaya Raj Mishra, Abbas Mardani, Fausto Cavallaro, Raghunathan Krishankumar and Dalia Streimikiene 15.1 Introduction 15.2 Prerequisites 15.3 Divergence measures for fuzzy sets 15.3.1 An example for developed fuzzy divergence measures 15.4 Divergence measures-based fuzzy TOPSIS method 15.4.1 Case study of renewable energy investment 15.5 Conclusions Appendix: Proof of the properties References

469 472 475 479 480 481 485 486 488

16. Multicriteria decision making for the selection of the best renewable energy scenario based on fuzzy inference system Jingzheng Ren, Yi Man, Ruojue Lin and Yue Liu 16.1 Introduction 16.2 Method 16.3 Application 16.4 Conclusions References

Part V Policy

491 493 496 504 505

509

17. How much is possible? An integrative study of intermittent and renewables sources deployment. A case study in Brazil Fernando Amaral de Almeida Prado, Jr 17.1 Introduction
understanding of the question 17.2 Irresistible expansion 17.2.1 Wind 17.2.2 Solar 17.3 Undesirable effects of the intermittent renewable resources expansion

511 515 516 516 516

xiv

Contents

17.3.1 Complexity 17.3.2 The operation problem with the increasing insertion of intermittent renewable resources 17.3.3 Economic effects 17.3.4 Externalities and the merit order effect 17.4 Rebound effect
social acceptance of intermittent renewable sources
the opponents 17.5 Conclusions Acknowledgments References

517 521 526 529 533 535 535 535

18. Renewable energy technologies: barriers and policy implications Jyoti Prasad Painuly and Norbert Wohlgemuth 18.1 Introduction 18.2 Literature on barriers to renewable energy 18.3 Barriers identification and policy frameworks 18.3.1 Economic barriers 18.3.2 Technical barriers 18.3.3 Awareness and information barriers 18.3.4 Financial barriers 18.3.5 Regulatory and policy barriers 18.3.6 Institutional and administrative barriers 18.3.7 Social and environmental barriers 18.3.8 End-use/demand-side barriers 18.4 Barriers identification framework 18.4.1 Selection of renewable energy technologies for the study of barriers 18.4.2 Identification of barriers for the study 18.5 Measures to overcome barriers 18.5.1 Renewable energy targets 18.5.2 Renewable energy promotion measures 18.5.2.1 Support mechanisms 18.5.3 Net metering/net billing 18.5.3.1 Fiscal incentives 18.5.3.2 Public financing of renewable energy 18.6 Current challenges References

539 541 544 545 546 547 547 548 548 548 549 550 553 553 554 554 555 555 557 558 558 558 560

19. Policies for a sustainable energy future: how do renewable energy subsidies work and how can they be improved? Qin Bao, Jiali Zheng and Shouyang Wang 19.1 Introduction

563

Contents

19.2 Renewable energy development and renewable energy subsidies 19.2.1 The development of renewable energy varies across countries 19.2.2 A brief review of the renewable energy subsidy policies in United States 19.2.3 A brief review of the renewable energy subsidy policies in European Union 19.2.4 A brief review of renewable energy subsidies in China 19.3 The mechanism of how renewable energy subsidy works 19.3.1 A model of renewable energy generation 19.3.2 Discussion and policy implications 19.4 Conclusions References

xv

567 567 570 571 575 577 577 580 581 582

20. Renewable energy-based power generation and the contribution to economic growth: the case of Portugal Joa˜o Paulo Cerdeira Bento 20.1 Introduction 20.2 Methodology 20.2.1 Econometric model and data 20.2.2 Testing for unit roots and detecting outliers 20.2.3 Testing for cointegration and estimating parameters 20.3 Empirical results 20.4 Conclusions Appendix 1 References Index

587 589 589 591 591 592 599 599 606 609

This page intentionally left blank

List of Contributors Fernando Amaral de Almeida Prado, Jr Sinerconsult Consultoria Treinamento e Participac¸o˜es Limitada, Sa˜o Paulo, Brazil K. Balaji Vellore Institute of Technology, Vellore, Tamil Nadu, India Qin Bao Center for Forecasting Science, Chinese Academy of Sciences, University of Chinese Academy of Sciences, Beijing, P.R. China Joa˜o Paulo Cerdeira Bento GOVCOPP - Research Unit in Governance, Competitiveness and Public Policy, Department of Economics, Management, Industrial Engineering and Tourism, University of Aveiro, Campus of Santiago, Aveiro, Portugal Yusuf Bicer Division of Sustainable Development (DSD), Hamad Bin Khalifa University (HBKU), Qatar Foundation (QF), Doha, Qatar Fausto Cavallaro Department of Economics, University of Molise, Campobasso, Italy Yuan Chang School of Management Science and Engineering, Central University of Finance and Economics, Beijing, P.R. China Jianqiang Cheng Department of Systems and Industrial Engineering, University of Arizona, Tucson, AZ, United States Hancheng Dai School of Environmental Science and Engineering, Peking University, Bejing, P.R. China Yi Dou Presidential Endowed Chair for “Platinum Society”, The University of Tokyo, Tokyo, Japan D. Dsilva Winfred Rufuss Vellore Institute of Technology, Vellore, Tamil Nadu, India Senqing Fan School of Chemical Engineering, Sichuan University, Chengdu, P.R. China Abolhassan Mohammadi Fathabad Department of Systems and Industrial Engineering, University of Arizona, Tucson, AZ, United States Minoru Fujii Center for Social and Environmental Systems Research, National Institute for Environmental Studies (NIES), Tsukuba, Japan Eskinder Demisse Gemechu Department of Mechanical Engineering, 10-263 Donadeo Innovation Centre for Engineering, University of Alberta, Edmonton, AB, Canada A.R. Go´mez-Calvet Finance Department, Facultad de Econom´ıa. Avda, Tarongers s/n, Valencia, Spain

xvii

xviii

List of Contributors

R. Go´mez-Calvet Business Department, Faculty of Social Sciencies, Universidad Europea de Valencia, Valencia, Espan˜a MD. Shouquat Hossain School of Environmental Science and Engineering, Peking University, Bejing, P.R. China S. Iniyan College of Engineering Guindy, Anna University, Chennai, Tamil Nadu, India Yuichiro Kanematsu Presidential Endowed Chair for “Platinum Society”, The University of Tokyo, Tokyo, Japan Farrukh Khalid Division of Sustainable Development (DSD), Hamad Bin Khalifa University (HBKU), Qatar Foundation (QF), Doha, Qatar Yasunori Kikuchi Presidential Endowed Chair for “Platinum Society”, The University of Tokyo, Tokyo, Japan; Institute for Future Initiatives, The University of Tokyo, Tokyo, Japan Raghunathan Krishankumar School of Computing, SASTRA University, Thanjavur, India Amit Kumar Department of Mechanical Engineering, 10-263 Donadeo Innovation Centre for Engineering, University of Alberta, Edmonton, AB, Canada Kexin Li College of New Energy and Environment, Jilin University, Changchun, P.R. China Ruojue Lin Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China Chao Liu Key Laboratory of Low-Grade Energy Utilization Technologies and Systems, Ministry of Education, School of Energy and Power Engineering, Chongqing University, Chongqing, P.R. China Jingyun Liu School of Chemical Engineering, Sichuan University, Chengdu, P.R. China Xiaorui Liu School of Environmental Science and Engineering, Peking University, Bejing, P.R. China Yue Liu Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China Yi Man Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China Abbas Mardani Department of Marketing, College of Business Administration, University of South Florida, Tampa, FL United States J.M. Mart´ınez-Duart Departamento de F´ısica Aplicada, C-XII, Universidad Auto´noma de Madrid, Campus de Cantoblanco, Madrid, Spain Aamir Mehmood Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China Arunodaya Raj Mishra Department of Mathematics, Government College Jaitwara, Satna, India Jyoti Prasad Painuly UNEP DTU Partnership, Technical University of Denmark, Copenhagen, Denmark

List of Contributors

xix

Kai Pan Department of Logistics and Maritime Studies, Hong Kong Polytechnic University, Kowloon, Hong Kong SAR, China Pratibha Rani Department of Mathematics, National Institute of Technology, Warangal, India Jingzheng Ren Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China D. Sakthivadivel Vellore Institute of Technology, Vellore, Tamil Nadu, India Deepesh Sonar Department of Mechanical Engineering, Government Polytechnic College, Ujjain, India Junnian Song College of New Energy and Environment, Jilin University, Changchun, P.R. China Dalia Streimikiene Lithuanian Energy Institute, Kaunas, Lithuania L. Suganthi College of Engineering Guindy, Anna University, Chennai, Tamil Nadu, India Lu Sun Center for Social and Environmental Systems Research, National Institute for Environmental Studies (NIES), Tsukuba, Japan Xiaoyu Tang Biogas Institute of Ministry of Agriculture, Chengdu, P.R. China Shouyang Wang Center for Forecasting Science, Chinese Academy of Sciences, University of Chinese Academy of Sciences, Beijing, P.R. China Shukun Wang Key Laboratory of Low-Grade Energy Utilization Technologies and Systems, Ministry of Education, School of Energy and Power Engineering, Chongqing University, Chongqing, P.R. China Yayin Wei School of Management Science and Engineering, Central University of Finance and Economics, Beijing, P.R. China Norbert Wohlgemuth Department of Economics, University of Klagenfurt, Klagenfurt, Austria Zeyi Xiao School of Chemical Engineering, Sichuan University, Chengdu, P.R. China Wei Yang College of New Energy and Environment, Jilin University, Changchun, P.R. China Jiali Zheng Center for Forecasting Science, Chinese Academy of Sciences, University of Chinese Academy of Sciences, Beijing, P.R. China Y. Zhou Key Laboratory for City Cluster Environmental Safety and Green Development of the Ministry of Education, Institute of Environmental and Ecological Engineering, Guangdong University of Technology, Guangzhou, P.R. China Z.X. Zhou Key Laboratory for City Cluster Environmental Safety and Green Development of the Ministry of Education, Institute of Environmental and Ecological Engineering, Guangdong University of Technology, Guangzhou, P.R. China

This page intentionally left blank

Part I

Technologies

This page intentionally left blank

Chapter 1

Solar energy technologies: principles and applications D. Sakthivadivel1, K. Balaji1, D. Dsilva Winfred Rufuss1, S. Iniyan2 and L. Suganthi2 1

Vellore Institute of Technology, Vellore, Tamil Nadu, India, 2College of Engineering Guindy, Anna University, Chennai, Tamil Nadu, India

Chapter Outline 1.1 Introduction 3 1.2 Photovoltaic technologies 5 1.2.1 Solar photovoltaic principles 6 1.2.2 Recent advancements in solar photovoltaic technologies 11 1.2.3 Applications of solar cells 18 1.3 Solar thermal collectors 18 1.3.1 Stationary collectors 20 1.3.2 Tracking concentrating collectors 21 1.4 Solar cooling technologies 23

1.1

1.4.1 Solar photovoltaic powered cooling system 1.4.2 Solar thermal powered cooling system 1.5 Solar pond 1.6 Solar cooking 1.7 Solar desalination 1.7.1 Indirect type desalination 1.7.2 Direct type desalination Nomenclature References

25 28 32 33 34 34 37 38 39

Introduction

The Sun is the primary source of sustenance for all living and nonliving things on this planet earth. Solar energy is the solitary renewable energy source with immense potential of yearly global insolation at 5600 ZJ [1], as compared to other sources such as biomass and wind. The Sun is a large, radiant spherical unit of hot gas which is composed of hydrogen (B70%), helium (B28%) and remaining are carbon (C), nitrogen (N2) and oxygen (O2) up to 1.5% and the other gases of 0.5% such as neon (Ne), iron (Fe), silicon (Si), magnesium (Mg) and sulphur (S). The process of fusion helps to power the Sun and the stars. When two atoms of hydrogen join together or in other words fuse, they form an atom of helium, while in the process, converting the mass of hydrogen to energy. The Sun is made up of gaseous molecules and emits the energy in the form of electromagnetic (EM) waves Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00001-7 © 2021 Elsevier Inc. All rights reserved.

3

4

PART | I Technologies

also known as radiation. The radiation is composed of two different forms of energy, namely temperature and photons. EM waves are a form of energy that promulgates as both electrical and magnetic waves propagating in energy packets called photons. The spectrums of EM radiation with different waves are based on their wavelengths (λ) and frequency (ʋ). Some of the EM spectrums are based on its wavelengths which include cosmic rays, xrays, ultraviolet (UV) rays, visible light, infrared (IR) rays, micro and radio waves. The photon energy of EM radiation is denoted by an electron volt (eV), where 1 eV is defined as the energy attained by an electron exited with a potential difference of 1 volt (V) [2]. The insolation of radiation at any location on the earth specified in W/m2 is not consistent but varies with the time of a day, latitude and season of a year. According to world radiation centre the intensity of solar radiation received outside of the atmosphere is 1367 W/m2 and is known as solar constant. Most of the solar energy is neither absorbed nor reflected back by the atmospheric layer before it is received on the earth’s surface through the amount of atmosphere called air mass (AM) which depends on the geographical site and the time of day and year. The AM value is zero for extraterrestrial radiations or solar constant and AM is one for the radiation normal to any surface or the Sun is at zenith (the shortest path to the surface). As per the industrial standards photovoltaic (PV) panels are tested at AM of 1.5 which is 50% longer than AM of 1. The solar spectrum epitomizes the intensity of solar radiation at each wavelength for any particular AM. Fig. 1.1 shows the solar spectrum at AM of 0 and AM of 1.5. The PV system is one type of a direct energy conversion device that converts solar energy particularly photons of energy into electricity through photoelectric effect as devised by Edmund Becquerel in the year 1839. As such, PV electricity is referred to as solar electric energy to differentiate it from

FIGURE 1.1 Spectral properties of solar radiation outside the earth’s atmosphere.

Solar energy technologies: principles and applications Chapter | 1

5

thermal energy that uses solar radiation in the form of heat to harvest electricity through indirect energy conversion techniques. All over the world PV manufacturing industry fabricate materials that can transform solar radiation into electricity at wavelengths with the supreme intensity to make the best use of conversion efficiency. The total quantity of solar energy received on any specific location greatly depends on latitude and climatic conditions. In general the equator area gets the maximum annual solar energy whereas the polar region receives the minimum. Also the amount of incident irradiance merely depends on the tilt angle (β) of the PV panel. In the following section the basic physics of solar electricity generation and crystalline technologies will be discussed, followed by different types of PV systems with its recent advancements. However, recent technologies like perovskite and quantum dot technologies are outlined as a separate section in detail.

1.2

Photovoltaic technologies

The world’s first invention of the silicon solar cell with a recorded efficiency of approximately 6% was developed by the Bell Laboratory scientists’ Pearson, Chapin and Fuller in the year 1954 and patented in 1957 [3,4]. During the initial period, that is during the 1960s’ and 1970s’, more amount of energy was needed to fabricate a solar cell than it could ever produce in the course of its lifespan. In subsequent years drastic enhancements have been taking place in the efficiencies and manufacturing methods by using different materials. However, silicon continues to rule the solar PV (SPV) market and power the electronics industry due to its abundant availability. At present silicon cells are having conversion efficiencies of 27.6% [5] as compared to an efficiency of 13% in the mid of 1970s’. Scientists across the globe have developed thin-film materials such as amalgamation of cadmium (Cd), indium (In), gallium (Ga), tellurium (Te), copper (Cu) to further curb the cost of silicon-based PV cells. These materials are less expensive to fabricate than pure silicon. Thin-film technology has achieved 20.3% efficiency, which is very high as compared to Si crystalline solar cells. At present the most efficient solar cells are four junction cells. As the name implies, it consists of four different layers of PV material that are properly stacked to convert four sections of solar radiation spectrum. The different types of SPV cells manufactured so far can be categorized into four authentic generations as depicted in Fig. 1.2 [6]. The modest and generally used flat module in household and streetlight applications for electricity generation is p
n junction flat module. This p
n junction is mostly made up of Si-based semiconductor materials combined together and compacted in a glass cover to isolate it from the environmental effects bounded with a metal casing. These types of direct sunlight to electricity-generating devices are simple nonconcentrating crystalline Si solar

6

PART | I Technologies

FIGURE 1.2 Generations of solar photovoltaic (PV) technologies.

cells, whereas concentrating PV (CPV) cells are making use of reflectors such as concentrating thermal collectors with high reflectivity mirrors, acrylic made lenses (linear Fresnel or Convex) to focus all the solar radiation falling directly to a point or a line of PV cells. On the other hand solar CPV technology is one of the emerging technologies wherein the solar light is made to focus 1000 times onto the p
n junction cells. In concentrated sunlight conditions, multijunction solar cells such as GaInP, GaAs and Ge have proven efficiencies of 44.4% whereas the Si solar cells have efficiencies ranging between 22.8% and 26.1% [5]. The CPV technology can be classified into the following two: low CPV (LCPV) cells and high CPV (HCPV) cells. LCPV technologies use low concentrated light and hence any type of PV cell can be used, whereas, for HCPV technologies, triple-junction silicon solar cells having high temperature withstanding capacity alone can be used [7]. The highest efficiency reported from Spectro lab for a 5-junction non-concentrator cell has been 38.8% and the highest efficiency reported for concentrating cells from Fraunhofer, Institute for Solar Energy has been 46.0% [7]. Hence one-half of the cost of electricity can be reduced for the locations having a higher intensity. A most important drawback of these CPV technologies is its dual-axis tracking systems. These systems are to be directed towards the Sun throughout the day to extract the maximum amount of solar radiation. Also the CPV system needs periodic cell cooling to achieve maximum operating efficiency. Hence these setbacks distress the usage of CPV systems for large electricity production plants.

1.2.1

Solar photovoltaic principles

The working principle of solar PV (SPV) cells is based on the PV or photoelectric effect for semiconductor materials. These formulate that, in certain circumstances, an electron (e2) of a semiconductor material can absorb an energy packet known as photon. The energy content possessed in the photon is given by the following equation:

Solar energy technologies: principles and applications Chapter | 1

E 5 hυ 5

hc λ

7

ð1:1Þ

When the solar radiation is incident on an effective surface area of a solar cell, the photons whose energy level is equal to or greater, then the band gap energy (Eg) of the semiconductor material gets absorbed. A very minimum amount of photons are redirected back from the exposed surface of the SPV cell. A very few photons are not able to penetrate as they are gridlocked from reaching the crystal layer by the metal grid that is used on the top surface of the semiconductor material to collect the electric current produced during the operation. Some photon energy which is less than “Eg” passes through the solar cell and gets dissipated as heat. Only when a photon of light with a suitable amount of energy enters a semiconductor solar cell near the p
n junction and approaches one of the silicon atoms, the following results occur subsequently: G

G

G

G

Excitation of e2 from valance band to the conduction band by leaving behind a hole called electron-hole pair (EHP) or generation. The generated carriers tend to move to the layer where it performs as a majority carrier, which means the e2 tends to drift to the layer of n-type material, while the hole has a tendency to move towards the layer of ptype material. The e2 can be absorbed by the current micro-grids on the front surface of the cell to produce an electric current in the external closed circuit and then return to the layer of p-type material. It repeatedly recombines with the hole in the layer of p-type material called recombination.

A semiconductor is a crystalline material in the solid phase, in which valence e2 and conduction e2 are permissible only in some energy values, with a forbidden gap of energies between both conduction and valance band called energy gap (Eg). Light energy having Eg greater than eV of energy possessed (usually between 0.5 and 5 eV) can excite the e2. The electrons in the outermost orbitals are at higher energy levels and are almost free to move in the orbitals called conduction electrons and the materials are known as conductors. On the other hand, all the electrons are tightly packed through chemical bonds between atoms in the orbitals, so all of them are called valence electrons and the material is an insulator. The number of electrons in the conduction band can be decided by the temperature and the energy content in solar radiation of a material called semiconductors. A pure Si-based semiconductor material at 0K acts as an insulator. The performance of the semiconductor materials can be decided by the level of doping, dopant materials and its type, direct or indirect gap energy type, intrinsic and extrinsic properties of the materials.

8

PART | I Technologies

Silicon (Si)-based semiconductor is used for the manufacturing of solar cells as Si is available abundantly in nature in the form of Quartz or sand (SiO2). The atomic structure of Si has four valence electrons in the outermost orbitals sharing with the nearest atoms by covalent bonds. Doping is a process of adding a foreign element to make this pure Si material into p-type or n-type materials for the formation of p
n junction. Replacing of an atom in the crystalline Si structure with III-group element Boron (B) or V-group element Phosphorus (P) having three and five electrons in the outermost orbitals, respectively. When a layer with majority holes and minority e2 (in p-type) is joined by a layer with a majority of conducting e2 and minority holes (in n-type), it forms the junction in which free e2 are combined with holes through the loss of energy. This junction is called the depletion region because all the bonds in this region are completely occupied and there is no existence of free electrons for electricity. Practically there are a many numbers of e2 in the p-type layer than holes, due to the fact that the e2 is in the conduction band of n-type layer. Hence there is a negatively charged portion in the p-type layer and a positively charged zone in the n-type layer. Eventually an electric field is exhibited in this zone which causes the voltage or potential difference across the junction of two layers. This p
n junction is otherwise termed to be naturally polarized. Hence this inhibits free electrons from the n-type layer crossing to holes in the p-type layer, because they would have to overcome the barrier potential. The significant parameters that characterize a cell are shown in this I
V curve (Fig. 1.3). They are the short-circuit current (ISC), the open-circuit voltage (VOC), the maximum power point current (IM), the maximum power point voltage (VM) and the fill factor (FF).

FIGURE 1.3 I
V characteristics of a solar cell.

Solar energy technologies: principles and applications Chapter | 1

9

1.2.1.1 Power of a solar cell The electrical power that can be extracted from a solar cell is directly proportional to its cell area (Acell) and the intensity of solar radiation (IT) that hits the effective surface area. It is denoted in Wp (Watt peak), tested under standard test conditions includes global irradiation of 1000 W/m2, AM of 1.5 and cell operating temperature of 25 C. 1.2.1.2 Fill factor The FF is the ratio between the maximum power or peak power (product of maximum voltage and current) from the solar cell and the product of the VOC and ISC. The following equation represents the FF: FF 5

Vm 3 Im VOC 3 ISC

ð1:2Þ

From Fig. 1.3 the FF is a measure of the area of the largest rectangle which will fit in the I
V curve. Higher voltage will result in a higher possible FF. For a perfect cell FF is equal to 1. However, parasitic resistance reduces FF and in real devices 0.7 , FF , 0.95. Maximum FF can be seen for the smallest series resistance (Rs) and the largest shunt resistance (Rsh), and its value is higher than 0.7 for good solar cells.

1.2.1.3 Conversion efficiency Conversion efficiency is defined as the ratio of the maximum power output (Pm) delivered by the PV cell to the input power (Pin) received at a given cell operating temperature T. Pm Vm 3 I m 5 Pin GT 3 Acell

ð1:3Þ

ISC 3 VOC 3 FF ISC 3 VOC 3 FF 5 Pin GT 3 Acell

ð1:4Þ

η5 η5

When a solar module is not exposed to the Sun or light source, it acts like a power electronic diode made up of Si-based p
n junction and follows the ideal diode equation. On the other hand, when a solar module is illuminated it yields an electric photo-generated load current (IL), which passes through the external circuit just opposite to the direction of the diode current (ID). Eventually the output current (I) is equal to the difference between IL and the diode current ID. The whole process is illustrated in an equivalent circuit of a solar cell is shown in Fig. 1.4. Commercial PV cells are made up of silicon (Si), gallium arsenide (GaAs), copper indium diselenide (CIS), copper indium gallium selenide (CIGS), cadmium telluride (CdTe) and others. The proper generation of electricity through the PV effect in the PV cells is enabled only when there is an

10

PART | I Technologies

FIGURE 1.4 Equivalent circuit of a solar cell.

appropriate form of p
n junction or its equivalent, such as a Schottky diode junction. Therefore the empathetic approach for the understanding of p
n junction is the core of how a PV cell transforms solar radiation into electricity. The p
n junction comprises of a layer of n-type material combined to a layer of p-type, with a continuous Si crystal structure across the junction. Layers that contain an additional number of free e2 are known as n-layer and the layer which has a huge number of free holes is called p-layer. Under thermal equilibrium conditions, the bond between a hole (p) and electron (n) densities at any selected point in the material, is given by n 3 p 5 ni 2

ð1:5Þ

The electrons in the n-layer and holes in the p-layer are under the concept of arbitrary diffusion in the Si crystalline structure. Hence each (electrons or holes) have a tendency to diffuse from a high concentration region to a low concentration region, according to Fick’s law. The barrier potential or concentration gradient is created across the junction of p-type and n-type materials due to the massive concentration differences. Eventually the e2 in the n-layer tend to diffuse across the junction into the p-layer and the holes in the p-layer diffuse across the junction into the n-layer, as shown in Fig. 1.5. The barrier junction is electrically neutral before the formation of the junction as both the layers are neutral. Consequently as e2 diffuse to the player crossing junction, they leave behind positively charged holes that are in the covalent bond with the Si lattice. Likewise, the holes diffuse to the nlayer, they leave behind a negatively charged hole-donor ions that are in the covalent bond with the Si lattice. These diffusions of e2 and holes across the junction produce a potential difference or electric field (E) across the junction as shown in Fig. 1.5. This field exerts forces on both the charged particles as per the relation F 5 q 3 E. This force ‘F’ causes the e2 and hole to drift in opposite sides as illustrated in Fig. 1.5. The holes drift from the n-layer to the p-layer through the junction in the same direction of the electric field. The e2 drifts from the p-layer to the n-layer through the junction in the direction opposite the field.

Solar energy technologies: principles and applications Chapter | 1

11

FIGURE 1.5 The drift and diffusion of an electron and hole across the p
n junction.

Carrying out an analysis of e2 and hole flow across the junction eventually leads to the formation of the familiar diode equation.
qV  I 5 Io 3 ekT 2 1 ð1:6Þ This can also model the V
I characteristic of the PV cell under illumination conditions, which is shown in Fig. 1.3. If there is no presence of external influences other than temperature, the flows of holes and the e2 are equal in both directions, resulting in zero net flow across the junction. This is called the law of detailed balance, which is consistent with Kirchoff’s current law. Important statements from Gauss’s law are the need for electric field lines to instigate at the positive charges and lay off at the negative charges. Hence the number of positive charges from n-side must be equal to the number of negative charges from the p-side [8].

1.2.2

Recent advancements in solar photovoltaic technologies

1.2.2.1 Perovskite solar cells The general meaning of Perovskite is the crystal structure of calcium titanate (CaTiO3), discovered in the year 1839 by the German mineralogist Gustav Rose and was entitled by the Russian mineralogist Lev Perovski. These

12

PART | I Technologies

materials use the most abundant elements on the earth, have low formation energies for deposition and are well-suited for bulk manufacturing techniques. Perovskite solar cells (PSCs) in recent times have been completely an emerging technology with environmentally realistic renewable energy alternatives to existing solar cell technologies for solving global contests in the area of power generation and climate change [9,10]. The aforementioned characteristics make the PSCs a best suit for terawatt (TW) power productions with low unit electricity generation costs. The performances of the PSCs are more comparable with that of the other thin-film technologies and enhancements in laboratory scale. PSC stability has made megawatt scale-up of this solar cell technology a deep area of research focus [11]. PSCs comprises of an active perovskite layer sandwiched between an electron transporting layer (ETL) and a hole transporting layer (HTL). If the light penetrated and passed through ETL, another layer called transparent conducting layer is placed in front of ETL called n
i
p and p
i
n structure. There are two basic structures namely mesoscopic and planar structures based on the different configurations such as n
i
p and p
i
n. The mesoscopic structure usually has an n
i
p configuration and the planar structures are further categorized into two sub-configurations namely n
i
p planar and p
i
n planar [12]. The n
i
p structure usually involves placing the perovskite material onto a transparent substrate sheltered with a thin layer of TiO2 and mesoporous TiO2 or Al2O3 (optional) framework layer [13]. The p
i
n structure encompasses depositing the perovskite material onto a transparent substrates which are enclosed with an HTL; for example poly (3,4-ethylene dioxythiophene):polystyrene sulfonic acid (PEDOT:PSS) [14,15]. There are different types of PSC manufacturing methodologies established and they fall under four main categories such as one-step, twostep, vapour-assisted solution method and thermal vapour deposition with a maximum PCE of 22.1% (current record), 20.26%, 16.48% and 17.6%, respectively. Jung et al. [16] have studied P3HT without dopant as a holetransport material using a wide-bandgap halide and found an operating efficiency of 22.7%. With an increase in efficiency up to 28% during the preceding years, hybrid organic
inorganic metal halide PSCs have become an important area of research for many scientists, academicians and researchers. The advantages of PSCs like long carrier diffusion lengths, largely adjustable bandgap methylammonium lead halides (MALHs) with a chemical formula of CH3NH3PbX3 which has a bandgap from 1.5 to 2.3 eV [17] and high light absorption coefficient ( . 104 per cm) along with the low-cost manufacturing techniques with the better efficiency creates PSCs comparable with Silicon (Si)-based solar cells, CdTe [18] and copper zinc tin sulfide (CZTS). The drawbacks such as device instability [19,20] arising due to the environmental factors namely thermal behaviour, illumination, ambient humidity [21], J-V hysteresis [22], lead toxicity, UV light and oxygen seems to be the major

Solar energy technologies: principles and applications Chapter | 1

13

problems that inhibit their further enhancement and the commercialization of PSCs. The instability with fluctuating temperature and pressure gains the superfluous anxieties for cell manufacturing. The future works such as elemental altering, tight compact sealing of the system and additional blocking layer inside the system might solve these encountered complications but further solidity tests under different tough environments are intensely required to obtain the expected mark in near future.

1.2.2.2 Other emerging photovoltaic technologies The viable contenders for a commercial application are CdTe (22.1%) and Cu (In, Ga)Se2 (CIGS) (23.4%) thin-film solar cell technologies that are as efficient as Si for below 100 3 applications. In addition to this for space/ lunar mission applications, III
V multijunction SPV with higher efficiencies (B44.4%) with a concentrator and (B37.9%) without concentrator is at superior regardless of higher costs. Substantial progress in research and development (R&D) and the status for each of these technologies provides confidence that 1 TW of power (approximately 3% of universal demand) will be produced by SPV systems in the near future (2020
30). The Cadmium (Cd) and Lead (Pb) are not environmentally friendly materials, and Gallium (Ga), Indium (In) and Telluride (Te) are not found in abundance on the earth. Thus what could be the possible PV solar materials that are emerging from R&D which are desirable to be expanded using the set of incumbent solar cell technologies. The important three explicit technologies of such emerging PV materials are discussed in detail in the following sections which include CdTe, CIGS and dye-sensitized solar cell (DSSC). 1.2.2.3 Cadmium telluride CdTe is a single junction direct-bandgap solar cell with a bandgap of 1.45 eV and the II
VI group elements in a chemical table referred to as metal chalcogenides with the maximum laboratory conversion efficiencies of 22.1% [23]. The manufacturing of CdTe can be achieved through three different possible ways. The layers of CdTe thin-film cells are manufactured by a compound deposition process at temperatures greater than 400 C that persist less than two and a half hours, as shown in Fig. 1.6. These structures consist of a p-type CdTe absorber layer combined with n-type CdS window layer to form a heterojunction intermixed interface region. The cadmium sulphide (CdS) layer is vapour-deposited onto a transparent conductive oxide (TCO), a front contact supported by a heat-treated glass. Subsequently the CdTe layer is placed on the CdS layer. CdTe material is naturally stable under the spectrum of solar radiation due to its bonding strength which is much higher than the energy of a photon in the solar spectrum. At this stage laser cut permitted through these layers to make the insulator into the

14

PART | I Technologies

FIGURE 1.6 Schematic diagram of cadmium telluride (CdTe) solar cell illustrates the different layers and its nomenclature.

module. Later quite a few cuts are made using the laser only for the CdS and CdTe layers, to enable positive (1) rear contact by sputter deposition. Eventually encapsulation is made using tempered rear glass to protect it from the environmental conditions. It is chemically stable with a robust nature makes them attractive even though they occur sporadically (CdTe).

1.2.2.4 Copper indium gallium selenide The closest profit-making contestants for global power applications are CdTe and CIGS thin-film solar cell technologies that are much efficient but use approximately 100 3 less absorber material when compared with Si. Multiple semiconductors from the I
III
VI2 column of the periodic table, such as copper indium diselenide (CIS) and copper
gallium
diselenide (CGS) are combined called copper
indium
gallium diselenide (CIGS) with a general formula of Cu(InxGa1-x)Se2. It is in the form of chalcopyrite due to its tetragonal crystal structure synthesized by preparing a molten mixture containing the desired amount of each element. CIGS can be made in a wide range of configurations and their equivalent phase diagrams have been extensively investigated worldwide. The bandgap of CIGS can be tuneable by varying the ratios between In: Ga and Se:S to attain 1.0
1.7 eV depending upon the fraction of the elements in the composite [24,25] and the existing efficient cells are manufactured with a bandgap in the range of 1.20
1.25 eV. Consequent enhancements in the efficiency were accomplished later and with the best ever efficiency of 23.4%, the highest for any thin-film technologies [5] and are comparable to commercial crystalline silicon cells and even better than organic PVs. Conversely CIGS is more expensive due to the scarcity of indium and difficulty in the manufacturing process. The CIGS can be developed in both substrate and superstrate configurations, but the substrate

Solar energy technologies: principles and applications Chapter | 1

15

FIGURE 1.7 Schematic diagram of copper indium gallium selenide solar cell.

structure offers the highest efficiency due to constructive processing conditions. On the other hand it involves a further glass or encapsulation layer to shelter the cell surface, which is not necessary for the superstrate structure. CIGS cells are usually manufactured using the following five steps [26]: a substrate such as a metal, a ceramic or a polymer sheet sometimes Na2CO3
CaO, is engaged to backing the whole layer of the solar cell. Subsequently the substrate is shielded with the back connection, which is typical of MoSe2. The coevaporation process is used to make the p-layer of CIGS and buffer n-layer made of a TCO such as zinc oxide (ZnO). At last an anti-reflective coating on the top is provided to enhance the efficiency of the cell. The cross-section structure of a CIGS cell is depicted in Fig. 1.7.

1.2.2.5 Dye-sensitized solar cells DSSCs consist of a sandwiched structure of anode and cathode including electrolyte to accomplish an energy conversion efficiency of 5%
9% [27]. The anode is made up of mesoporous metal oxides such as titanium dioxide (TiO2) for high porous structure and increased surface area, and zinc oxide (ZnO), deposited on the back of the glass and the cathode is made by platinum-coated TCOs. The metal oxide electrode is dipped in a natural mixture of photosensitive ruthenium-polypyridine dye (molecular sensitizer) and solvent for a whole night and the electrolyte is filled within the sandwiched electrodes. A thin conductive sheet (platinum) is spread over a thin layer of iodide electrolyte and these two plates are joined and sealed together to prevent leakage of electrolyte. When this arrangement is exposed under the sunlight due to the absorption of photons energy, excitation of e2 take place in the system leads to circulating current in the closed-loop through the load. Absorption of a photon by dye (S) and molecule are excited (S ) to get into TiO2. Eventually the excited molecule gets oxidized (loss of e2) to form S1 as shown in the following equation:

16

PART | I Technologies

FIGURE 1.8 Schematic of dye-sensitized solar cell.

S 1 Photon-S -S1 1 e2

ð1:7Þ

This excited electron is injected to the conduction band of the semiconductor. A TCO layer collects the e2 from the conduction band and flows through the external circuit. Oxidized dye molecule S1 is reduced to its original form (S) by collecting the e2 from the organic electrolyte. The electrolyte solution contains the iodide redox system in which the iodide ions are oxidized to tri-iodide ions. The flow of e2 will be acknowledged through the counter electrode for the reformation of iodide. This step requires catalytic presence, that is Pt. The absorption of light energy is by the dye molecules and charge separation by e2 injection from the dye to the TiO2 at the semiconductor electrolyte interface is shown in Fig. 1.8. The physical and chemical composition and structure of the electrolyte predominantly affect the material stability and the conversion efficiency of the DSSC [27]. At present high conversion efficiency between 12.3% [5] and 14.6% [28] is obtained through metal-free sensitizers with porphyrin dyes as light-absorbing capacity due to its prevailing absorption in the visible spectrum of solar radiation. Cadmium alloyed Quantum dots (QDs) based DSSCs have a maximum efficiency of 6.36% [29]. In recent times perovskite sensitizers based DSSCs have reached a maximum conversion efficiency of 20% [28].

1.2.2.6 Quantum dot solar cells The colloidal crystalline concept has been widely proposed during the early days of the 1960s and over the past few decades. Some novel materials or nanostructures with variable bandgap or intermediate band are invented in the present scenario that can be altered to match the spectral distribution of

Solar energy technologies: principles and applications Chapter | 1

17

FIGURE 1.9 Schematic diagram of quantum dot sensitized solar cells (QDSCs), comprising a fluorine-doped tin oxide (FTO), photoanode, quantum dot (sensitizer), electrolyte and counter electrode (CE).

solar radiation without naming this as QDs. Then the term QD was given by Reed group in 1988 [30]. But in early 1985, Louis Brus established a quantum model of spherical QDs upon an effective mass model [31]. Tuning the size, the QDs can help control the absorption and emission spectrum of these photons and determines the effectiveness of QDs [32]. Artificial atoms or molecules with energy gap and energy levels of nanoscale semiconductor with the size of 2
100 nm [25,33,34] are the QDs and these materials fall in the II
VI, III
V or IV
VI groups of the periodic table. The energy bandgap increases with a decrease in the size of the QDs (Fig. 1.9). QDs are emitting up to three e2 by single-photon absorption unlike only one for crystalline semiconductor material solar cells [35] and now it has been increased to seven e2 due to the absorption of higher energy photons (up to eight times of Eg) by inverse Auger recombination [36]. The movement of the e2 and holes are restricted in the three directions of the space. When the sufficient amount of Eg is supplied, the movement of e2 from valance band to the conduction band in semiconductors occurs. Whereas in QDs, the valance and conduction bands are thus close together like continuous bands [33] but no longer are treated as a continuous band but as discrete energy levels, and the properties of those atoms lie between semiconductors and discrete atoms. The different organic layers are deposited/molded in a 2D sheet or 3D arrays are in the actual structure of QDs [32] with distinct electronic and optical properties. This can be treated to create junctions on economical substrates such as a metalloid crystalline core, plastics or glass. Initially 3D silver (Ag) atoms are deposited on GaAs substrate using the two-step process [37]. These can be easily joined with organic polymers and DSSCs. At present the QDs are pooled with DSSCs enriched the matrix cell

18

PART | I Technologies

efficiency of 1% to 11.6% [38,39]. Further enhancement in efficiency is possible through doping with materials like Si which can increase the efficiency from 11.3% to 16.6% [25,40]. Despite toxic in nature (only a few QDs) and finite size, it is an immature technology and the lack of manufacturing and production techniques are the setbacks to this technology and making it trouble in hitting into the market still today. Nowadays quantum dots are promising nanostructure materials in photonics and biomedical applications.

1.2.3 G

G

G

G

G

Applications of solar cells

Distinct SPV cells can be used for operating torch and flashlights, electronic watches and others. Ability to exhibit high transparency, the polymer solar cells are used as well in windows, stretchable electronics and so on. Solar power generation using SPV systems can be used for residential, commercial, industrial, agricultural and traction applications Recent research focuses on the electrical vehicle driven by solar energy which is a need of the hour technology Solar cells are the prime important source of energy for lunar missions in powering space vehicles such as satellites and Hubble

1.3

Solar thermal collectors

Solar thermal collectors (STC) are used to convert solar energy into thermal energy that can be stored for later use. STCs have drawn attention among the researcher in the last decade due to their easy construction and ability to deliver heat for domestic and industrial purposes. This heat energy may be used for cooking, refrigeration, cooling, desalination, drying and melting metals. The solar energy utilization has been classified broadly as low-, mediumand high-temperature system. Low-temperature system such as flat plate collectors works at a maximum temperature of 100 C, whereas medium temperature system such as line focusing technology works at a maximum temperature of 400 C. The high-temperature system such as central receivers and paraboloidal dish collectors works at more than 400 C. The STCs can be grouped into two major categories such as stationary (flat plate collector, evacuated tube collector and compound parabolic) and tracking concentrating collector (parabolic trough, parabolic dish, heliostat field collector and Linear Fresnel collector). The concentrating collectors are further divided into subcategories such as distributed receiver and central receivers to concentrate the sunlight using point focus and line-focus optics. The hot water generated from STC can be used for different application as mentioned in Table 1.1 Generally stationary STCs are used in residential and commercial water heating or space heating applications. Whereas the

Solar energy technologies: principles and applications Chapter | 1

19

TABLE 1.1 Type of solar collectors and its application. Type of solar collector

Absorber type

Temperature range ( C)

Application

Flat plate solar air heater

Flat

30
75

Space heating and drying

Flat plate collector

Flat

40
85

Evacuated tube collector

Flat

50
150

Solar water heating, space heating, solar refrigeration, desalination and industrial process heat

Compound parabolic collector

Tubular

60
220

Parabolic trough collector

Tubular

60
300

Linear Fresnel collector

Tubular

60
250

Parabolic dish collector

Point

100
500

Heliostat field collector

Point

200
2000

PV/T liquid collector

Flat

30
80

Simultaneous hot water and electricity

PV/T air collector

Flat

30
65

Simultaneous hot air and electricity

Power generation

PV/T, photovoltaic
thermal.

concentrating STCs are used to generate power. The lifetime and efficiency of a PV system are very less as compared to the concentrating collector. Hence concentrating STCs plays a vital role in the effective utilization of solar energy. The STCs performance depends on isolation, the inlet temperature of the heat transfer fluid (HTF), thermo-physical properties of HTF, HTF flow (natural or forced) atmospheric temperature, wind speed, relative humidity, absorber material, coating and geometry, insulation material and collector angles. Various research was conducted in the last two decades to improve the performance of the system. One of the major developments is the tracking mechanism (single-axis and two-axis) to improve solar concentration. Nowadays, hybrid cogeneration systems, for example PV/T collectors are showing greater attention among all the researchers for the integration with buildings.

20

PART | I Technologies

1.3.1

Stationary collectors

The stationary collectors are permanently fixed and tilted according to the location based on the latitude and longitude to absorb more heat energy from the Sun. Flat plate solar water heater (FPSWH) consists of five major components such as glazing, absorber plate, absorber tube, insulation and casing (Fig. 1.10). The radiation from the Sun passes through the glazing and strike the absorber plate. The glass is used to reduce the convection and radiation losses from the collector. The absorber plate is coated with black paint to absorb more radiation. The radiation wave hits the absorber plate surface thereby absorber material is energized and the heat is then transferred to the absorber tube due to conduction heat transfer. The absorber tubes at the bottom and top are connected using a header tube. The HTF passes inside the absorber tube to exchange the heat from the tube wall to the fluid. The hot HTF can be stored in the storage tank for later use. To reduce the heat loss from the collector to ambient, insulation has been made between absorber tube and casing. The performance of the FPSWH was improved to increase the conduction and convection heat transfer mechanism [41]. To reduce the material corrosion the evacuated type tube solar collector (ETSC) has been developed. The heat pipe (individual absorber tube) has been modified with

FIGURE 1.10 Stationary solar collectors: (A) flat plate solar water heater; (B) flat plate solar air heater; (C) compound parabolic collector; (D) evacuated tube collector.

Solar energy technologies: principles and applications Chapter | 1

21

vacuum sealing arrangements and the arrangements are shown in Fig. 1.10. The ETSC not only increases the life span but it also increases the outlet temperature of the collector by reducing conduction and convection losses. The compound parabolic collector is made to reflect all the incident insolation using parabola-shaped reflector within the limits to the absorber tube [42]. The orientation is associated with its acceptance angle. The dual sections of a parabola are used to accommodate all the incoming insolation at different angles as shown in Fig. 1.10, and it avoids the necessity of a Sun tracker. For a stationary collector, the minimum acceptance angle is 47 degrees. The insolation enters the aperture and hits the absorber tube placed at the bottom by various internal reflections. The performance of the collector is improved by changing the absorber tube geometry, HTF properties and vacuum manifolds. Flat plate solar air heater (FPSAH) is used to produce hot air for space heating and drying application. The system components and arrangements are similar to FPSWH as shown in Fig. 1.10. In FPSAH the absorber plate surface has been extended using fins to contact with air molecules. The air will be forced through a blower to maintain the outlet temperature at a constant level. The insulating material plays an important role in the performance enhancement [43]. PV panels may be integrated with the collector to operate the blower for achieving a better performance. PV panel only converts 20% of the radiation into useful electricity and 80% of the radiation is just heating the PV cell [44]. This heat energy affects the performance of the PV. Cooling the PV cell from both top and bottom sides increase the heat transfer rate thereby improving the performance of the system. Hence the hybrid PV/T collector combines two individual components (PV cell and collector) to a single device for cogeneration, which increases the PV cell efficiency with which the electricity and heat energy can be generated simultaneously. The performance of the PV/T system strongly depends on collector length, properties of HTF, cell density and duct depth. Further the hybrid solar collectors can be used for drying, water heating, solar cooling and desalination applications.

1.3.2

Tracking concentrating collectors

Concentrating collectors are used for converting solar energy into heat energy, and the radiation emitted by the Sun is optically concentrated using mirrors or lens before converting into heat. Different concentrating collectors and their arrangements are shown in Fig. 1.11. The concentrating collector achieves a higher temperature of the HTF as compared to stationary collectors for the same collecting surface. The heat energy from concentrating collector shows a great potential to reduce the demand for a thermal substitute in the industry. The concentration ratio can be defined as the ratio of aperture to the absorber area. A higher concentration ratio leads to higher thermal

22

PART | I Technologies

FIGURE 1.11 Tracking concentrating solar collectors: (A) parabolic trough collector; (B) linear Fresnel collector; (C) parabolic dish collector; (D) heliostat collector.

efficiency. Hence the reflectors and receivers are the major performance influential components. The shape of the reflectors may be cylindrical, parabolic or segmented. The receivers may be (flat, convex, concave, cylindrical) with glazing or without glazing. The glazing is used to reduce the heat loss and the system performance can be augmented using a tracking system [45]. PTCs are designed by bending a reflector into a parabolic shape. The receiver (tube with black coated) is placed in the entire focal line. One-axis tracking method is used to point the PTCs towards the Sun, and the rays from the Sun incident on the reflectors thereby it hits the entire length of the absorber tube. The absorber material is energized uniformly, and this energy is further transferred to the HTF. Different HTF such as diathermic oil, molten salts or water-based mixtures can be used according to the application. To overcome the structural constraints in the PTCs, the Linear Fresnel Reflector (LFR) has been developed. LFR strips are aligned horizontally and reflect the Sun rays to the linear receiver. The receiver will be stationary which uses curved type or flat type which are relatively cheaper than parabolic type. Parabolic dish collector concentrates all the incoming solar radiation to a focal point (receiver) on the dish along with two axes tracking mechanisms. The dish absorbs solar energy and transfers it to the HTF. This heat energy

Solar energy technologies: principles and applications Chapter | 1

23

can be converted into electricity in the focal point of the individual dish using a small generator and this heat energy from the individual system can be grouped together for the central energy conversion system. A central receiver system has an array of flat mirrors or heliostats which reflect (reflector surface: 50 to 150 m2) the incoming radiation to the receiver placed at the center. Altazimuth tracker has been generally used in this application and the mirrors can track the Sun. Conclave mirror is highly suitable to reflect a large amount of incoming radiation to a focal point to produce a higher amount of steam. Recent research focuses on changing the various heat transfer fluids, integrating energy storage techniques and so on.

1.4

Solar cooling technologies

In global energy end-use, 36% contribution is from buildings operationconstruction and it is also responsible for approximately 40% of carbon dioxide (CO2) emission globally [46]. Several strategies have been reported to reduce the peak demand load and overall energy consumption in suburban and commercial buildings. Some of them include, the usage of renewable energy, efficient cooling system, appliances and daylighting, insulation and use of phase change materials can reduce the peak load demand [47]. Solar energy technology is one of the promising renewable energy technologies for the development of net-zero energy building and zero peak building. The peak demand in the built environment occurs during high thermal stress conditions; hence the solar space cooling system plays a vital role in peak demand reduction. One of the main reasons for the development of a solar cooling system is the harmonious nature of demand and supply. The solar energy conversion technologies are one of the affordable forms of renewable energy and it can be easily integrated with different types of building. Instead of generating power from the solar and distributing to the utility, it is advised to construct the system to reduce the peak energy which is normally used in high thermal stress conditions. To reduce the peak demand and emission, further research on the solar-powered cooling systems is inevitable and it is the best option during daytime which can also be used for cold storage. Solar cooling refers to the devices and processes that convert heat harnessed from the Sun into a useful cooling system. The solarpowered cooling system has the advantage of providing zero-emission with eco-friendly working fluids. The solar cooling system includes three components (solar collector, heat sink and refrigeration/air-conditioning unit) as shown in Fig. 1.12. Solar cooling is subjected to solar energy either by hot water or electricity. There are various technologies such as compression, absorption [48], adsorption [49], desiccant [50], thermoelectric [51], ejector [52], Stirling [53], vacuum cooling technology [54], evaporative cooling system and

24

PART | I Technologies

FIGURE 1.12 Components of a solar cooling.

TABLE 1.2 Architecture of solar cooling system. Solar energy conversion system

Cooling technologies

Application

SPV (electricity)

Vapour compression Thermoelectric Ground source heat pump

Solar thermal Flat plate Evacuated tube Concentrated

Absorption Adsorption Desiccant Ejector

Space cooling: residential and commercial buildings, laboratories, etc. Food storage: vegetables, fruits, meat, fish, etc. Processing industry: dairy, paramedical, chemical, etc. Electronic cooling and devices cooling

hybrid [55] to produce cooling effect from solar energy [56], [57] and they are described in Table 1.2. Due to the harmful effects of conventional refrigerants, the researchers are focussing on the development of eco-friendly alternate options such as sorption cooling. Even well-established companies such as Thermax, Voltas and Blue Star are contributing to the eco-friendly absorption machine. These industries are involved in developing the commercial solar cooling system with capacities ranging from 4.5 kW. There are a lot of building setups such as office, residential and workshops that uses the sorption system (absorption, adsorption and desiccant). Solar sorption systems for residential applications operate hot water and water vapour at 60 C
90 C and 150 C, respectively. The average cooling capacity for the residential building would be in the range of 3
10 kW. The coefficient of performance (COP) of an absorption system is in the range of 0.15
0.6. The liquid sorption chiller, using H2O/LiBr and NH3/H2O are widely tested for residential building but the usual payback period is more

Solar energy technologies: principles and applications Chapter | 1

25

than 10 years. The adaptation of other technologies with the residential building is still under research. The dry absorption system is more suitable for small capacity applications. The development of hydrogen—metal hydride-based cooling system [58]; CO2—adsorbent based cooling system [59] and NH3—salts-based cooling system [60] are emerging nowadays to boost the performance and efficiency. Several works have been done on the performance evaluation of sorption cooling with metal hydrides. But a very few attempts were made in the development of CO2—adsorbent-based and NH3—salts-based sorption cooling systems. Therefore strong theoretical and experimental studies are required to identify the suitable sorption cooling system among hydrogen, NH3- and CO2-based system for required operating conditions. The solid desiccant is usually powered by an air heater with which the COP above 1. The liquid desiccant (LiCl
water) for the same heat source has a COP of 0.6. The COP of integrated vapour compression and the desiccant system is more than five [61]. Further integration possibilities with VCR, evaporative cooling and sorption system are still under research. The earlier mentioned systems could be integrated with building partially or fully with a storage system to produce continuous cooling. The ground source heat pump (GSHP) is an eco-friendly energy-saving technology for heating and cooling, in which the ground is used as a heating or cooling source. The annual installation of the GSHP system has increased in the rate of 10%
30% [62]. Worldwide more than 1 million GSHP systems were installed during 2008 and this system is very popular in European countries, United Kingdom, Turkey and North America due to its merits [63
66]. The effective integration of ground sources with the conventional system not only increases efficiency but also reduces CO2 emissions [67]. The solar cooling technologies are not economically viable, and further research and development in the existing system or developing a new system will play a vital role in the effective utilization of solar energy. However, solar energy can be used to reduce the peak demand which is normally used in high thermal stress conditions utilizing low-grade heat from solar systems.

1.4.1

Solar photovoltaic powered cooling system

1.4.1.1 Solar vapour compression cooling system Cooling assists the cold storage requirements in food processing industries using ammonia as a working fluid. In comparison with conventional vapour compression refrigeration (VCR) systems, significant energy and environment savings could be achieved using solar powered systems. Combining SPV and VCR into a single Solar Vapour Compression Cooling system (SVCC) system addresses both the issues such as cold storage and space cooling spontaneously. The biggest advantage of the SVCC system is its

26

PART | I Technologies

FIGURE 1.13 Vapour compression cooling system.

simple construction with a higher COP. The pictorial representation of the SPVC system consists of a mechanical compressor, condenser and evaporator as shown in Fig. 1.13. The demand for cooling is increased when the intensity of solar energy is high. However, due to the intermittent nature of solar energy, thermal energy storage is inevitable for continuous operation. PV cells convert solar radiation into electrical energy. The electricity generated by these modules is in the form of direct current (DC) where a DC motor-driven compressor (to drive the refrigerant) is used for the vapour compression system or it uses an inverter to convert the produced DC to AC. Then the compressed hot refrigerant is passed through the condenser coil and liberates the heat to the atmosphere. This causes the hot refrigerant to condense back into a warm liquid. The warm liquid is then carried back to the evaporator by passing through an expansion device which decreases the temperature and pressure of the liquid. The cold liquid flows into the evaporator coil and absorbs the heat from the indoor environment and it becomes a lowpressure gas. This low-pressure gas flows to the compressor and the cycle is repeated. Fig. 1.13 shows the working of solar-powered vapour compression refrigeration system for space cooling applications.

1.4.1.2 Solar thermoelectric cooling system A solar thermoelectric cooler is a semiconductor where the components are connected in parallel (thermal) and series (electrical) as shown in Fig. 1.14. The heat is flowing from cold junction to hot junction and it is triggered by the Peltier effect when the current flows across the thermoelectric components. This Peltier effect arises at the junction if the heat is flowing in a specified direction. High potential electricity is supplied to the junction providing to provide a heating and cooling effect at either side. The intensity of cooling and heating depends on the magnitude of the electricity and

Solar energy technologies: principles and applications Chapter | 1

27

FIGURE 1.14 Solar thermoelectric cooling system.

temperature difference. The high magnitude SPV cells can be used in the thermoelectric cooler. The thermoelectric effect was used in different applications such as electronic cooling, medical devices cooling, space cooling and refrigerator. The system is compatible with low capacities. However, the systems size up to 700 W are available for refrigeration and recirculating chillers.

1.4.1.3 Solar ground source heat pump system Waste heat from the PV panel is extracted using heat transfer fluid to increase the PV cell performance and this technique is called PV-thermal (PVT). All the electricity, cooling and heating (tri-generation) demand issues in the building can be solved by combining GHSP with PVT. The system description and working principles of the integrated system are presented in Fig. 1.15. The major objective is to generate onsite electricity from PV and to reduce grid electricity consumption. This integration powers the electrical component such as compressor and pump using electricity from SPV and collects the heat energy for hot water generation. A storage system can also be used with the above-mentioned system to reduce the fluctuations in demand and supply. The GSHP has five components namely ground heat exchanger, evaporator, compressor, condenser and expansion device. In a heating mode, the evaporator is coupled with the ground to exchange the heat. As the temperature of the ground is relatively higher than the atmospheric temperature, the liquid refrigerant in the evaporator absorbs the heat energy from the ground using heat transfer fluid. The low-temperature vapour enters the compressor powered by an SPV where the pressure and temperature of the refrigerant increases. The high-temperature vapour refrigerant loses its heat to the heat

28

PART | I Technologies

FIGURE 1.15 Schematic diagram of PVT-GSHP system: (A) winter condition; (B) summer condition; (C) GSHP - winter condition; (D) GSHP - summer condition.

distribution system without any change in the pressure inside the condenser and again the phase change takes place from vapour to liquid. The heat distribution system supplies the heat to the building and water. The pressure and temperature of the refrigerant are further reduced by an expansion device and it enters the evaporator. In the cooling mode, the condenser unit is coupled with the ground, the refrigerant drops the heat energy to the ground through heat transfer fluid circulated in the ground heat exchanger. The ground temperature is relatively low as compared to the atmosphere. The liquid refrigerant from the expansion device absorbs the heat energy from the cool-distributed system in the evaporator, and the phase change occurs from liquid to vapour. The distribution system is used to cool the space. The compressor increases the pressure and temperature of the refrigerant which is leaving from the evaporator and it enters the condenser. The cycle is closed and repeated for continuous production.

1.4.2

Solar thermal powered cooling system

The attention towards the thermal powered cooling system is growing in the past two decades. The two major benefits of this technology are low carbon

Solar energy technologies: principles and applications Chapter | 1

29

FIGURE 1.16 Solar sorption cooling process.

emission and minimal electrical requirements. Eco-friendly working fluid further leads to reduce global warming and ozone layer depletion. Also it has advantages like low operating and maintenance cost, fewer moving parts in the system as compared to a conventional system that makes them viable in a remote area (nongrid).

1.4.2.1 Solar sorption cooling system Absorption denotes a process in which one substance is completely mixed with another substance. These substances may be in different states and it forms a homogenized mixture which is called a strong solution. The absorption system has mainly five components namely evaporator, condenser, expansion device, generator or desorber and heat exchanger (Fig. 1.16). The exothermic and endothermic reactions are taken place during the absorption and desorption process respectively. Hence the working fluid is the major determining factor for the absorption system with regard to its coefficient of performance. The performance of commercialized working pairs such as LiBr/H2O, LiCl/H2O and H2O/NH3 were analyzed using solar energy and inferred that it requires a minimum cut-off temperature of 75 C. They contain some major drawbacks which include corrosiveness and crystallization of LiBr and the toxicity. The present research scenario focusses on the generator of a solar absorption refrigeration system at various concentrations of H2O/LiBr. The adsorption denotes surface interaction between solid (absorbent) and gas (refrigerant). The physical and chemical interaction during the adsorption process takes place due to the adsorption forces. The physical interaction is governed by van der Waals’ principle, which postulates that the adsorption takes place as the gas molecules contact the absorbent surface. Whereas in the chemical interaction, the adsorption takes place due to the exchange of electrons between gases and solid. Hence the working pair (solid/gas) is a highly determining factor for the performance of the system (activated

30

PART | I Technologies

carbon
methanol, silica gel
water or activated carbon
ammonia for the physical interaction of NH3 salts with alkaline compounds or metal hydrides and hydrogen with low-hysteresis intermetallic for the chemical interaction). Finding the optimum working pair is the key research for the development of the adsorption system. Initially the refrigerant gets evaporated in an evaporator producing refrigeration effect. Then the low-pressure refrigerant transferred to the absorber and gets adsorbed on adsorbent at low temperatures. Since the process of adsorption is exothermic; a certain amount of heat is liberated into the atmosphere. The adsorber is then disconnected from the evaporator and is sensibly heated to attain the required condenser temperature and corresponding pressure. After reaching the equilibrium conditions, refrigerant gets desorbed from adsorbent using solar thermal energy. The desorbed refrigerant is condensed back. The operation completes by the expansion of refrigerant through the expansion valve.

1.4.2.2 Solar desiccant cooling system Zeolite and silica gel are commonly used solid desiccation material in the desiccant wheel to dehumidify the ambient air. During this process (Fig. 1.17), the air is heated due to the exothermic release of heat which is then humidified and enters the heat exchanger where the cooling effect is provided inside the living space. Solar energy is used to desorb the water from the desiccant material in the desiccant wheel and this process continues. In liquid desiccation, there are various components such as absorber, re-generator and cooler (evaporative type). H2O-LICL and H2O-CaCl2 were used as an absorbent solution to dehumidify the air and then the air is passed through the humidifier where it gets humidified to the required

FIGURE 1.17 Principle of solid desiccant cooling system.

Solar energy technologies: principles and applications Chapter | 1

31

temperature (Fig. 1.18). The cooled air is then distributed to space and finally, the solar energy is used to remove the water from the absorbent. Normally desiccant systems are commercially available for small capacities space cooling applications (30 kW). However, the solar integrated desiccant system is not yet commercialized though a few industries are in the process of developing desiccant systems. Current research focusses on improving the COP by integrating with the vapour compression system which will be a potential candidate for a humid climate.

1.4.2.3 Solar ejector cooling system Solar thermal energy is used to produce the refrigerant vapour inside the generator. This system has a generator, ejector and pump as the major components (Fig. 1.19). Low boiling point fluids such as CFC and HFCF (early days); HFE and HC (last decade) are used as refrigerant in this system. Even though the COP of this system for space cooling applications is less than

FIGURE 1.18 Principle of liquid desiccant cooling system.

FIGURE 1.19 Solar ejector cooling system.

32

PART | I Technologies

0.3, it is still an attractive option due to its simple construction and low operating cost whereas, for refrigeration, the COP is less than 0.12. The ejector cooling system has been classified as single-stage, multistage and hybrid systems. The hybrid system with integration of solar adsorption system is under research focussing on the techniques to augment the COP. Design complications of the ejector and operating temperature restrictions are the major challenges of this system which makes as the key parameters to consider in future research. From the detailed discussion of the different types of solar cooling systems including types, application and COP, it is concluded that all the above cooling systems have great potential for commercial development due to their energy-saving potential and environment-friendly nature. However, the development of an optimum solar cooling system in accordance with the application is a need of the hour.

1.5

Solar pond

A solar pond is a structure to save and store heat energy from the Sun. Later the heat can be retrieved for various applications such as industrial process heating and power generation. It works on the principle of creating a salt density gradient in the lower surface of the water thereby preventing the natural convection to occur between the water layers and surpassing the heat transfer from the top layer of the water with the atmosphere. Various salts were used in the bottom layer of the solar pond to create the salt gradient. Thus maintaining three layers (Fig. 1.20) namely the upper convective zone (UCZ), nonconvective zone (NCZ) and lower convective zone (LCZ) where the salinity is proportional to depth. LCZ which has high-density salts acts as a storage medium. Earlier research in the area of solar pond focussed on varying the pond solution and concentrations. Sodium carbonate, sodium chloride, magnesium chloride, urea were used as pond solutions.

FIGURE 1.20 Solar pond depicting the upper, middle and lower convective zones.

Solar energy technologies: principles and applications Chapter | 1

33

Nowadays research is being done by using nanofluids and phase change materials in a solar pond. To prevent the algae growth inside the solar pond, copper ethylamine was used. Recently structural modification has got a momentum in solar pond research. Trapezoidal, rectangular and u-shape pond with external heat exchanger was used as structural modifications. The temperature of the trapezoidal structure solar pond was higher as compared to the rectangular structure. The rotatable cover was used in a solar pond to act as a reflector during daytime and insulator during the night-time. Plastic glazing, ball bearing, pebbles were integrated with a solar pond to increase the temperature of the lower convective zone owing to the higher heat capacity. Future research in the area of a solar pond may focus on optimizing the configuration, developing a hemispherical structure to reduce the shading effect, effective turbidity control techniques, nanofluid based effective heat extraction methods, efficient hybrid solar pond integration with solar collector, air conditioning, solar chimney, desalination and power generation systems.

1.6

Solar cooking

Solar cooking is one of the applications of solar thermal technology which was initiated by a German scientist Tschirnhausen during the year 1651
1708. It works on the principle of utilizing heat energy from the Sun for cooking purposes. Lens and reflectors were used to focus and reflect the solar radiation on to the system. The whole system is insulated to avoid heat loss to the surroundings. Depending on the structure, solar cookers were categorized into box type, concentrating type and panel type. Based on the cooking method, solar cookers were categorized into direct and indirect types. In direct-type cookers, heat from the Sun will be used directly to heat the cooking vessel whereas in indirect types there will be a steam medium through which the heat will be transferred to the cooking vessel. Vacuum tube and olive oil are used as a heat transfer medium in indirect type solar cookers. Among the various classifications, box-type cookers are quite famous and common (Fig. 1.21). The time required for solar cooking is higher as compared to modern technology and hence solar cooking is not widely implemented in household cooking. To boost the efficiency of solar cookers, they were integrated with mirrors, reflectors, double glazing to concentrate more solar radiation on the cooking vessel. Attaching fins with cooker decreases the cooking time. SPV systems, solar concentrators and energy storage materials (sensible heat and latent heat) are also integrated with the solar cookers to improve the efficiency of solar cookers.

34

PART | I Technologies

FIGURE 1.21 Schematic of solar cooker.

1.7

Solar desalination

Utilizing solar energy to convert the brine water to clean water is solar desalination. Solar desalination is categorized into direct and indirect types. If the input water that needs to be desalinated directly absorbs the solar energy to desalinate, it is termed as direct type. Whereas in indirect type systems, the radiation will be first absorbed by a heat source (e.g. solar collectors) and it will be transferred then to brine water.

1.7.1

Indirect type desalination

In indirect type plants, there are two systems (solar collector and desalination systems) integrated together to produce freshwater. Humidification and dehumidification techniques, membrane desalination, multistage flash (MSF) techniques, multieffect desalination and vapour compression technique comes under the indirect type desalination system.

1.7.1.1 Humidification and dehumidification desalination The working principle is quite similar to the formation of rain from lakes, sea and others. Saline water is heated by solar energy and humidifies the air available in the atmosphere. Then once it gets dehumidifies, the resultant water is the desalinated water. Here the ambient air is heated using air heaters and water is engaged simultaneously with the warm air heated directly through water heaters assisted by an external source (solar, thermal, geothermal, electrical, etc.) to get humid water vapour which will then be dehumidified (condensed) to get freshwater (Fig. 1.22). As it is one of the reliable and economical techniques to obtain freshwater, research is being done to enhance the performance of the system adopting a parametric investigation of working fluids (air and water) used in the system. For the system which uses air as the working fluid, the current research focusses on increasing the overall efficiency by preheating the air thereby reducing the work done to increase the temperature of the heat transfer fluid, adding baffles to increase

Solar energy technologies: principles and applications Chapter | 1

35

FIGURE 1.22 Schematic of humidifier and dehumidifier desalination system.

FIGURE 1.23 Multistage flash desalination system.

the heat transfer rate through effective heat transfer mechanisms and so on. Whereas for the system which uses water as working fluid, the current research focusses on modifications in absorber tube (coating, design, mechanism), integrating multiple heaters, collectors and so on, to enhance the performance of humidification and dehumidification system.

1.7.1.2 Multistage flash desalination In MSF desalination, the saline water temperature is increased more than its saturation temperature with a flash stepwise setup that reduces the pressure and thus maintaining low pressure at each successive stage (with the help of vacuum pump) to form vapour which will then be condensed to obtain freshwater (Fig. 1.23). The main drawback of this technique is that it requires a large amount of energy as an input which is common in the forms of thermal and electrical energy. Also maintaining a vacuum in the system is much complicated. Nowadays the required energy (thermal and electrical) is being assisted from fossil fuels which results in increasing the carbon footprints. Hence the research in the present scenario is being done to assist the required thermal energy by integrating the system with nuclear power plants, solar power plants (solar collectors and concentrators).

36

PART | I Technologies

1.7.1.3 Vapour compression desalination The heat from solar radiation is used to heat the saline water producing a vapour which is then compressed thermally or mechanically to raise its pressure and temperature. The high pressurized resultant fluid heats the feed water in the rest of the stages. Since the desalination is obtained by compressing the vapour, this type of desalination is called a vapour compression desalination technique (Fig. 1.24). The main drawback of this technique is that it required high-grade energy for compression. Hence the recent research focusses on reducing the compressor energy by substituting additional support from solar concentrators, collectors to boost its performance. Going towards solar energy substitutes reduces the amount of carbon dioxide emission using traditional compressing techniques. 1.7.1.4 Osmotic desalination driven by solar energy There are two types of osmotic desalination techniques named reverse and forward osmosis. Reverse osmosis is one of the most dominating desalination techniques which provide more than 65% of drinking water worldwide. This is done by a semipermeable membrane to remove the salinity from the saline water. Due to the osmotic pressure, the solvent is moved from low solute concentration to high solute concentration separating freshwater from the saline water. When a pressure more than the osmotic pressure is applied on the other side, the freshwater is collected out. It requires approximately 4 kWh of energy to yield a 1 m3 of distilled water and hence it is termed as an energyexhaustive process. The research around the world is being done to reduce the specific energy consumption by powering the pressure through SPV and organic Rankine cycle. Recent research focusses on improving the performance of the solar organic Rankine cycle by varying the operating temperature, configuration, working fluid, pressure assisting technology and so on.

FIGURE 1.24 Vapour compression desalination system.

Solar energy technologies: principles and applications Chapter | 1

37

FIGURE 1.25 Schematic of (A) forward and (B) reverse osmosis desalination process.

Another osmotically driven desalination technique that consumes less energy is forward osmosis (Fig. 1.25). It also has membranes (semipermeable) to purify the saline water from disbanded solutes by an osmotic pressure grade developed by a draw solution that has higher concentration as compared to the feed water through the diffusion process. Once the draw solutes separate the salinity from the feed saline water, it is further separated (second separation) to get the desalinated water from draw solution again by evaporation and freezing process and thus regenerating the draw solution again to the system. The second separation process is an energy-consuming process in this whole system thus researchers were focusing on reducing the energy consumption to separate pure water from the draw solution stepping ahead to rely on solar power. Further research and development are certainly required to attain a momentum for this technology using solar power.

1.7.2

Direct type desalination

Solar still utilizes direct solar radiation from the Sun to desalinate saline water. It works on the principle of evaporation and condensation process. The still, consisting of a basin (where the saline water is fed) is fully insulated along all its sides and closed with the transparent glass cover to permit the solar energy. As the radiation strikes the saline water in the basin through the glass cover, evaporation of water occurs producing the vapour from the water upwards which will be trapped by the glass cover above the basin and thus the condensate is formed in the lower side the glass cover and are collected as the distillate yield (Fig. 1.26). Even though solar stills are simple and reliable, the main drawback of solar stills is its productivity. Research is being done to augment the yield by integrating solar stills with fin, SPV system and collectors to improve the

38

PART | I Technologies

FIGURE 1.26 Schematic of a solar still desalination system.

heat transfer and to preheat the brine water which enters inside the basin keeping in mind that preheating the saline water increases the distillate yield. The research was also done by modifying the structure of solar still with hemispherical structure, tubular structure, pyramid structure and so on, to analyze the productivity. However, single slope conventional solar still seems to be good in the technoeconomic view annually. Energy storage techniques (sensible and latent) and nanocomposites were also integrated with solar still to obtain the distillate yield in the nocturnal hours which resulted in an increase in the productivity yielding freshwater throughout the night depending on the sensible and latent heat energy storage material according to the ambient conditions.

Nomenclature E h c ni q k T V GT Acell Popt I Io (T) V

Photon energy for its frequency, ν, or its wavelength, λ Planck constant The speed of light The density of electrons or holes in intrinsic (impurity-free) material The charge of an electron The Boltzmann constant The junction temperature (K) The applied voltage across the junction from the p-side to the n-side Solar irradiance (W/m2) Cell area (m2) Expressed (W) Real current flowing outside the cell Diode reverse bias saturation current (dependant of the temperature, T) Voltage difference across the terminals of the cell

Solar energy technologies: principles and applications Chapter | 1

39

References [1] Moriarty P, Honnery D. Global renewable energy resources and use in 2050. In: Letcher TM, editor. Managing global warming: an interface of technology and human issues. London, UK: Academic Press; 2018. p. 221
35. [2] Bell RB, Fernandes RP, Andersen PE, editors. Index. Oral, head and neck oncology and reconstructive surgery. New York, NY: Elsevier; 2018. p. 919
43. [3] Chapin D-M, Fuller C-S, Pearson G-L. A new silicon p-n junction photocell for converting solar radiation into electrical power. J Appl Phys 1954;25(5):676
7. [4] Chapin DM. Solar energy converting apparatus. US Patent 2,780,765. 1957. [5] NREL. Best research-cell efficiency chart, ,https://www.nrel.gov/pv/cell-efficiency. html.. [6] Jayawardena KDGI, Rozanski LJ, Mills CA, Beliatis MJ, Nismy NA, Silva SRP. ‘Inorganics-in-organics’: recent developments and outlook for 4G polymer solar cells. Nanoscale 2013;5(18):8411
27. [7] Surampudi R, Blosiu J, Stella P, Elliott J, Castillo J, et al. Solar power technologies for future planetary science missions. Washington, D.C.: National Aeronautics and Space Administration (NASA). 2017. [8] Messenger R, Mcconnell R. Handbook of energy efficiency and renewable energy. Choice Rev Online 2008;45(05) 45-2629-45
2629. [9] Yan J, Saunders BR. Third-generation solar cells: a review and comparison of polymer: fullerene, hybrid polymer and perovskite solar cells. RSC Adv 2014;4(82):43286
314. [10] Wang R, Mujahid M, Duan Y, Wang Z-K, Xue J, Yang Y. A review of perovskites solar cell stability. Adv Funct Mater 2019;29(47):1808843. [11] Shi Z, Jayatissa AH. Perovskites-based solar cells: a review of recent progress, materials and processing methods. Materials (Basel) 2018;11(5). [12] Kim BJ, Lee S, Jung HS. Recent progressive efforts in perovskite solar cells toward commercialization. J Mater Chem A 2018;6(26):12215
36. [13] Ma J, Chang J, Lin Z, Guo X, Zhou L, Liu Z, et al. Elucidating the roles of TiCl4 and PCBM fullerene treatment on TiO2 electron transporting layer for highly efficient planar perovskite solar cells. J Phys Chem C 2018;122(2):1044
53. [14] Dong H, et al. Improving electron extraction ability and device stability of perovskite solar cells using a compatible PCBM/AZO electron transporting bilayer. Nanomaterials 2018;8(9). [15] Pang S, et al. Efficient bifacial semitransparent perovskite solar cells using Ag/V2O5 as transparent anodes. ACS Appl Mater Interfaces 2018;10(15):12731
9. [16] Jung EH, et al. Efficient, stable and scalable perovskite solar cells using poly(3-hexylthiophene). Nature 2019;567(7749):511
15. [17] Jeon NJ, et al. Compositional engineering of perovskite materials for high-performance solar cells. Nature 2015;517(7535):476
80. [18] Rangel-C´ardenas J, Sobral H. Optical absorption enhancement in CdTe thin films by microstructuration of the silicon substrate. Materials (Basel) 2017;10(6). [19] Berhe TA, et al. Organometal halide perovskite solar cells: degradation and stability. Energy Env Sci 2016;9(2):323
56. [20] Chen W, et al. Efficient and stable large-area perovskite solar cells with inorganic charge extraction layers. Science 2015;350(6263):944
8. [21] Raga SR, Jung M-C, Lee MV, Leyden MR, Kato Y, Qi Y. Influence of air annealing on high efficiency planar structure perovskite solar cells. Chem Mater 2015;27(5):1597
603.

40

PART | I Technologies

[22] Unger EL, et al. Hysteresis and transient behavior in current
voltage measurements of hybrid-perovskite absorber solar cells. Energy Env Sci 2014;7(11):3690
8. [23] Ba¨tzner DL, Romeo A, Terheggen M, Do¨beli M, Zogg H, Tiwari AN. Stability aspects in CdTe/CdS solar cells. Thin Solid Films 2004;451
452:536
43. [24] Niki S, et al. CIGS absorbers and processes. Prog Photovolt Res Appl 2010;18 (6):453
66. [25] Lucen˜o-S´anchez JA, D´ıez-Pascual AM, Capilla RP. Materials for photovoltaics: state of art and recent developments. Int J Mol Sci 2019;20(4). [26] Feurer T, et al. Progress in thin film CIGS photovoltaics—research and development, manufacturing, and applications. Prog Photovolt Res Appl 2017;25(7):645
67. [27] Singh R, Polu AR, Bhattacharya B, Rhee H-W, Varlikli C, Singh PK. Perspectives for solid biopolymer electrolytes in dye sensitized solar cell and battery application. Renew Sustain Energy Rev 2016;65:1098
117. [28] Carella A, Borbone F, Centore R. Research progress on photosensitizers for DSSC. Front Chem 2018;6:481. [29] Pan Z, Zhao K, Wang J, Zhang H, Feng Y, Zhong X. Near infrared absorption of CdSexTe1
x alloyed quantum dot sensitized solar cells with more than 6% efficiency and high stability. ACS Nano 2013;7(6):5215
22. [30] Reed MA, Randall JN, Aggarwal RJ, Matyi RJ, Moore TM, Wetsel AE. Observation of discrete electronic states in a zero-dimensional semiconductor nanostructure. Phys Rev Lett 1988;60(6):535
7. [31] Brus L. Electronic wave functions in semiconductor clusters: experiment and theory. J Phys Chem 1986;90(12):2555
60. [32] AbouElhamd AR, Al-Sallal KA, Hassan A. Review of core/shell quantum dots technology integrated into building’s glazing. Energies 2019;12(6). [33] Bera D, Qian L, Tseng TK, Holloway PH. Quantum dots and their multimodal applications: a review. Materials (Basel) 2010;3(4):2260
345. [34] Tian J, Cao G. Semiconductor quantum dot-sensitized solar cells. Nano Rev 2013;4 (1):22578. [35] Schaller RD, Klimov VI. High efficiency carrier multiplication in PbSe nanocrystals: implications for solar energy conversion. Phys Rev Lett 2004;92(18):186601. [36] Schaller RD, Sykora M, Pietryga JM, Klimov VI. Seven excitons at a cost of one: redefining the limits for conversion efficiency of photons into charge carriers. Nano Lett 2006;6 (3):424
9. [37] Smith AR, Chao K-J, Niu Q, Shih C-K. Formation of atomically flat silver films on GaAs with a ‘silver mean’ quasi periodicity. Science 1996;273(5272):226
8. [38] Goetzberger A, Wittwer V. Fluorescent planar collector-concentrators: a review. Solar Cell 1981;4(1):3
23. [39] Almosni S, et al. Material challenges for solar cells in the twenty-first century: directions in emerging technologies. Sci Technol Adv Mater 2018;19(1):336
69. [40] Sogabe T, Shen Q, Yamaguchi K. Recent progress on quantum dot solar cells: a review. J Photonics Energy 2016;6(4):1
27. [41] Balaji K, Iniyan S, Muthusamyswami V. Experimental investigation on heat transfer and pumping power of forced circulation flat plate solar collector using heat transfer enhancer in absorber tube. Appl Therm Eng 2017;112:237
47. [42] Kalogirou SA, Karellas S, Braimakis K, Stanciu C, Badescu V. Exergy analysis of solar thermal collectors and processes. Prog Energy Combust Sci 2016;56:106
37.

Solar energy technologies: principles and applications Chapter | 1

41

[43] Poongavanam GK, Kumar B, Duraisamy S, Panchabikesan K, Ramalingam V. Heat transfer and pressure drop performance of solar glycol/activated carbon based nanofluids in shot peened double pipe heat exchanger. Renew Energy 2019;140:580
91. [44] Michael JJ, Iniyan S, Goic R. Flat plate solar photovoltaic
thermal (PV/T) systems: a reference guide. Renew Sustain Energy Rev 2015;51:62
88. [45] Kalogirou SA. Solar thermal collectors and applications. Prog Energy Combust Sci 2004;30(3):231
95. [46] Petrullo M, Jones SA, Morton B, Lorenz A. World green building trends 2018—smart market report. 2018. [47] Sadineni SB, Boehm RF. Measurements and simulations for peak electrical load reduction in cooling dominated climate. Energy 2012;37(1):689
97. [48] Zeyghami M, Goswami DY, Stefanakos E. A review of solar thermo-mechanical refrigeration and cooling methods. Renew Sustain Energy Rev 2015;51:1428
45. [49] Anand S, Gupta A, Tyagi SK. Solar cooling systems for climate change mitigation : a review. Renew Sustain Energy Rev 2015;41:143
61. [50] Jani DB, Mishra M, Kumar P. A critical review on application of solar energy as renewable regeneration heat source in solid desiccant—vapor compression hybrid cooling system. J Build Eng 2018;18:107
24. [51] Prieto A, Knaack U, Auer T, Klein T. COOLFACADE : state-of-the-art review and evaluation of solar cooling technologies on their potential for fac¸ade integration. Renew Sustain Energy Rev 2019;101:395
414. [52] Sarbu I, Sebarchievici C. Review of solar refrigeration and cooling systems. Energy Build 2013;67:286
97. [53] Tassou SA, Lewis JS, Ge YT, Hadawey A, Chaer I. A review of emerging technologies for food refrigeration applications. Appl Therm Eng 2010;30(4):263
76. [54] Sun DW, Zheng L. Vacuum cooling technology for the agri-food industry: past, present and future. J Food Eng 2006;77(2):203
14. [55] Allouhi A, Kousksou T, Jamil A, Bruel P, Mourad Y, Zeraouli Y. Solar driven cooling systems : an updated review. Renew Sustain Energy Rev 2015;44:159
81. [56] Anyanwu EE. Design and measured performance of a porous evaporative cooler for preservation of fruits and vegetables. Energy Convers Manag 2004;45(13
14):2187
95. [57] Kim DS, Infante Ferreira CA. Solar refrigeration options—a state-of-the-art review. Int J Refrig 2008;31(1):3
15. [58] Sharma VK, Kumar EA. Studies on la based intermetallic hydrides to determine their suitability in metal hydride based cooling systems. Intermetallics 2015;57:60
7. [59] Syed SN, Sharma VK. Thermodynamic simulation of CO2: adsorbent based sorption refrigeration system. FME Trans 2020;48(2):187-94. [60] Sharma R, Kumar EA. Thermodynamic analysis of advanced resorption cooling/heating systems based on NH3
halide salts using measured PCIs. Int J Refrig 2018;105:109
19. [61] Jani DB, Mishra M, Sahoo PK. A critical review on application of solar energy as renewable regeneration heat source in solid desiccant—vapor compression hybrid cooling system. J Build Eng 2018;18:107
24. [62] Aresti L, Christodoulides P, Florides G. A review of the design aspects of ground heat exchangers. Renew Sustain Energy Rev 2018;92:757
73. [63] Goetzler W, Zogg R, Lisle H, Burgos J, Navigant Consulting Inc. Ground-source heat pumps : overview of market status, barriers to adoption, and options for overcoming barriers. Oak Ridge, TN: USDOE Office of Energy Efficiency and Renewable Energy (EERE); 2009. http://dx.doi.org/10.2172/1219308.

42

PART | I Technologies

[64] Bayer P, Saner D, Bolay S, Rybach L, Blum P. Greenhouse gas emission savings of ground source heat pump systems in Europe: a review. Renew Sustain Energy Rev 2012;16(2):1256
67. [65] ODPM. Low or zero carbon energy sources: strategic guide. London, UK: NBS, Office of the Deputy Prime Minister. 2006; p. 28. [66] Haehnlein S, Bayer P, Blum P. International legal status of the use of shallow geothermal energy. Renew Sustain Energy Rev 2010;14(9):2611
25. [67] Sivasakthivel T, Murugesan K, Sahoo PK. A study on energy and CO2 saving potential of ground source heat pump system in India. Renew Sustain Energy Rev 2014;32:278
93.

Chapter 2

Bioenergy for better sustainability: technologies, challenges and prospect Senqing Fan1, Jingyun Liu1, Xiaoyu Tang2 and Zeyi Xiao1 1

School of Chemical Engineering, Sichuan University, Chengdu, P.R. China, 2Biogas Institute of Ministry of Agriculture, Chengdu, P.R. China

Chapter Outline 2.1 Introduction 2.2 Technologies 2.2.1 Microorganisms 2.2.2 Feedstocks

2.1

43 45 45 48

2.2.3 Fermentation technologies 2.3 Challenges 2.4 Future prospects References

54 61 62 63

Introduction

The increasing industrialization and motorization of the world has led to a steep rise for the demand of petroleum-based fuels and the progressive depletion of conventional fossil fuels with increasing energy consumption and greenhouse gas (GHG) emissions have led to a move toward renewable and sustainable energy sources. Biomass has significant potential to boost energy supplies for people, and it can be directly burned for heating or power generation. As a renewable resource, biomass could be sustainably developed in the future. Biomass is carbon neutral, appearing to have formidably positive environmental properties. Biomass can be converted into liquid fuels or gas substitutes made from plant matter and residues, such as municipal wastes, agricultural crops and forestry by-products. Several energy products can be obtained by various biomass feedstocks. In this section, we will describe the bioalcohols and biogas by fermentation technology. Ethanol and butanol are the typical bioalcohols, which can be used as transportation fuel. Ethanol could be combined and blended with petrol within unmodified spark-ignition engines or burned in its pure form within modified spark-ignition engines. Gasoline
ethanol mixtures, E5 (5 wt.% Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00002-9 © 2021 Elsevier Inc. All rights reserved.

43

44

PART | I Technologies

ethanol and 95 wt.% gasoline), E15 (15 wt.% ethanol and 85 wt.% gasoline) and E85 (85 wt.% ethanol and 15 wt.% gasoline) are typical ones used in transport sector. Butanol can be blended with gasoline and transported in the present pipeline. It can also be directly used in existing motor engines and vehicular infrastructures without mechanical tailoring [1]. The ethanol fermentation technology originated thousands of years ago in China for alcohol consumption. Biobutanol fermentation was first applied during World War I. At that time the fermentation substrate was corn, which was also the food for human beings. Because of the wide production of bioethanol and biobutanol based on corn biomass, the debate of “food versus fuels” was fiercer. Meanwhile the generation of bioalcohols by chemical way was more popular for its low cost and so the bioalcohols by fermentation was declined gradually. In the recent several years the bioalcohols production by fermentation was come up again because of the fossil energy crisis (Table 2.1). Biogas is a mixture of gas mainly consisting of methane (CH4) and CO2. It is produced using modern bioenergy technologies for anaerobic digestion of a series of biomass resources, including the organic food waste, municipal sewage wastewater, sewage, industrial organic effluents, energy crops, and agriculture residues (crop straw, livestock and poultry manures, horticultural residues, etc.). Biogas can be purified into biomethane by separating the CO2 and other gases. Using separation and purification to obtain biomethane could enable more convenient utilization to fuel transportation and for injection into the natural gas network [2,3]. Besides biogas after simple desulfurization and drying can also be converted to electricity and heat in cogeneration units (combined heat and power), or the biogas is burnt to produce heat. Thus biogas and biomethane produced from biogas are flexible renewable fuels that can be stored. Motor fuel, electricity, and heat can be produced from them, which make them important functions in the context of sustainable energy supply. With the increasing demand for energy, many countries have put efforts on the development of biogas technology and

TABLE 2.1 Some properties of bioalcohols and gasoline. Bioethanol

Biobutanol

Gasoline

Calorific value (MJ/L)

21.2

29.2

32.5

Air
fuel ratio

9

11.2

14.6

Evaporation heat (MJ/kg)

0.92

0.43

0.36

Theoretical octane number

129

96

91
99

Engine octane number

102

78

81
89

Soluble in water

Soluble

Insoluble

Insoluble

Bioenergy for better sustainability Chapter | 2

45

consider it as an alternative solution while facing the energy crisis caused by the depletion of fossil resources. Besides over use of traditional fossil fuel will contribute to global warming due to the GHG emission. In this case, biogas, as the modern biofuel based on biotechnology, has huge potential on environmental improvement by mitigating GHG emissions and is regarded as clean energy resource. Biogas production has gradually become industrialization and commercialization as energy demand and environment protection becomes more urgent. Traditionally biogas was principally yielded in small, household-based digesters for cooking and heating in developing countries. In contrast the expansion of biogas in developed countries has been industrially driven as a commercial medium- and large-scale biogas plants (a single digester volume is more than 300 m3, and total volume is more than 300 m3) [4]. Various support programs have been carried out to develop biogas production as one of the most sustainable alternative energy sources to reduce consumption and reliance on fossil fuels, mitigate the fossil fuel crisis, cut carbon emissions and hazardous materials, avoid deforestation, and improve soil fertility.

2.2 2.2.1

Technologies Microorganisms

The common microorganisms used for ethanol fermentation are Saccharomyces cerevisiae and Zymomomas mobilis. Ethanol fermentation by S. cerevisiae is a process in which the cells use its own enzyme system to convert monosaccharide or disaccharide into ethanol and carbon dioxide through anaerobic respiration. At the same time, the ATP required for its own life activities is also produced. The metabolic process of ethanol fermentation by S. cerevisiae can be approximately divided into four stages. In the first stage glucose is phosphorylated to produce unstable fructose-1,6diphosphate. In the second stage fructose-1,6-diacid is split into two molecules of glyceraldehyde-3-phosphate. In the third stage glyceraldehyde3-phosphate is oxidized to produce pyruvate and release energy. In the fourth stage pyruvate continues to degrade to produce aldehydes, which is further produced ethanol by reduction reaction. The metabolism of glucose in Z. mobilis can be divided into five stages. In the first stage glucose generates 6-phosphogluconic acid under the action of high energy phosphate bond in ATP. In the second stage 6-phosphogluconic acid generates 2-keto-3deoxy-6-phosphogluconic acid under the action of dehydratase. In the third stage 2-keto-3-deoxy-6-phosphogluconic acid generates 2-keto-3-deoxy-6phosphogluconic acid under the action of deoxyketosylate aldolase. In the fourth stage like the Embden-Meyerhof-Parnas pathway in yeast cells, glyceraldehyde-3-phosphate is oxidized to form pyruvate and release energy.

46

PART | I Technologies

In the fifth stage pyruvate generated in the third and fourth stages is degraded to acetaldehyde, which is further reduced to ethanol. The conventional microorganism applied in the fermentation for butanol is Clostridium, it belongs to the class Schizophyllum
Eubacteriae
Eubacteriae
Bacillus
Clostridium [5
9]. Different Clostridium strains are slightly different in morphological changes and metabolite composition but butanol is almost always included in. Here the Clostridium acetobutylicum was taken as the example. C. acetobutylicum is strictly or moderately anaerobic, spore-forming, with movability, and Gram-positive bacteria. The common morphology of C. acetobutylicum is short-rod or series-rod shape (0.5
0.7 μm) 3 (2.6
4.7 μm) and endospore (dormant bodies of microorganisms and able to keep their vitality against adverse environment). Endospore is usually formed after 3 days. Suitable growth condition is temperature 30 C
37 C and pH 5.6
7.0 (4.3
4.8 at the later stage of the fermentation process). There are some starch granules inside of the cells and so the cell can be dyed dark blue by iodine solution. Generally speaking there are two stages in the metabolic process of Clostridium, acidogenesis and solventogenesis (Fig. 2.1). In the body of the cells the glucose as the initial carbon source in the pathway. After glycolysis the glucose is converted into pyruvate and then the acidogenesis begins. During this stage the pyruvate and CoA is converted into acetyl-CoA in the catalysis of pyruvate-ferredoxin oxidoreductase with the production of hydrogen. AcetylCoA is converted into acetyl phosphate under the catalysis of phosphotransacetylase, and then acetic acid is generated under the catalysis of acetate kinase, accompanied by the production of ATP. Part of acetyl-CoA is catalyzed into butyryl-CoA in the existing of thiolase(acetyl-CoA acetyltransferase), 3hydroxybutyryl-CoA dehydrogenase, crotonase, and butyryl-CoA dehydrogenase. Then the butyryl-CoA is converted to butyryl phosphate with the catalysis of phosphobutyryltransferase. Finally butyryl phosphate was dephosphorylated under the catalysis of butyrate kinase and the butyric acid is formed. During this stage the pH of the fermentation broth is decreased gradually owing to the accumulation of acetic acid and butyric acid. The onset of solventogenesis involves a switch in the carbon flow from the acid-producing pathways to the solvent-producing pathways. During the stage the acetyl-CoA is partly converted to acetaldehyde under the catalysis of acetaldehyde dehydrogenase and then catalyzed to ethanol by ethanol dehydrogenase. Part of the acetyl-CoA is firstly turned into acetoacetyl-CoA by thiolase (acetyl-CoA acetyltransferase) and then converted to acetone with the catalysis of acetoacetyl-CoA transferase and acetoacetate decarboxylase. Butyryl-CoA is converted to butyraldehyde under the catalysis of butyraldehyde dehydrogenase, and then catalyzed to butanol by butanol dehydrogenase. At the same time, part of the acetic acid and butyric acid generated during the acidogenesis is converted to acetyl-CoA or butyrylCoA and participate in the production of butanol under the catalysis of

Bioenergy for better sustainability Chapter | 2

47

FIGURE 2.1 The metabolite pathway of the Clostridium acetobutylicum.

acetoacetyl-CoA:acetate/butyrate:CoA transferase. Generally the ratio of final acetone, butanol, and ethanol concentration from fermentation is 3:6:1 and the ratio of carbon dioxide and hydrogen is 6:4 [10]. Different from alcohols production by a certain kind of microorganisms, biogas fermentation is taken place by several kinds of microorganisms. The microorganisms involved with biogas fermentation can be classified as “fermentation bacteria,” “acid-producing bacteria,” and “methane producing bacteria.” Fermentation bacteria mainly play a role in the hydrolysis and acid production stage. Extracellular enzymes produced by fermentation bacteria can hydrolyze insoluble organic matters such as carbohydrates, lipids, and proteins into soluble organic matters such as polysaccharides, long-chain fatty acids, and amino acids. Fermenting acid-producing bacteria can also decompose

48

PART | I Technologies

soluble organic matter into volatile organic acids, alcohols, and ketones. Fermentation bacteria can not completely degrade carbohydrate, fat, and other organic matter, and will produce organic acids, solvents, vitamins, hydrogen, and other by-products. Methanogens are strictly anaerobic bacteria. According to the types of substrate utilization, methanogens can be divided into three categories: hydrogen nutrition bacteria, acetic nutrition bacteria, and methyl nutrition bacteria. Hydrogen nutrition bacteria utilize hydrogen as electron acceptor to reduce carbon dioxide to methane. Hydrogen is produced by the metabolism of other microorganisms in the system, such as clostridia. Acetic nutrition bacteria utilize acetic acid to produce methane and carbon dioxide. During biogas fermentation system, more than 70% of the methane is produced by this pathway. Methyl nutrition bacteria can utilize the substrate containing methyl group to produce methane. For the time being there are six kinds of bacteria being able to produce methane: methanobacteriales, methanococcales, methanocellales, methanomicrobiales, methanopyrales, and methanosarcinales.

2.2.2

Feedstocks

The bioalcohols can be divided into three generations according to the feedstocks. Currently the fermentation for bioalcohols is mainly based on firstgeneration feedstock-starch such as maize, wheat, rice, corn, and cassava which could be utilized directly or the hydrolysates-monosaccharide. The United States and Brazil are the two main suppliers of the first-generation bioethanol. While the high cost of the starch and the food/feed versus fuel debate make the bioalcohols production by fermentation less competitive to be widely used. Molasses, a by-product of sugar industry, is mainly composed of sucrose (25%
35%), glucose & fructose (15%
25%), colloid (5%
12%), ash (5%
16%), total nitrogen (2%
5%), and water (20%
30%). Cheese whey, a liquid effluent obtained from the cheese manufacturing process, presents high volumetric production and high organic loads. Typical chemical oxygen demand of it is 50
102 kg/m3 and the biological oxygen demand value is 27
60 kg/m3 [11]. This relatively high organic load is mainly owing to its lactose content (48
60 g/L), which may be extremely low to be used in most industrial fermentation processes but works well in the biobutanol production. Application of the above substrate would avoid the food/feed versus fuel debate to some degree (Table 2.2). TABLE 2.2 World bioethanol production in 2005 and 2006 (million ton). Year

United States

Brazil

China

India

France

Others

2005

12

12

0.8

0.24

0.12

1.24

2006

14.6

14

0.8

0.24

0.2

1.32

Bioenergy for better sustainability Chapter | 2

49

The second-generation bioalcohols production technologies that use energy dense lignocellulosic biomass are under development. Lignocellulosic biomass is considered to be a renewable feedstock, and it is the most abundant carbohydrate on Earth [12]. There are various lignocellulosic biomass such as agricultural and agroindustrial residues, forestry waste, and others. As reported lignocellulosic biomass has a complex structure mainly composed of three polymers
cellulose (30%
50%), hemicellulose (20%
40%), and lignin (15%
25%). Cellulose is a polysaccharide composed of linear chains of hundreds to thousands of glucose units. Hemicellulose is a heterogeneous polymer composed of several different types of monosaccharides, which are pentasaccharides and hexasaccharides, including xylose, arabinose, mannose, and galactose. The specific composition varies with the cell tissue and type of plants. Cellulose is entrapped by hemicellulose and lignin forming a relatively stable structure. Some microorganisms are able to utilize cellulose and hemicellulose directly while most microorganisms including clostridia are not [13,14]. Thus before the utilization, the lignocellulose has to be pretreated. The general pretreatment methods include solvent fractionation, acid treatment, steam explosion, alkali treatment, oxidation treatment, enzymatic hydrolysis, and so forth. However, the pretreatment of lignocellulose is complicated and costing for the rigid structure caused by lignin (Fig. 2.2). The promising third-generation feedstock is microalgae. It is mainly used as the feedstock for the biodiesel (Fig. 2.3). Studies showed that the residue of the microalgae after the production of biodiesel was still rich in carbohydrate such as starch and cellulose and have the possibility to be used as the fermentation substrate for bioalcohols. Also the microalgae have to be pretreated by enzymolysis, hot alkali, or ultrasound before used as the carbon source for fermentation. Nevertheless, the pretreatment of microalgae is simpler than that of the lignocellulose because of the absence of lignin in microalgae. Several features of algal physiology are relevant for evaluating their possible incorporation into renewable biofuel applications. The advantages of microalgae can be summarized as follows: (1) the solar energy accumulated in algae could be up to 6
12 times that of terrestrial plants because they are inherently more efficient solar energy converters (3%
8% greater than terrestrial); (2) unlike terrestrial, the nonexistent of intractable biopolymers relieves the need for pretreatments to breakdown cellular products; (3) the metabolic and ecological diversity of them allows selection of taxa which are suitable for their growth in locally available culture environment or have morphological characters that allow cost-effective harvesting; and (4) manipulating the end-products of them through the biosynthetic regulation of chemical composition by nutrient and environmental stresses [15]. The feedstocks for biogas can be divided into three kinds: municipal waste, industrial waste, and agricultural waste. China’s first central sewage treatment plant was built in Shanghai in 1923. In the 1950s central sewage

50

PART | I Technologies

FIGURE 2.2 Schematic representation of bioalcohols by fermentation from corn, sugar, and cellulose. CGM, Corn Gluten Meal; CGF, Corn Gluten Forage; DDGS, Distillers Dried Grains with Solubles.

FIGURE 2.3 Potential pathways from microalgae to fuels.

Bioenergy for better sustainability Chapter | 2

51

treatment plants in Xi’an, Fushun, Anshan, and Chengdu built biogas plants for the treatment of residual sludge, the primary municipal waste used in biogas production (along with food waste). Sludge yield is 2.3
7.0 m3 sludge (moisture content 96%) per thousand cubic meters of sewage in China, an average yield of 3.59 m3 per thousand cubic meters of sewage, or 0.14 tons of dry sludge dry sludge per thousand cubic meters of sewage. By 2010, 2842 central sewage treatment plants had been built in cities and towns throughout China, with a total treatment capacity of 127.6 million m3/day, and producing more than 20 million tons/day of sludge with moisture content of 80% [16]. The sludge from sewage treatment plants has high moisture content, large volume, and organic matter content as high as 50%
70%. Adoption of biogas technology can help to reduce and stabilize sludge, while simultaneously recovering methane. The sewage treatment plants applying anaerobic digestion technology to treat sludge have a treatment capacity of sewage ranging from 25 to 1000 thousand m3/day, including 8 small sewage treatment plants (treatment capacity , 50 thousand m3/day), 7 middle-sized sewage treatment plants (treatment capacity of 50
100 thousand m3/day), 20 large sewage treatment plants (treatment capacity of 100
400 thousand m3/ day), and 11 super-large sewage treatment plants (treatment capacity $ 400 thousand m3/day). Anaerobic digestion units were operating in 25 sewage treatment plants, and were under construction or starting in 6 more plants. Fifteen sewage treatment plants had built anaerobic digestion units, but they were not running or were shut down in five small plants, one middle-sized, eight large, and one super-large sewage treatment plant. Surveyed plants cited the following reasons for nonfunctioning units: complexity of the process, missing equipment, understaffing, poor management, and reduced biogas and electricity production, which resulted in no significant benefits [17]. Most anaerobic digestion units covered in the survey used the continuous stirred tank reactor (CSTR) process at mesophilic temperatures. Biogas produced from the treatment of sludge digestion was mainly used for heating of the digester and as domestic fuel for the sewage treatment plants. Six sewage treatment plants used biogas for electricity generation, and four sewage treatment plants used biogas directly for fuel for the blower [17]. Food waste is also used for biogas production. Cities in China produce about 60 million tons of food waste every year. Since 2010, the Chinese government has carried out a development model that supported test projects for food waste treatment in more than 60 pilot cities throughout the country. The construction of plants for the treatment of food waste is undergoing rapid development. Many cities are actively exploring treatment methods for food waste and biogas technology has achieved good results to this end. In 1967 the first biogas plant for the treatment of distillery wastewater was established in the Nanyang alcohol plant, Henan province, China, with a digester volume of 2 3 2850 m3, a treatment capacity of 500 m3/day, and a biogas output of 10,000 m3/day [18]. In the treatment of industrial waste,

52

PART | I Technologies

biogas technology is mainly used to treat organic wastewater with high concentrations of organic materials, such as distillery wastewater, pharmacy wastewater, food process wastewater, starch wastewater, and slaughterhouse wastewater. Alcohol wastewater, starch wastewater, and winery waste water together account for approximately 71% of biogas production. Alcohol wastewater treatment alone accounts for 38.4% of all production, followed by starch and glucose industry wastewater, which accounts for 18.0%. Finally there is winery wastewater, accounting for 11.6% of production. The application of biogas technology in slaughterhouse wastewater, citric acid wastewater, pharmaceutical wastewater, beverage waste water, molasses alcohol wastewater, monosodium glutamate wastewater, petrochemical wastewater, and papermaking wastewater represented 5.7%, 5.2%, 4.9%, 3.9%, 3.7%, 3.4%, 3.0%, and 2.2% of production, respectively [19]. The upflow anaerobic sludge blanket (UASB) is the most popular technology for anaerobic treatment, accounting for 51% of all anaerobic processes in the survey plants. This proportion was close to the world average value of 55%. However, in China approximately 32% of biogas plants still use CSTR or anaerobic contact (AC) processes because wastewater with high concentrations of suspended solids is not beneficial to the formation of granular sludge in UASB. Expansion granular sludge beds and internal circulation anaerobic reactors have been in rapid development in recent years, accounting for approximately 11% of all anaerobic processes [19]. Data calculated from a statistical summary in 2014 from the Department of Education, Science & Technology, Ministry of Agriculture showed that the annual average volumetric biogas production rate of biogas plants treating industrial waste was 1.05 m3/m3/day, much higher than that of biogas plants treating agricultural waste. Biogas produced from the treatment of industrial waste was mainly used for fuel for the boiler; the rest was used for electricity generation. Agricultural waste for biogas production is mainly animal waste (only 30 biogas plants in China use straw as feedstock). The production of livestock and poultry have moved toward larger and more specialized production units so manure quantities increased in a few concentrated areas, posing a serious threat to soil, water, air, livestock, and poultry. Biogas technology is an effective measure to solve some of these problems. The application of biogas technology for the treatment of animal waste began in the 1980s [20]. Biogas plants treating cow dung were built at the Fenghuangshan cow farm in Chengdu and an animal husbandry farm affiliated with the Zhejiang Agricultural University in Hangzhou. The former operated an underground plug flow reactor, and the latter used underground hydraulic digesters, both involve digestion at ambient temperature. Early biogas plants treating animal waste mainly used hydraulic digesters, with volumes of 20
300 m3. A plant consisted of several or even hundreds of hydraulic digesters. The digesters which were buried underground were running at ambient temperature. Therefore the efficiency of these biogas plants was low, with hydraulic

Bioenergy for better sustainability Chapter | 2

53

retention times of about 40 days, and a volumetric biogas production rate of 0.13
0.3 m3/m3/day. Moreover, it was difficult to remove digestate. However, as this process has some advantages, it is still in use in small- and medium-scale biogas plants even today. Advantages include the reduced need for pretreatment of feedstock, low area requirement (can build units below facilities housing animals), low cost, and no fuel costs. Underground hydraulic digesters have serious disadvantages such as the loss of microorganisms, difficulty in controlling the temperature, low efficiency, and the difficulty in digestate removal. In response to these problems, several new processes have been developed including ACs, over-ground plug flow reactors, and upflow blanket filter (UBF) reactors. In the late 1980s some biogas plants were built above ground with mesophilic digestion. The volumetric biogas production rate of the biogas plants reached 0.69
1.5 m3/m3/day. In the 1980s the fundamental purpose of building biogas plants was to produce biogas, with digestate as a by-product. In the 1990s with the development of large-scale livestock and poultry farms, water pollution originating from breeding operations was becoming more and more prominent. At that time, the main purpose for building biogas plants was to treat wastewater, with the production of biogas becoming a by-product. The efficiency of biogas production was not high because the influent concentration was low, and it was difficult to increase the digestion temperature. These types of plants were first developed in South China. For example, the Shenzhen agriculture and animal husbandry company built swine wastewater treatment plants in 1990 using the UBF process with an influent chemical oxygen demand (COD) of approximately 9500 mg/L and digestion occurring at ambient temperature (16
33 C). The plants reached a volumetric biogas production rate of 0.8
1.3 m3/m3/day. The wastewater treatment plants of the Hangzhou Dengta general livestock farm (built in 2000) applied a combination of the UASB process and a sequencing batch reactor to treat 3000 m3/day swine wastewater. These types of plants were also built in North China. For example, swine wastewater treatment plants built at the Ma Sanjia mechanization pig farm in Liaoning in 1994 applied the UASB process with influent COD of 4000
5000 mg/L [digested at ambient temperature (9 C
13 C)], and achieved a volumetric biogas production rate of 0.2
0.3 m3/m3/day [21]. After the year of 2000, in response to an energy shortage, many biogas plants were built with biogas production as their primary objective. The biogas plants used animal manure as feedstock digested at mesophilic temperatures and at high concentration (Total solid . 8%), applying the process of CSTR, they reached a volumetric biogas production rate of more than 1.0 m3/m3/day. The biogas was used to generate electricity, and the surplus heat from the generator was applied to heat the digester. Biogas plants for agricultural waste are classified as small, medium, large, and super-large, with biogas production of 5
150, 150
500, 500
5000, and more than 5000 m3/day, respectively [22]. Data calculated from a 2014 statistical

54

PART | I Technologies

summary from the Department of Education, Science & Technology, Ministry of Agriculture showed that annual average volumetric biogas production rates of small-, medium-, large-, and super-large biogas plants treating animal waste were 0.214, 0.276, 0.512, and 0.485 m3/m3/day, respectively. The larger the scale of the biogas plants, the higher the efficiency of biogas production, which can be explained by the fact that the small- and medium-scale biogas plants use traditional processes such as the underground hydraulic digester and plug flow reactor, while large and superlarge biogas plants have adopted advanced processes such as CSTR and upflow solids reactors. Most of the biogas produced from the treatment of agricultural waste was used for domestic fuel for animal farms and neighboring farmers; only a small amount of the biogas was used for electricity generation. The number of biogas plants for the treatment of agricultural waste increased rapidly after 2006, as did the installed electricity capacity. By 2014, 102,716 biogas plants for the treatment of agricultural waste had been built in China, with a total volume of 16.25 million m3, an annual biogas output of 2.01 billion m3, an installed electricity capacity of 177.80 MW, and annual electricity production of 426.85 million kWh. Although the number of small-scale biogas plants was greatest, the large-scale biogas plant produced the largest biogas output.

2.2.3

Fermentation technologies

Batch and fed-batch fermentation are the two typical processes for biogas production, because the viscosity of the substrate is higher and the slurry cannot be effectively flowing. Bioalcohols fermentation can be operated with batch, fed-batch, or continuous mode, which depends on production capacity. In general batch or fed-batch fermentation is more suitable for alcohols production at small scale. While batch fermentation has been widely practiced in industry, fed-batch fermentation was performed mainly in laboratories [23]. Traditional batch fermentation has the advantages of low temperature requirements, simple operating conditions, high carbon source utilization efficiency, and easy to handle the contamination of bacteria and degradation of bacteria. While there were also some disadvantages in batch fermentation such as huge consumption of labor and materials and long-term duration. Besides the lag-phase in the cells’ growth process, the preculture before every batch fermentation, the washing time of machine and sterilization time indeed extend the total duration of the fermentation and thus decrease the efficiency. High concentration alcohols accumulated in the batch fermentation would do harm to cell membrane resulting in the inactive of cells. To solve the problems discussed earlier, researcher developed many kinds of fermentation process. Continuous fermentation is a good selection for alcohols production at large scale as a biofuel or bulk chemical, due to its high productivity, less

Bioenergy for better sustainability Chapter | 2

55

labor, and relatively low maintenance costs. CSTR is an improvement of the batch fermentation because of the addition process of the inflow of fresh medium and outflow of broth (Fig. 2.4). The fresh medium and fermentation broth can be mixed well with the help of the stirring blade. Thanks to the outflow of the broth, the alcohols concentration is always at a relatively low level which cannot inhibit the cell growth and so are the other metabolite. In this case, the survival environment of the cells in CSTR fermentation is much better than that of the batch fermentation. An important parameter in CSTR fermentation is the dilute rate (D), the ratio of flow rate to fermenter volume. It represents the update rate of the broth. When the D is at a high level, the broth is updated quickly. If the D is extremely high, the cell concentration in the broth will be extremely low which could decrease the carbon source consumption rate and alcohols productivity. If the D is extremely small, the metabolite will accumulate continuously which probably inhibit the growth of cells. When the D is equal to the specific cell growth rate, the fermentation environment inside the fermenter is relatively uniform and stable. Single-stage continuous fermentation may provide higher volumetric productivity but lower product concentration than batch fermentation. Besides some of the fresh medium will always flow out of the fermenter owing to the stirring resulting in the waste of the carbon source. The use of a singlestage continuous reactor does not yet seem practical at an industrial scale. Currently two- or multi-stage continuous fermentation systems are being investigated [24]. In two-stage continuous fermentation for butanol production, the first stage is generally maintained at a relatively high dilution rate for acid production and the second, at a low dilution rate for solvent production. Or the acidogenesis and solventogenesis can be controlled in different fermenter by adjusting the pH of the relative fermentation broth. Some researcher reported that the average total solvent concentration 15 g/L and productivity 0.27 g/L/h were achieved during a two-stage fermentation process (1600 hours in all) when the dilution rate of first stage was 0.12 per hour and the second stage was 0.022 per hour [25]. Besides an overall

FIGURE 2.4 The schematic diagram of the continuous stirred tank reactor (CSTR).

56

PART | I Technologies

acetone
butanol
ethanol (ABE) concentration of 25.32 g/L was obtained by C. acetobutylicum in a two-stage continuous fermentation whereas a much lower concentration of 15.98 g/L in the single-stage fermentation [26]. The cells will also be withdrawn when the broth is expelled from the fermenter during the CSTR fermentation leading to the low level cell concentration and poor fermentation performance. To solve the problem, application of cell immobilization is carried out in continuous fermentation, which results in reduction of reactor volume and operational period, thereby improving process economics [27]. During fermentation with cell immobilization, cells were fixed on the carrier materials or inside of the carrier materials by chemical or physical way [28]. Common used immobilization methods are entrapment and surface adsorption. During cell immobilization with cell entrapment, the cells are trapped in the small network space made of the insoluble gel polymer. Cell leakage can be avoided by the small network of the gel polymer and the diffusion of fresh medium and broth can be achieved by the same way. Cell immobilization by entrapment is relatively easy in terms of the operating conditions and almost no bad effect on the cell activity. Natural polymer such as k-carrageenan and calcium alginate are recognized as a nontoxic immobilization materials [29,30]. While reports said that the main drawback of natural polymers for entrapment lies in their poor physical strength and durability. Synthetic polymers such as polyvinyl alcohol offer not only higher mechanical strength and durability but also greater chemical stability in acidic and alkaline solutions and have been used as an immobilization matrix for various purposes [29,31]. The attachment of cells on the support is achieved by the physical adsorption or ionic adsorption. Physical adsorption is using the high adsorption-capacity materials such as silica gel, activated carbon, porous glass, and cellulose to attach cells. Physical adsorption is with simple operating, mild condition, and the potential of support recycling. But the cells attached on the support surface are easily detached owing to the weak combination of cell and the carrier. In the method of ionic adsorption, the combination of cell and the support is obtained by the linking of the negative charge on the cell surface and the positive charge on the support surface. Cell membrane of most microorganisms are negative charged and it offers the opportunity for the positive charged carriers to attach them. Surface attachment is a natural process where microbial cells attach to the support (adsorbent) or aggregates without the use of chemical. The commonly used immobilization methods for clostridia are gel entrapment and surface attachment [32]. While as reported, Clostridium cells are usually secret sticky substance such as proteins and polysaccharides, which are easily attached to the support. Also microscopy of the secretions showed that there are some cells attached on it apparently. Accordingly the Clostridium cells are considered more suitable to be immobilized by surface attached instead of entrapment. Besides compared with gel entrapment cell immobilization, the biofilm can be directly contacted with

Bioenergy for better sustainability Chapter | 2

57

fermentation broth, which reduces greatly the mass transfer resistance of both substrates and metabolites through entrapment materials and broth. Also it was reported that the bacterial biofilm was more tolerant to physical stress, chemical agents, and toxic solvents compared with suspended cells. The extracellular polymer substances secreted by cells in the biofilm can make a more suitable and stable environment for the cells to grow [33]. The steps of biofilm formation is as follows: (1) the free cells are contacted with the carriers owing to the movement of cells; (2) the cells contacted with the carriers are attached on the surface of the carriers owing to the affinity of carriers to the cells; and (3) the attached cells are growing on the surface of the carries with biofilm formed [34]. The common materials for immobilization can be divided into two categories, inorganic materials (e.g., slate, coal, clay brick, tygon rings, ceramic balls, carbon nanotube, carbon fiber, activated carbon chlorinated polyethylene, loofah sponge, and porous nylon) and organic materials (e.g., corncob residues, wood pulp, coconut fibers, corn stalk, cashew apple bagasse, and bagasse) [9,30,33,35
38]. It has been reported that the yield of hydrogen was five times higher than that of free cell fermentation with one-eighth of reactor volume required, during fermentation with ceramic balls as cell carriers [32]. The cell mass immobilized on the support is reported to be almost five times higher than that of the suspended fermentation [39]. In the fermentation process, cells are usually immobilized in a fixed bed reactor. Single stage and multistage is also applicable in the immobilization fermentation. Two or more fixed bed reactors are connected in series in some fermentation process [40]. Besides immobilization fermentation can be also integrated with batch, fed-batch, or continuous fermentation or any other kinds of fermentation process. During alcohols fermentation in batch or CSTR, the accumulation of alcohols in the broth inhibits the cell growth and fermentation. To solve the problem, several in situ separation technologies have been investigated to remove alcohols and controlling their concentrations in fermentation broth, such as adsorption, extraction, gas stripping, membrane distillation (MD), and pervaporation [41
47]. Another aim in research on in situ recovery is to develop a separation method, which consumes less energy than the conventional distillation. In adsorption, alcohols are preferentially transferred from the feed liquid to a solid adsorbent material. The common used adsorbents are activated carbon, diatomite, zeolite (SiO2/Al2O3), polymeric (typically ion-exchange) resins, and so on. In general adsorbents should have a high alcohols adsorption capacity, affinity, and selectivity, and be inexpensive and easy to be regenerated for reuse. The adsorbent used in the separation of alcohols from fermentation broth must have strong hydrophobicity based on the low alcohols concentration in the broth. Activated carbon is common used to remove the organic pollutes and it is also a good adsorbent candidate for the removal of butanol alcohols. While the regeneration, stability and uniformity of activated carbon are more difficult than that of the

58

PART | I Technologies

silicon-based adsorbent. Hydrophobic silicalite has a high selectivity to butanol and the separation factor is in 130
630. Separation of ABE by adsorption is with high efficiency, low energy requirement, and high productivity but the design of the adsorption process is complicated because of the interaction between the solvents and adsorbents and the complex adsorption equilibrium system. Besides the addition of adsorbents may do harm to the cell activity and thus result in the bad fermentation performance. The equipment used for desorption of the solvents from adsorbents will raise the initial cost of the fermentation process. Extraction is achieved by the difference of solubility or partition coefficient of substance in two immiscible (or slightly soluble) solvents and substance can be transferred from one solvent to another. After repeated extraction, most of the compounds were extracted. An advantage of extraction over other recovery methods may be the high capacity of the solvent and the high selectivity of the alcohol/water separation. Extraction, however, is a comprehensive operation, and the design of an extraction apparatus can be complex. There are several extraction method, namely, liquid
liquid extraction, membrane extraction, supercritical CO2 extraction, and so on [42,48]. For the liquid
liquid extraction and membrane extraction, oleyl alcohol, dibutyl phthalate, benzyl benzoate, biodiesel, and ionic liquid are usually used as the extractant. The separation of ABE from broth with extraction is achieved by the higher solubility of ABE in organic extractant than that of water. ABE can be separated from the fermentation broth and concentrated in the organic extractant but the substrate and the nutrients in the broth still existed in the broth which can improve the carbon source consumption rate and the product generation rate. Studies showed that the application of in situ extraction in the fermentation process indeed mitigate the inhibition of butanol to cells and enhance the fermentation performance. While the toxicity of the extractant to the cells hinders the couple of extraction and fermentation process and so how to decrease the toxicity of the extractant may be an important matter to be studied in the future. The separation by gas striping has many advantages such as the simple operating condition, cheap equipment and the full study of its application in fermentation. It is recognized as one of the most economical separation technologies. Gas striping can participate in various fermentation processes such as batch fermentation, fed-batch, single-stage continuous fermentation, multistage continuous fermentation, and immobilized fermentation (Fig. 2.5). During the gas striping process, the purge gas flow through the top space of the fermenter quickly and carry out the volatile solvent products and then the volatile solvent is harvested by condensation. Nitrogen is usually used as the purge gas in this process while the off-gas of the fermentation-hydrogen and carbon dioxide can also be applied to carry out the volatile solvent. Gas striping has an advantage over the adsorption and extraction because of the unnecessary additions, no carry-out of nutrients, and metabolic intermediate

Bioenergy for better sustainability Chapter | 2

59

FIGURE 2.5 The schematic diagram of fermentation coupled with gas stripping.

and efficient removal of the butanol inhibition to cells. Parameters including feedstock type, purge gas flow rate, operating temperature, and condensation conditions have obvious effect on the fermentation and separation performance during the gas striping-fermentation coupled system. Butanol concentration of 14.13 g/L, butanol productivity of 0.29 g/L/h, total ABE concentration of 18.90 g/L, and sugar utilization of 90.5% were achieved during a gas striping-fermentation coupled system with the molasses used as the carbon source [49]. MD is a process in which a microporous, hydrophobic membrane is applied to separate solvents from aqueous solutions at different temperatures. The separation is achieved by the difference vapor pressure of the solvent caused by the temperature difference between both sides of the membrane. Vapor molecules can transport through the membrane from higher vapor pressure to lower vapor pressure side of the membrane. There are several MD modes, namely, direct contact MD, vacuum MD, air gap MD, and sweeping gas distillation [50]. The commonly used membrane in the process is hydrophobic porous membrane like polyvinylidene fluoride (PVDF), polytetrafluoroethylene (PTFE), polyethylene, and polypropylene [46,50,51]. The membranes applied in MD should be porous (porosity should be in 60%
80%), inexpensive and with good mechanical properties, nontoxicity to cell growth, and membrane wetting should be avoided. During MD for alcohols removal in situ from the broth, the pores that existed in membrane are frequently blocked by the cells and their secretion decreased the total flux and separation factor.

60

PART | I Technologies

Pervaporation is a membrane separation process with a dense hydrophobic membrane applied (Fig. 2.6). Separation of alcohols from fermentation broth by pervaporation is achieved by the different dissolution
diffusion rate of solvents in the membrane. Usually organic solvents will be first adsorbed by the membrane and then desorbed in the downstream in the form of steam, which will be harvested by condensation. The microorganisms, carbon source, and nutrients cannot permeate through the membrane during the separation process and return to the fermenter for the further fermentation. Compared with the former separation technologies, pervaporation can efficiently remove the inhibition of alcohols to cells and improve the carbon source conversion efficiency and productivity. Besides it does not have any bad effect on the fermentation environment and the cell growth. As reported, the most important cell in pervaporation is the membrane (Fig. 2.7). Properties of membranes largely determines the separation performance and thus influences the fermentation process. Usually polydimethylsiloxane (PDMS) and poly(1-trimethylsilyl-1-propyne) (PTMSP) are applied to the separate solvents from broth [52]. However, the flux of the PTMSP membrane reduced as a function of operating time because of chain relaxation that decreased the free volume and also pores of the PTMSP are often blocked by cells [52,53]. Currently PDMS is the most widely used organic separation membrane [54
56]. Although PDMS is with high selectivity and stability, its mechanical properties are pretty poor. To solve the problem many support layers are developed such as PTFE, polyamide (PA), PVDF,

FIGURE 2.6 The principle of pervaporation membrane separation.

Bioenergy for better sustainability Chapter | 2

61

FIGURE 2.7 The schematic diagram of pervaporation membrane separation coupled with fermentation.

polyethersulfone, polyacrylonitrile, polyethyleneimine, and ceramic [45,47,57
60]. Different thickness of PDMS selective layer is coated on this matrix by crosslinking with tetraethyl orthosilicate. In our previous study ABE fermentation in a continuous and closed-circulating fermentation system with PDMS/PA composite membrane bioreactor was carried out. During the process, there was no any other mass exchange except the addition of water and glucose and the removal of ABE. Fermentation lasted for 300 hours and the accumulated butanol concentration was up to 61 g/L [45]. The integrated system of fermentation and pervaporation removed the butanol inhibition to cells efficiently and did not leave much fermentation residue. To enhance the strengths of these organic polymeric membranes, many inorganic fillers, namely, silica and graphene oxide, covalent organic frameworks, multiwalled carbon nanotubes, metal organic frameworks, zeolites, porous organic cages, and many others, have been researched and added into polymeric for the enhancement of pervaporation performance [61].

2.3

Challenges

Currently the raw materials for the production of bioalcohols are mainly corn, wheat, and sugar cane, which are also human food and livestock feed. This inevitably involves the safety of food and feed. There is less arable land per capita in the world, and the sensitivity of food security is strong. At present, researchers are actively exploring new energy plant resources and proposing to use marginal land to solve the problem of planting energy plants. However, the situation of marginal land and the results of energy efficiency evaluation of new energy plants are still unclear. The idea of developing energy agriculture is obvious, but how to start and how to develop it is extremely difficult. Besides batch and CSTR are the main fermentation processes for bioalcohols production. These processes are ineffective and in situ removal by membrane integrated with fermentation would be the promising

62

PART | I Technologies

ways for bioalcohols production. The poor stability of membrane separation performance and higher cost of the membrane are the two obstacles for the integrated process. Biogas production could increase from the extended use of various organic waste streams, such as food waste, crop residues, sewage sludge from waste water treatment, or microalgae and macroalgae. However, in developing countries such as China, agricultural residues are still the main sources of biogas production feedstocks. Straw used as the sole feedstock to produce biogas still faces significant challenges and requires a breakthrough in technology. Improved pretreatment technologies (such as hydrolysis) aiming at increasing feedstock biodegradability can be expected, opening the possibility to use those materials with high cellulose content. Although anaerobic digestion is well established and to an extent demonstrated technology, some improvements and cost reduction could be expected from improved biological processes (optimization, improving design, and process integration), dry fermentation, and thermophilic processes that increase biological efficiency and biogas yield. Codigestion of municipal, industrial, and agricultural waste will drive the design of future biogas plants. Dry fermentation will attract widespread attention for its potential in large-scale production of biogas combined with only small amounts of digested effluent. New techniques to improve the biological digestion process through ultrasonic treatment or enzymatic reactions, the use of new enzymes and substrates, the use bacterial strains with a greater tolerance to process changes, and feedstock type can also contribute to the advancement in biogas production. Process improvements could result in a reduced need to clean the gas and removing contaminant [62].

2.4

Future prospects

Fermentation coupled with in situ separation can promote fermentation and it would be a promising technology for bioalcohols production. In the coupled process, more kinds of substrate material should be used and the appropriate pretreatment and hydrolysis methods should be applied, because several by-products during pretreatment and hydrolysis may be accumulated in the fermentation broth. Membrane with better separation performance and stability should be developed. Furthermore, the coupled process scaled up should be integrated with more functional unit with higher efficiency and less energy consumption. Biogas will play an important role in renewable energy supply in the future and economics are the key determining factor affecting the development of biogas production. In Europe the use of heat has emerged as an opportunity to increase the income and thus the profitability of biogas plants. The use of upgraded biogas in transport applications has increased as result of the new opportunities for the use of biogas and benefited from various

Bioenergy for better sustainability Chapter | 2

63

support schemes and programs. Technological improvements in biogas upgrading technologies to biomethane could lead to lower energy intensity and improved cost performance that could make biomethane cost competitive with fossil fuel use in transport. In China upgrades to biogas plants for the production of biological methane for municipal gas and vehicle fuel will provide new pathways.

References [1] Campos-Fernandez J, Arnal JM, Gomez J, Dorado MP. A comparison of performance of higher alcohols/diesel fuel blends in a diesel engine. Appl Energy 2012;95:267
75. [2] Chen Q, Liu T. Biogas system in rural China: upgrading from decentralized to centralized? Renew Sustain Energy Rev 2017;78:933
44. [3] Scarlat N, Dallemand J-F, Fahl F. Biogas: developments and perspectives in Europe. Renew. Energy 2018;129:457
72. [4] Xue S, Song J, Wang X, Shang Z, Sheng C, Li C, et al. A systematic comparison of biogas development and related policies between China and Europe and corresponding insights. Renew Sustain Energy Rev 2020;117:109474. [5] Keis S, Shaheen R, Jones DT. Emended descriptions of Clostridium acetobutylicum and Clostridium beijerinckii, and descriptions of Clostridium saccharoperbutylacetonicum sp nov and Clostridium saccharobutylicum sp nov. Int J Syst Evol Microbiol 2001;51: 2095
103. [6] Zhang X, Feng XH, Zhang H, Wei YC. Utilization of steam-exploded corn straw to produce biofuel butanol via fermentation with a newly selected strain of Clostridium acetobutylicum. Bioresources 2018;13:5805
17. [7] Gong FY, Bao GH, Zhao CH, Zhang YP, Li Y, Dong HJ. Fermentation and genomic analysis of acetone-uncoupled butanol production by Clostridium tetanomorphum. Appl Microbiol Biotechnol 2016;100:1523
9. [8] Bramono SE, Lam YS, Ong SL, He JZ. A mesophilic Clostridium species that produces butanol from monosaccharides and hydrogen from polysaccharides. Bioresour Technol 2011;102:9558
63. [9] Lipovsky J, Patakova P, Paulova L, Pokorny T, Rychtera M, Melzoch K. Butanol production by Clostridium pasteurianum NRRL B-598 in continuous culture compared to batch and fed-batch systems. Fuel Process Technol 2016;144:139
44. [10] Jones DT, Woods DR. Acetone-butanol fermentation revisited. Microbiol Rev 1986;50:484
524. [11] Erguder TH, Tezel U, Guven E, Demirer GN. Anaerobic biotransformation and methane generation potential of cheese whey in batch and UASB reactors. Waste Manage 2001;21:643
50. [12] Jang YS, Malaviya A, Cho C, Lee J, Lee SY. Butanol production from renewable biomass by clostridia. Bioresour Technol 2012;123:653
63. [13] Lee SY, Park JH, Jang SH, Nielsen LK, Kim J, Jung KS. Fermentative butanol production by clostridia. Biotechnol Bioeng 2008;101:209
28. [14] Jin C, Yao MF, Liu HF, Lee CFF, Ji J. Progress in the production and application of nbutanol as a biofuel. Renew Sustain Energy Rev 2011;15:4080
106. [15] Shuba ES, Kifle D. Microalgae to biofuels: ‘promising’ alternative and renewable energy, review. Renew Sustain Energy Rev 2018;81:743
55.

64

PART | I Technologies

[16] Wang HC. The development process and the future prospects of Chinese sewage treatment industry. J Environ Protect 2012;15:19
22 [in Chinese]. [17] Wu J, Jiang J, Zhou HM, Bi L. Current operation status of sludge anaerobic digestion system in municipal wastewater treatment plants in China. China Water Wastewater 2008;24 (22):21
4 [in Chinese]. [18] Yang KJ, Zhang DT. The development of large and medium scale biogas plants in China in 10 years for China biogas. Beijing, China: China Science and Technology Press; 1990 [in Chinese]. [19] Wang K.J. The application prospect of anaerobic biological technology in the field of agriculture and industry in China. Agricultural mechanization and the new rural construction. In: Proceedings of the 2006 academic annual conference of the Chinese society of agricultural machinery, Zhengjiang, Jiang Su Province, November 13
16; 2006. p. 92
97 [in Chinese]. [20] Zhou MJ, Peng WH, Xiong CY. The development of the process of biogas fermentation. Department of agricultural environment protection and rural energy of MOA China biogas society ten years for China biogas. Beijing, China: China Science and Technology Press; 1990. p. 22
8 [in Chinese]. [21] Xu JQ, Yang KJ, Liu YH, Huang ZL, Xu KN. The industrial scale experiment for the integrated treatment system with biogas fermentation of wastewater from intensive pig farms. China Biogas 1991;9(3):26
9 [in Chinese]. [22] Ministry of agriculture of the People’s Republic of China, (MOA). Classification of scale for biogas engineering. NY/T 667-2011 [in Chinese]. [23] Xue C, Zhao XQ, Liu CG, Chen LJ, Bai FW. Prospective and development of butanol as an advanced biofuel. Biotechnol Adv 2013;31:1575
84. [24] Zheng J, Tashiro Y, Wang QH, Sonomoto K. Recent advances to improve fermentative butanol production: genetic engineering and fermentation technology. J Biosci Bioeng 2015;119:1
9. [25] Mutschlechner O, Swoboda H, Gapes JR. Continuous two-stage ABE-fermentation using Clostridium beijerinckii NRRL B592 operating with a growth rate in the first stage vessel close to its maximal value. J Mol Microbiol Biotechnol 2000;2:101
5. [26] Bankar SB, Survase SA, Singhal RS, Granstrom T. Continuous two stage acetonebutanol-ethanol fermentation with integrated solvent removal using Clostridium acetobutylicum B 5313. Bioresour Technol 2012;106:110
16. [27] Ezeji TC, Qureshi N, Blaschek HP. Butanol fermentation research: upstream and downstream manipulations. Chem Rec 2004;4:305
14. [28] Kumar G, Mudhoo A, Sivagurunathan P, Nagarajan D, Ghimire A, Lay CH, et al. Recent insights into the cell immobilization technology applied for dark fermentative hydrogen production. Bioresour Technol 2016;219:725
37. [29] Chen WH, Chen YC, Chaiprapat S. Activation of immobilized Clostridium saccharoperbutylacetonicum N1-4 for butanol production under different oscillatory frequencies and chemical buffers. Int Biodeterior Biodegrad 2016;110:129
35. [30] Tripathi A, Sami H, Jain SR, Viloria-Cols M, Zhuravleva N, Nilsson G, et al. Improved bio-catalytic conversion by novel immobilization process using cryogel beads to increase solvent production. Enzyme Microbial Technol 2010;47:44
51. [31] Kheyrandish M, Asadollahi MA, Jeihanipour A, Doostmohammadi M, Rismani-Yazdi H, Karimi K. Direct production of acetone-butanol-ethanol from waste starch by free and immobilized Clostridium acetobutylicum. Fuel 2015;142:129
33.

Bioenergy for better sustainability Chapter | 2

65

[32] Lee SM, Cho MO, Park CH, Chung YC, Kim JH, Sang BI, et al. Continuous butanol production using suspended and immobilized Clostridium beijerinckii NCIMB 8052 with supplementary butyrate. Energy Fuel 2008;22:3459
64. [33] Plangklang P, Reungsang A, Pattra S. Enhanced bio-hydrogen production from sugarcane juice by immobilized Clostridium butyricum on sugarcane bagasse. Int J Hydrog Energy 2012;37:15525
32. [34] Parsek MR, Greenberg EP. Sociomicrobiology: the connections between quorum sensing and biofilms. Trends Microbiol 2005;13:27
33. [35] Keskin T, Giusti L, Azbar N. Continuous biohydrogen production in immobilized biofilm system versus suspended cell culture. Int J Hydrog Energy 2012;37:1418
24. [36] Survase SA, van Heiningen A, Granstrom T. Wood pulp as an immobilization matrix for the continuous production of isopropanol and butanol. J Indust Microbiol Biotechnol 2013;40:209
15. [37] Zhou W, Liu J, Fan S, Xiao Z, Qiu B, Wang Y, et al. Biofilm immobilization of Clostridium acetobutylicum on particulate carriers for acetone-butanol-ethanol (ABE) production. Bioresour Technol Rep 2018;3:211
17. [38] Raganati F, Olivieri G, Russo ME, Marzocchella A. Butanol production by Clostridium acetobutylicum in a continuous packed bed reactor fed with cheese whey. In: Pierucci S, Klemes JJ, editors. Icheap-11: 11th International conference on chemical and process engineering, Pts 1
42013. p. 937
42. [39] Liu JY, Zhou WC, Fan SQ, Qiu BY, Wang YY, Xiao ZY, et al. Cell degeneration and performance decline of immobilized Clostridium acetobutylicum on bagasse during hydrogen and butanol production by repeated cycle fermentation. Int J Hydrog Energy 2019;44:26204
12. [40] Raganati F, Procentese A, Olivieri G, Russo ME, Gotz P, Salatino P, et al. Butanol production by Clostridium acetobutylicum in a series of packed bed biofilm reactors. Chem Eng Sci 2016;152:678
88. [41] Qureshi N, Maddox IS. Reduction in butanol inhibition by perstraction: utilization of concentrated lactose/whey permeate by Clostridium acetobutylicum to enhance butanol fermentation economics. Food Bioprod Process 2005;83:43
52. [42] Groot WJ, Soedjak HS, Donck PB, Vanderlans RGJM, Luyben KCAM, Timmer JMK. Butanol recovery from fermentations by liquid-liquid-extraction and membrane solventextraction. Bioprocess Eng 1990;5:203
16. [43] Ezeji TC, Qureshi N, Blaschek HP. Production of acetone, butanol and ethanol by Clostridium beijerinckii BA101 and in situ recovery by gas stripping. World J Microbiol Biotechnol 2003;19:595
603. [44] Qureshi N, Meagher MM, Huang J, Hutkins RW. Acetone butanol ethanol (ABE) recovery by pervaporation using silicalite-silicone composite membrane from fed-batch reactor of Clostridium acetobutylicum. J Membr Sci 2001;187:93
102. [45] Chen CY, Xiao ZY, Tang XY, Cui HD, Zhang JQ, Li WJ, et al. Acetone-butanol-ethanol fermentation in a continuous and closed-circulating fermentation system with PDMS membrane bioreactor. Bioresour Technol 2013;128:246
51. [46] Lin DS, Yen HW, Kao WC, Cheng CL, Chen WM, Huang CC, et al. Bio-butanol production from glycerol with Clostridium pasteurianum CH4: the effects of butyrate addition and in situ butanol removal via membrane distillation. Biotechnol Biofuels 2015;8. [47] Liu G, Wei W, Wu H, Dong X, Jiang M, Jin W. Pervaporation performance of PDMS/ ceramic composite membrane in acetone butanol ethanol (ABE) fermentation-PV coupled process. J Membr Sci 2011;373:121
9.

66

PART | I Technologies

[48] Qureshi N, Eller F. Recovery of butanol from Clostridium beijerinckii P260 fermentation broth by supercritical CO2 extraction. J Chem Technol Biotechnol 2018;93:1206
12. [49] Wechgama K, Laopaiboon L, Laopaiboon P. Enhancement of batch butanol production from sugarcane molasses using nitrogen supplementation integrated with gas stripping for product recovery. Indust Crop Prod 2017;95:216
26. [50] Kujawska A, Kujawski J, Bryjak M, Kujawski W. ABE fermentation products recovery methods-a review. Renew Sustain Energy Rev 2015;48:648
61. [51] Banat FA, Simandl J. Membrane distillation for dilute ethanol—separation from aqueous streams. J Membr Sci 1999;163:333
48. [52] Fadeev AG, Selinskaya YA, Kelley SS, Meagher MM, Litvinova EG, Khotimsky VS, et al. Extraction of butanol from aqueous solutions by pervaporation through poly(1-trimethylsilyl-1-propyne). J Membr Sci 2001;186:205
17. [53] Lopez-Dehesa C, Gonzalez-Marcos JA, Gonzalez-Velasco JR. Pervaporation of 50 wt % ethanol-water mixtures with poly(1-trimethylsilyl-1-propyne) membranes at high temperatures. J Appl Polym Sci 2007;103:2843
8. [54] Yang DC, Cheng C, Bao MT, Chen LJ, Bao YM, Xue C. The pervaporative membrane with vertically aligned carbon nanotube nanochannel for enhancing butanol recovery. J Membr Sci 2019;577:51
9. [55] Liu GP, Hung WS, Shen J, Li QQ, Huang YH, Jin W, et al. Mixed matrix membranes with molecular-interaction-driven tunable free volumes for efficient bio-fuel recovery. J Mater Chem A 2015;3:4510
21. [56] Liu GP, Gan L, Liu SN, Zhou HL, Wei W, Jin WQ. PDMS/ceramic composite membrane for pervaporation separation of acetone-butanol-ethanol (ABE) aqueous solutions and its application in intensification of ABE fermentation process. Chem Eng Process 2014;86:162
72. [57] Zhang WD, Sun W, Yang J, Ren ZQ. The study on pervaporation behaviors of dilute organic solution through PDMS/PTFE composite membrane. Appl Biochem Biotechnol 2010;160:156
67. [58] Sun D, Yang QC, Sun HL, Liu JM, Xing ZL, Li BB. Effects of PES support layer structure on pervaporation performances of PDMS/PES hollow fiber composite membranes. Desalin Water Treat 2016;57:9123
35. [59] Li J, Chen XR, Qi BK, Luo JQ, Zhuang XJ, Su Y, et al. Continuous acetone-butanolethanol (ABE) fermentation with in situ solvent recovery by silicalite-1 filled PDMS/PAN composite membrane. Energy Fuels 2014;28:555
62. [60] Trinh LTP, Lee YJ, Bae HJ, Lee HJ. Pervaporative separation of butanol using a composite PDMS/PEI hollow fiber membrane. J Indust Eng Chem 2014;20:2814
18. [61] Ong YK, Shi GM, Le NL, Tang YP, Zuo J, Nunes SP, et al. Recent membrane development for pervaporation processes. Prog Polym Sci 2016;57:1
31. [62] Scarlat N, Dallemand J-F. Technology development report heat and power from biomass. 2016. Low carbon energy observatory deliverable D. 2.1. JRC102407.

Chapter 3

Organic Rankine cycle driven by geothermal heat source: life cycle techno-economic
environmental analysis Chao Liu1, Shukun Wang1 and Jingzheng Ren2 1

Key Laboratory of Low-Grade Energy Utilization Technologies and Systems, Ministry of Education, School of Energy and Power Engineering, Chongqing University, Chongqing, P.R. China, 2Department of Industrial and Systems Engineering, The Hongkong Polytechnic University, Hongkong, SAR, P.R. China

Chapter Outline 3.1 Introduction 68 3.2 Organic Rankine cycle system description and working fluid selection 70 3.3 Methods and models 73 3.3.1 Thermodynamic and technical analysis 73 3.3.2 Heat exchanger model 76 3.3.3 Economic and exergoeconomc analysis 76 3.3.4 Life-cycle environmental analysis 81 3.3.5 Multicriteria integrated assessment and decisionmaking 84 3.4 Thermodynamic and economic results 85 3.4.1 Effects of design parameters on thermodynamic performance 85 3.4.2 Effects of design parameters on economic performance 89

3.4.3 Effects of design parameters on exergoeconomic performance 92 3.4.4 Sensitivity analysis on the economic performance and inlet temperature of geothermal source 95 3.5 Life-cycle and carbon footprint analysis of the organic Rankine cycle 99 3.5.1 Environmental evaluation of life cycle 99 3.5.2 Environmental evaluation of components 103 3.5.3 Environmental evaluation of working fluids 104 3.5.4 Analysis of emission reductions 105 3.5.5 Sensitivity analysis 106 3.6 Comparison between different layouts of organic Rankine cycle systems 108 3.7 Results of multifactor evaluation 112 3.8 Conclusions 118 References 122

Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00003-0 © 2021 Elsevier Inc. All rights reserved.

67

68

3.1

PART | I Technologies

Introduction

In recent years, energy is progressively becoming a crucial factor in economic systems and it plays a dominant role in social development. With the increasing demand for energy and strict emission reduction requirement, the energy framework dominated by fossil fuels is no longer feasible due to the shortage of conventional energy and severe environmental pollution. In this regard, the development of renewable energy, that is solar energy, wind energy, biomass energy, and geothermal energy, and harnessing of waste heat are regarded as potential solutions. Geothermal energy, as a representative of renewable energies, has shown enormous potential and extensive application prospects. The organic Rankine cycle (ORC) is an effective and reliable technology that can convert low- to medium-grade waste heat into electricity, and it has been proved to be suitable for geothermal power systems [1]. Marty et al. [2] evaluated a combined heat and power system for a geothermal plant and proposed an optimization approach to solve mixed-integer nonlinear programming problems. The characteristic dimensions for each component were selected as optimization variables. Results indicated that the proposed approach was robust, fast, and stable. Imran et al. [3] performed a comparative assessment on different layouts of ORC systems for geothermal heat source applications. The basic ORC, recuperated ORC, and regenerative ORC were chosen as candidates. Results indicated that R245fa had the lowest costs of 2423 $/kW for the basic ORC and 2567 $/kW for the regenerative ORC. Liu et al. [4] conducted a parametric analysis of an ORC using R600a/R601a as the working fluid in a geothermal field. Results showed that such ORC generated 4%
11% more power and required nearly the same turbine size as the ORC with pure R600a. Walraven et al. [5] conducted an economic optimization of an air-cooled ORC in a geothermal field. They found that the parameters, including the electricity price and annual electricity price, had strong influences on the efficiency of the ORC. Until the present, the ORC with a simple structure and easy maintenance has been drawing much attentions. Many studies have been performed to comprehensively evaluate the ORC system performance and improve its efficiency. For the past few decades, the work mostly focused on the following aspects: system structure improvement (such as subcritical ORC, supercritical/transcritical ORC, regenerative ORC, recuperative ORC, dual-pressure ORC, dual-loop ORC and combined heating and power/combined cooling, heating and power system with ORC), design and selection of major components, selection principle of organic fluids (such as pure fluids including dry, wet and isentropic fluids or zeotropic mixtures), dynamic characteristics analysis, other different evaluations, and multiobjective optimization. Recently researchers have devoted themselves to the studies of the economic and environmental feasibility of the ORC. Zhang et al. [6] performed

Organic Rankine cycle driven by geothermal heat source Chapter | 3

69

an economic comparison of the basic ORC with different heat exchangers. The production cost of unit electricity was selected as the optimization objective and four different layouts of heat exchangers were considered. Results indicated that the finned-tube heat exchanger as evaporator and shell-and-tube heat exchanger as condenser are the most cost-effective configuration for a water-cooled ORC driven by waste flue gas. Quoilin et al. [7] conducted a thermoeconomic optimization for a small-scale ORC. The plate heat exchanger was adopted as evaporator in their model. Moreover apart from the conventional economic evaluation, the concept of exergoeconomics (a combination of exergy and economics), which is a relatively popular area in engineering, was proposed. Shokati et al. [8] conducted a comparative study on different layouts of ORC systems by using an exergoeconomic method. Optimization results showed that the dual-pressure ORC produced the maximum electrical power among the basic ORC, dual-fluid ORC, and Kalina cycle. However, the Kalina cycle had the minimum unit cost of produced power. Fergani et al. [9] and Ashouri et al. [10] also selected the exergoeconomic analysis to evaluate the corresponding ORC systems. On the other hand the environmental impact of energy systems is an increasing concern. Liu et al. [11] adopted a life cycle assessment approach to evaluate the environmental impact of ORC systems. Results showed that the construction stage contributed the most to the global warming potential (GWP) and eutrophication potential. Heberle et al. [12] investigated the environmental impact of a geothermal power plant in Germany using three different ORC systems. Results revealed that a two-stage ORC is favorable under the conditions of higher inlet temperature of geothermal fluid. Economic and environmental indicators are often selected as important parameters in multiobjective analyses [13,14]. A few studies [12,15
17] were performed to evaluate the technical feasibility of different layouts of ORCs, particularly for the ORC with an internal heat exchanger (IHX-ORC) and dual-pressure ORC. Studies [18
21] indicated that the additional equipment of IHX and the dual-pressure evaporator improved the thermodynamic performance including net power output, thermal efficiency, and exergy efficiency for subcritical ORC systems. With respect to economic feasibility, an investigation indicated that the additional IHX would increase the electricity production cost (EPC), payback period and total capital cost [22]. Conversely, Li [23] and Zhang et al. [24] revealed that the IHX-ORC would have more benefits compared with the basic ORC when the heat source temperature and load were increased to a high level. As for the dual-pressure ORC, both Wang et al. [25] and Shokati et al. [8] indicated that the additional evaporator would increase the capital cost of the basic ORC under a conventional economic viewpoint and exergoeconomic perspective. The brief review discussed above indicates that the application of ORCs in the geothermal field is a active topic, and the different evaluation methods

70

PART | I Technologies

for various ORC layouts are still drawing much attentions because of the complex impact of the thermal parameters and different viewpoints. However, few works conducted all-sided evaluation of an ORC adopted in the geothermal field to recovery the low-grade waste heat and their research interests are confined to one specific aspect. Thus a intensive study is conducted for a basic ORC adopted in geothermal field, including the single objective of thermodynamic, economic, and environmental analysis and the multicriteria integrated assessment. Moreover, the comparative study on the different system layouts and working fluids is also carried out and shows a great significance. In this study, a parametric analysis of various ORC systems using environmental-friendly working fluids is performed first. The evaluation indicators are based on three aspects describing the thermodynamic performance, economic cost and environmental benefits of the proposed systems and fluids. Consequently the evaporating pressure and pinch-point temperatures in evaporator and condenser can be determined according to the corresponding models. The net power output, EPC, exergy cost for unit power production and emission reduction are used as the optimization objectives in the thermodynamic, economic, exergoeconomic, and environmental analyses. Then, a comparison of different ORC systems is conducted. Finally, a multifactor evaluation is carried out to evaluate the comprehensive feasibility of a basic ORC with different working fluids.

3.2 Organic Rankine cycle system description and working fluid selection The schematic of a basic ORC system is shown in Fig. 3.1. The system consists of a preheater, an evaporator, a turbine with generator, a condenser and a pump. The working principle of an ORC could be illustrated as follows: the working fluid at saturated liquid state is pressurized to its evaporating pressure by the pump at the outlet of the condenser. Then, the working fluid absorbs heat from geothermal water having passed both the preheater and evaporator. After being heated to saturated or superheated vapor, the working fluid expands in the turbine, which drives the generator to produce electricity. Finally it is cooled again in the condenser by air. The corresponding T
s diagram and the limiting condition of the turbine inlet temperature are illustrated in Fig. 3.2. For both isentropic and dry fluids, the turbine inlet temperature is determined by the state of saturated vapour (Fig. 3.2A). However, for a wet fluid, to prevent cavitation in the turbine, the specific entropy of the state at the turbine inlet depends on the condensation temperature (Fig. 3.2B). The given conditions and input parameters of the ORC are summarized in Table 3.1. To prevent silica precipitation of the geothermal fluid in the reinjection wells, the temperature constraint of the discharged geothermal

Organic Rankine cycle driven by geothermal heat source Chapter | 3

71

FIGURE 3.1 Schematic diagrams for a basic ORC system.

FIGURE 3.2 T-s diagrams of a basic subcritical ORC: (A) dry or isentropic fluid and (B) wet fluid.

fluid is selected as 70 C [15,26]. The condensing temperature is set as 30 C and fresh air is used as the cooling medium. The exergy unit cost of geothermal fluid is assumed as 1.3 $/GJ [8]. Some assumptions for simplification of the calculation problem are as follows: G G G G

The system works at a steady state. There are no heat losses in the components. The pressure drops in the heat exchangers and pipes are neglected. The working fluid at the outlet of the condenser is in saturated liquid state.

72

PART | I Technologies

TABLE 3.1 Given conditions and input parameters.

G

Parameter

Unit

Value

Inlet temperature of geothermal water, Ths,in



C

110

Pressure of geothermal water, Phs

kPa

1200

Constraint outlet temperature of geothermal water, Ths,outlimit



C

70

Mass flow rate of geothermal water, mhs

kg/s

10

Condensing temperature, Tcon



C

30

Range of pinch point temperature in evaporator, ΔTe



C

5
30

Range of pinch point temperature in condenser, ΔTc



C

5
10

Inlet temperature of cooling air, Tcs,in



C

20

Isentropic efficiency of turbine, ηtur




Isentropic efficiency of pump, ηpump




0.75

Operation time of each year, top

hour

7000

Life-cycle time, LT

year

15

Annual loan interest rate, i




0.05

On-grid electricity price, Celec

$/kWh

0.15

Unit cost of exergy of the geothermal source, chs

$/GJ

1.3

Environmental temperature, T0



20

Environmental pressure, p0

kPa

C

0.85

101.325

The dry and isentropic working fluid at the inlet of the turbine is saturated vapour while the wet working fluid is superheated vapour.

The selection of working fluid is a crucial factor in system design because the system performance primarily depends on the working fluid properties. Currently the environmental requirements for organic fluids are increasingly stringent, and many environmental criteria adopted from the European Union gradually come into practice, for example ‘MAC Directive’ and ‘F-gases Regulation’. Thus the selection of environmental-friendly working fluids is of great significance. In this study fluids with zero ozone depletion potential and low GWP, namely R1270, R1234yf, R290, R1234ze(E) and 600a, are chosen as candidates. In addition, fluid R134a with higher GWP is also selected because it is wildly used in the ORC field. The properties of selected working fluid are shown in Table 3.2.

73

Organic Rankine cycle driven by geothermal heat source Chapter | 3

TABLE 3.2 Properties of working fluids. ASHRAE 34 R1270 R1234yf

Molecular mass (g/ mol)

Critical temperature, Tcrit ( C)

Critical pressure, pcrit (MPa)

ODP

GWP 100 years

42.08

91.1

4.56

0

2

114.04

94.7

3.38

0

4

R290

44.10

96.7

4.25

0

3

R134a

102.03

101.1

4.06

0

1370

R1234ze (E)

114.04

109.4

3.64

0

6

58.12

134.7

3.63

0

B20

R600a

ASHRAE 34, American society of refrigerating engineers: Standard 34 for designation and safety classification of refrigerants; GWP, global warming potential; ODP, ozone depression potential.

3.3

Methods and models

In this section, the methods and models of a basic ORC system are established. The flowchart of the system evaluation framework is shown in Fig. 3.3. First the themodynamic model of the basic ORC is described. Then the heat exchanger model is introduced for the basic ORC to estimate the heat exchanger areas of the evaporator and condenser. Based on the earlier described two foundation models, the economic and exergoeconomic models are carried out to evaluate the capital cost and cost per exergy unit of the net power produced. After that, the environmental model of the basic ORC is conducted based on the whole life-cycle of the ORC system. And finally, under comprehensively consideration of the energetic, exergetic, economic and environmental factors, a multicriteria integrated assessment of the basic ORC is conducted.

3.3.1

Thermodynamic and technical analysis

The energy and mass conservation equations and exergy flow rate under steady-state condition for each component are listed in Table 3.3. The general equations are written as follows [8]: X X X X Q_ 1 m_ i hi 5 W_ 1 m_ o ho ð3:1Þ X X m_ i 5 m_ o ð3:2Þ _ i 5 m_ i ½ðhi 2 h0 Þ 2 T0 ðsi 2 s0 Þ Ex

ð3:3Þ

74

PART | I Technologies

FIGURE 3.3 Flowchart of the ORC system evaluation framework.

The exergy rate balance for each component can be defined as E_ F 5 E_ P 1 E_ L 1 E_ D

ð3:4Þ

where E_ F and E_ P are the fuel and product exergy rates, respectively, E_ L is the exergy rate associated with component losses and E_ D is the exergy destruction rate. The net power output, thermal efficiency, total exergy loss, and exergy efficiency can be determined as follows: G

Net power output W_ net 5 W_ tur 2 W_ pump

ð3:5Þ

TABLE 3.3 Models of energy balance and exergy for each component in a basic ORC. Component

Energy balance model

Preheater

Qpre 5 mwf ðh5 2 h4 Þ 5 mhs hhs;pp 2 hhs;out
 Qevap 5 mwf ðh1 2 h5 Þ 5 mhs hhs;in 2 hhs;pp

Evaporator Turbine Condenser Pump




Wtur 5 mwf ðh1 2 h2 Þ ηtur 5 ðh1 2 h2 Þ=ðh1 2 h2s Þ




 Qcond 5 mwf ðh2 2 h3 Þ 5 mcs hcs;out 2 hcs;in Wpump 5 mwf ðh4 2 h3 Þ ηpump 5 ðh4s 2 h3 Þ=ðh4 2 h3 Þ

Exergy of fuel, EF,k

Exergy of product, EP,k

Exergy loss, I

E_ hs;pp 2 E_ hs;out

E_ 5 2 E_ 4

_I pre 5 E_ F;pre 2 E_ P;pre

E_ hs;in 2 E_ hs;pp

E_ 1 2 E_ 5

_I evap 5 E_ F;evap 2 E_ P;evap

E_ 1 2 E_ 2

_ tur W

_I tur 5 E_ F;tur 2 E_ P;tur

E_ 2 2 E_ 3

E_ cs;out 2 E_ cs;in

_I cond 5 E_ F;cond 2 E_ P;cond

_ pump W

E_ 4 2 E_ 3

_I pump 5 E_ F;pump 2 E_ P;pump

76 G

PART | I Technologies

Thermal efficiency ηth 5

G

W_ net
 m_ hs hhs;in 2 hhs;out

Total exergy loss I_tot 5 I_pre 1 I_evap 1 I_tur 1 I_cond 1 I_pump

G

ð3:6Þ

ð3:7Þ

Exergy efficiency ηex 5

3.3.2

W_ net E_ hs;in 2 E_ hs;out

ð3:8Þ

Heat exchanger model

The heat exchanger is an essential component of an ORC and accounts for a large portion of the capital cost. When the cooling type is selected as air-cooled, the cost of heat exchanger would even reach approximately 80% [5]. Thus the heat exchanger type selection is very important. Among all types of heat exchangers, the plate heat exchanger is suitable for high pressure and temperature fluids. Additionally it is the preferred type for geothermal power systems [27]. In this study the plate heat exchanger is chosen for the preheater and evaporator, while the finned-tube heat exchanger is chosen for the condenser. The geometric parameters of the selected heat exchangers are listed in Table 3.4. The log mean temperature difference approach is adopted to calculate the average temperature difference between the heat-transfer fluids. Table 3.5 gives the heat-transfer coefficient correlations corresponding to the two heat exchangers. In the condenser, the overall heat-transfer coefficient is determined by Young’s correlation because the heat-transfer coefficient of cooling air is much smaller than that of the working fluid [14].

3.3.3

Economic and exergoeconomc analysis

The capital cost model for each component in the ORC system is listed in Table 3.6. K1, K2, K3, C1, C2, C3, B1 B2, Fm and Fbm are the cost coefficients corresponding to each component and the values are given in Table 3.7. The total capital cost of the ORC system can be calculated by X Ztot 5 Zi ð3:9Þ The cost of maintenance and operation (γmo) is taken into consideration and is assumed as 1.5% per year. The capital recovery factor (CRF) can be calculated by

Organic Rankine cycle driven by geothermal heat source Chapter | 3

77

TABLE 3.4 Parameters of the heat exchangers. Heat exchanger type

Parameter

Unit

Value

Plate heat exchanger

Effective channel length, Leff

m

1.25

Width of flow channel, w

m

0.55

Plate thickness, δ

mm

0.80

Mean flow channel gap, b

mm

5.00

Chevron angle, β

degrees

π/6

Material




CS

Inner tube diameter, di

mm

10.00

Outer tube diameter, do

mm

13.00

Fin collar outside diameter, dt

mm

25.30

Fin root diameter, db

mm

10.55

Fin pinch, Y

mm

2.60

Fin height, H

mm

5.00

Fin thickness, δf

mm

0.15

Transversal tube pitch, S1

mm

25.00

Longitudinal tube pitch, S2

mm

21.00

Tube length, L

m

3.00

Tube material




CS

Fin material




Al

Finned-tube heat exchanger

CRF 5  The EPC can be determined by


ið11iÞLT  ð11iÞLT 2 1

Ztot CRF 1 γ mo EPC 5 top Wnet

ð3:10Þ

 ð3:11Þ

The exergoeconomic analysis aims to specify the cost formation process and determine the exergy unit cost of product streams of the proposed system. The specific exergy costing method is adopted in this study. Under steady-state conditions, the material streams and heat as well as work interactions can reflect the entering and existing exergy streams for each component in the ORC. The relative costs with these exergy streams are represented by the following equations:

TABLE 3.5 Heat-transfer coefficient correlations for heat exchangers. Heat exchanger type Plate heat exchanger

Finned-tube heat exchanger

Region

Heat-transfer coefficient correlation

References

Chisholm-Wanniarachchi correlation

Single-phase


0:646 0:583 1=3 Nu 5 Dh α=λ l 5 0:724 6β=π Re Pr

[28]

Yan-Lin correlation

Two-phase boiling

Young correlation

Single-phase: air-side

h
1=2 i 1=3 0:3 0:5 Nu 5 Dh α=λl 5 1:926Prl Boeq Reeq 1 2 xm 1 xm ρl =ρv 
 Boeq 5 q= Geq rfg Reeq 5 Geq Dh =η l
0:667
1=3
0:164
0:075 Nu 5 0:1507 dt Gmax =μ cp =μ Y =H Y =δf dt =db 5 1:2B1:6; and db 5 13:5B16mm; Nu 5 0:1378ðdt 3 Gmax =μÞ0:718 ðcp =μÞ1=3 ðY =HÞ0:296 dt =db 5 1:7B2:4; and db 5 12B41mm

[28]

[6]

Organic Rankine cycle driven by geothermal heat source Chapter | 3

79

TABLE 3.6 Capital cost model for each component in an ORC system. Component Preheater

Evaporator

Turbine Condenser

Pump

Investment model
 
2 logCpre 5 K1; pre 1 K2; pre log10 Apre 1 K3;pre log10 Apre
 
2 logFpre 5 C
1;pre 1 C2;pre log10 ppre 1 C3;pre log10 ppre Zpre 5 Cpre B1;pre 1 B2;pre Fm;pre Fpre CEPCI2017 =CEPCI2001
 
2 logCevap 5 K1;evap 1 K2;evap log10 Aevap 1 K3;evap log10 Aevap
 
2 logFevap 5 C1;evap 1 C2;evap log10 pevap 1C3;evap log10 pevap
Zevap 5 Cevap B1;evap 1 B2;evap Fm;evap Fevap CEPCI2017 =CEPCI2001  2 logCtur 5 K1;tur 1 K2;tur log10 ðWtur Þ 1 K3;tur log10 ðWtur Þ Ztur 5 Ctur Fbm;tur Ftur CEPCI2017 =CEPCI2001  2 logCcond 5 K1;cond 1 K2;cond log10 ðAcond Þ 1 K3;cond log10 ðAcond Þ
 
2 logFcond 5 C1;cond 1 C2;cond log10 pcond 1C3;cond log10 pcond
Zcond 5 Ccond B1;cond 1 B2;cond Fm;cond Fcond CEPCI2017 =CEPCI2001
 
2 logCpump 5 K1;pump 1 K2;pump log10 Wpump 1 K3;pump log10 Wpump
 
2 logFpump 5 C1;pump 1 C2;pump log10 ppump 1 C3;pump log10 ppump
Zpump 5 Cpump B1;pump 1 B2;pump Fm;pump Fpump CEPCI2017 =CEPCI2001

X

C_ in;k 1 C_ q;k 1 Z_k 5

X

C_ out;k 1 C_ w;k

ð3:12Þ

where C_ in , C_ q , C_ out and C_ w are the cost rates associated with the entering exergy streams, exergy transfer in the form of heat, existing exergy streams and exergy transfer in the form of work, respectively. In addition, C_ 5 cE_ is the exergy costing principle, and Z_ is the cost rate for capital investment and maintenance and operation, which can be defined as
 Z_k 5 CRF 1 γ mo Zk =top ð3:13Þ In the exergoeconomic analysis, the exergy cost rate balances and auxiliary equations for each component are provided in Table 3.8. The average cost unit exergy of fuel (cF,k) and product (cP,k), relative cost difference (rk), cost rate of exergy destruction (C_ D ) and exergoeconomic factor (fk) are chosen to evaluate the exergoeconomic performance of the system components. The calculation details are as follows: cF;k 5

C_ F;k _ F;k Ex

ð3:14Þ

cP;k 5

C_ P;k _ P;k Ex

ð3:15Þ

_ D;k C_ D;k 5 cF;k Ex

ð3:16Þ

TABLE 3.7 Coefficients in equations for evaluating the investment of system components [6,29]. Component

K1

K2

K3

C1

C2

C3

B1

B2

Fm

Fbm

Plate heat exchanger

4.6656

20.1557

0.1547

0

0

0

0.96

1.21

1.00

/

Finned-tube heat exchanger

4.3247

20.3030

0.1634

0

0

0

1.63

1.66

1.25

/

(5 , p , 140 bar)

20.00164

20.00627

0.0123

Pump

3.3892

0.0536

0.1538

0

0

0

1.89

1.35

1.50

/

(10 , p , 100 bar)

20.3935

0.3957

20.00226

20.177

/

/

/

/

/

/

6.2

Turbine

2.705

1.440

Organic Rankine cycle driven by geothermal heat source Chapter | 3

81

TABLE 3.8 Exergy cost rate balances and auxiliary equations for each component in a basic ORC. Component

Energy balance model

Investment model

Preheater

C_ 4 1 C_ hs;pp 1 Z_ pre 5 C_ 5 1 C_ hs;out

chs;pp 5 chs;out

Evaporator

C_ 5 1 C_ hs;in 1 Z_ evap 5 C_ 1 1 C_ hs;pp

chs;in 5 chs;pp

Turbine

C_ 1 1 Z_ tur 5 C_ 2 1 C_ Wtur

c1 5 c2

Condenser

C_ 2 1 C_ cs;in 1 Z_ cond 5 C_ 3 1 C_ cs;out

c2 5 c3 ; ccs;in 5 0,

Pump

C_ 3 1 C_ Wpump 1 Z_ pump 5 C_ 4

cWpump 5 cWtur

FIGURE 3.4 Life cycle boundary of an ORC system.

rP;k 5 fk 5

cP;k 2 cF;k cF;k

Z_k _ CD;k 1 Z_k 1 C_ L;k

ð3:17Þ ð3:18Þ

The cost per exergy unit of the net power produced can be calculated as follows: cP;sys 5

3.3.4

C_ P;sys Wnet

ð3:19Þ

Life-cycle environmental analysis

3.3.4.1 Life-cycle boundary The life cycle boundary of an ORC is depicted in Fig. 3.4. The whole lifecycle time consists of three different phases, namely construction, operation and decommissioning. During the construction phase, the main concern is

82

PART | I Technologies

about the manufacturing of components and working fluids and the transport of these components. During the operation phase, the concern is on the emission result from the leak of working fluids and there is no other material or energy consumed. At the decommissioning phase, the emissions from recycling and processing of components and fluids are estimated.

3.3.4.2 Carbon footprint analysis The carbon footprint approach, which is a method of estimating the emission of carbon dioxide equivalent (ECE) of a product during its life-cycle time, is adopted to evaluate the global warming impact of the ORC. The ECE is composed of direct emissions (ECEdirect) and indirect emissions (ECEindirect). The direct emissions are composed of the working fluid’s annual leak during the operation stage and fluid loss at the disposal stage of the system. The indirect emissions come from the manufacturing emissions and disposal processes of materials and fluids as well as the energy consumption. The equations of these emissions can be represented as [30] ECEdirect 5 Mfc ðLTUεALR 1 εEOL ÞGWP

ð3:20Þ

ECEindirect 5 ECEtrans 1 LTUwAEC UECE 1 Mmaterial UðECM 1 ECR Þ 1 ðMfc 1 LTUεALR UMfc ÞUECFM 1 Mfc Uð1 2 εEOL ÞUECFR

ð3:21Þ

where Mfc is the quality of fluid charge in the system (kg); εALR is the annual leak rate of working fluid; εEOL is the end-of-life loss rate of working fluid; ECEtrans is the emission from transportation calculated as ECEtrans 5 Mtot x Distance x eCO2,trans, where ‘Distance’ is the average transport distance by truck obtained from the statistics in China, which is equal to 410.78 km; eCO2,trans is the unit emission for transportation discharged by truck, which is 23.77 g CO2,eq/kg  km [31]; wAEC is the annual power consumption (kWh); ECE is the ECE produced by unit power consumption (kg CO2,eq/kWh); Mmaterial is the mass of materials (kg); ECM is the ECE produced by unit material manufacturing (kg CO2,eq/kWh); ECM is the ECE produced by unit material disposal (kg CO2,eq/kWh) and ECFM and ECFR are the ECE produced from working fluid manufacturing (kg CO2,eq/kWh) and disposal (kg CO2,eq/kWh), respectively. The total ECE and emission reduction can be determined as follows: G

Total ECE ECEtot 5 ECEdirect 1 ECEindirect

G

ð3:22Þ

Emission reduction of CO2,eq ER 5 Wnet top LTUeelec 2 CFtot

ð3:23Þ

Organic Rankine cycle driven by geothermal heat source Chapter | 3

83

where eelec is the CO2,eq generated by geothermal power plant for 1 kWh electricity production, which is 0.131 kg CO2,eq/kWh [32].

3.3.4.3 Data sources The mass of steel consumed for the different components of the ORC can be calculated by the following equations. For preheater (plate heat exchanger): Mpre 5 ρVpre 5 ρδApre

ð3:24Þ

For evaporator (plate heat exchanger): Mevap 5 ρVevap 5 ρδAevap

ð3:25Þ

Mtur  31:22UWtur

ð3:26Þ

For turbine [33]:

For condenser (finned-tube heat exchanger): 
  
2 π do 2 di2 Acond L π dt2 2 di2 δf Mcond 5 ρVfinned-tube 5 ρ U 1 U Y 4 πdo 4

ð3:27Þ

For pump [33]: Mpump  14UWpump

ð3:28Þ

The steel consumed for the turbine and pump is based on the power output and electricity consumption, respectively. The charged mass of working fluid is estimated empirically according to the power output of the ORC, which is 5.57 kg/kW [33]. The CO2,eq emissions of manufacturing and disposal processes for different working fluids are listed in Table 3.9. With

TABLE 3.9 CO2,eq emissions of working fluids during manufacturing and disposal processes [14,34,35]. Working fluid R1270 R1234yf

CO2,eq emissions of working fluid during manufacturing process (kg CO2,eq/kg)

CO2,eq emissions of working fluid during disposal process (kg CO2,eq/kg)

0.5

0.06

13.7

2.04

R290

0.67

0.05

R134a

5

1.55

13.7

1.16

0.9

3.03

R1234ze(E) R600a

84

PART | I Technologies

TABLE 3.10 CO2,eq emissions of materials during manufacturing and disposal processes [30]. Material item

CO2,eq emissions of materials during manufacturing process (kg CO2,eq/kg)

CO2,eq emissions of materials during recycling process (kg CO2,eq/kg)

Aluminum

4.50

0.07

Copper

1.64

0.07

Plastics

2.61

0.01

Steel

1.43

0.07

respect to the material, aluminum, copper, plastic and steel are taken into consideration according to an estimated percentage of composition of 12%, 19%, 23% and 46%, respectively. The details of ECE from manufacturing and recycling processes of materials are given in Table 3.10.

3.3.5

Multicriteria integrated assessment and decision-making

When comprehensively considering the system performance based on different criteria, including energetic, exergetic, economic, and environmental aspects, the multicriteria integrated assessment is a more effective tool to solute these problems. The analytic hierarchy process (AHP) is a representative way to evaluate the complex multifactor or multicriteria decision problems. In this study, AHP is adopted to determine the weight allocation of multiindicators. A series of primary criteria and secondary criteria comprises the AHP framework. First the structure of the AHP framework consists of three hierarchies, as shown in Fig. 3.5. There are four primary criteria for the goal of multifactor feasibility assessment: energetic, exergetic, economic, and environmental criteria. In addition, there are eight indicators for the four criteria: net power output, thermal efficiency, total exergy loss, exergy efficiency, EPC, cost per exergy unit of the net produced power, total CO2,eq emissions, and CO2,eq emission reductions. Then after the normalization and gradation processes, the effect of the eight factors can be superposed by multiplying their weights with the gradation level to generate a single evaluation indicator, which is called the feasibility level (FL). The FL can be determined as FLðXÞ 5

n X

ðxi Uwi Þ; n 5 1; 2; 3; . . .; N

ð3:29Þ

i51

where xi is the factor i and wi is the ith factor weight. A higher multifactor evaluation indicator (FL) means that the working fluid is more suitable for

Organic Rankine cycle driven by geothermal heat source Chapter | 3

85

FIGURE 3.5 Hierarchy network structure of multifactor evaluation for an ORC.

the ORC. The steps of the AHP can be found in our previous work [14]. xi also represents each indicator’s normalization value and can be calculated as follows: Xi ; Xi AðE1; E2; EX2; EN2Þ Xopt

ð3:30Þ

Xopt ; Xi AðEX1; EC1; EC2; EN1Þ Xi

ð3:31Þ

xi 5 xi 5

3.4 3.4.1

Thermodynamic and economic results Effects of design parameters on thermodynamic performance

The thermodynamic parameters of a basic ORC with different working fluids are optimized based on the objective of net power output and under the same heat load of geothermal source. The optimized parameters of each state are given in Table 3.11. The lowest evaporating pressure is approximately 1290 kPa when R600a is chosen as the candidate fluid. The range of evaporating pressure is from approximately 2000 to 4000 kPa for the other fluids, that is R1270, R1234yf, R290, R134a and R1234ze(E), whereas R1270 shows the highest evaporating pressure of 3972.03 kPa. Furthermore it should be noted that only R600a and R1234ze(E) exhibit no superheat during the evaporation process. The other fluids, that is R1270, R1234yf, R290 and R134a, all have the superheated region, and the corresponding superheat degrees are 11.91 C, 3.81 C, 6.59 C, and 6.12 C, respectively. Figs. 3.6
3.8 depict the thermodynamic performance of the basic ORC with various working fluids. The ORC using R1234yf exhibits the greatest power output and the highest thermal and exergy efficiencies if only the thermodynamic performance is considered, as shown in Figs. 3.6 and 3.7. The corresponding values of these three parameters are 173.19 kW, 10.30%, and

TABLE 3.11 Optimization parameters of each state based on the net power output in a basic ORC. Working fluid

State

T ( C)

p (kPa)

m (kg/s)

h (kJ/kg)

s (kJ/kg  K)

E (kW)

R1270 1

95.46

3972.03

4.55

653.58

2.34

856.77

2

33.79

1304.98

4.55

609.90

2.36

624.10

3

30.00

1304.98

4.55

277.20

1.26

573.91

4

32.87

3972.03

4.55

284.31

1.27

598.52

5

83.55

3972.03

4.55

455.16

1.78

692.87

1

98.51

3382.00

10.42

405.07

1.61

662.10

2

33.23

783.51

10.42

385.25

1.62

420.41

3

30.00

783.51

10.42

240.51

1.14

370.46

4

32.40

3382.00

10.42

243.72

1.14

395.87

5

94.70

3382.00

10.42

368.18

1.51

583.30

1

89.55

3309.42

4.54

655.23

2.35

775.48

2

33.59

1079.00

4.54

612.99

2.37

551.11

1079.00

4.54

278.83

1.27

500.86

R1234yf

R290

3

30,00

4

32.28

3309.42

4.54

284.94

1.27

521.95

5

82.96

3309.42

4.54

451.04

1.77

610.61

R134a 1

87.85

2731.68

8.58

440.00

1.71

600.89

2

33.58

770.20

8.58

418.60

1.73

386.22

3

30.00

770.20

8.58

241.72

1.14

335.99

4

31.54

2731.68

8.58

243.92

1.15

350.29

5

81.73

2731.68

8.58

325.72

1.39

429.75

1

81.86

2088.99

9.08

427.76

1.68

527.13

2

34.10

578.33

9.08

407.92

1.69

316.47

3

30.00

578.33

9.08

240.78

1.14

266.18

4

31.16

2088.99

9.08

242.53

1.14

278.26

5

81.86

2088.99

9.08

320.40

1.38

357.30

1

78.16

1293.61

4.40

655.25

2.36

446.10

2

41.70

404.72

4.40

616.20

2.38

245.69

3

30.00

404.72

4.40

271.24

1.25

193.84

4

30.70

1293.61

4.40

273.42

1.25

201.11

5

78.16

1293.61

4.40

399.39

1.63

259.55

R1234ze(E)

R600a

88

PART | I Technologies

FIGURE 3.6 Optimal net power output of the basic ORC with various working fluids as well as the corresponding thermal efficiency.

53.64%, respectively. When the net power output is selected as the optimization objective, R600a shows the lowest thermal efficiency and highest total exergy loss, as shown in Figs. 3.6 and 3.7, respectively. To explore the exergy distribution of the components in the basic ORC, the exergy losses for the different components are illustrated in Fig. 3.8. When the geothermal water inlet and outlet temperatures are set as 110 C and 70 C, the exergy loss in the heat exchanger is the greatest, followed by those in the turbine and pump, as plotted in Fig. 3.8. The exergy loss in the preheater region is nearly equal to that in the evaporating region except for R1234yf, which accounts for 21%
26%. With respect to working fluid R1234yf, the evaporating pressure is almost near the critical pressure; and thus the two-phase region during the evaporation process is smaller than that for the other fluids, and the exergy loss in the region is also much smaller. The corresponding exergy losses for fluid R1234yf are 35.63, 5.83, 25.66, 27.01, and 5.87 kW for the preheater, evaporator, turbine, condenser, and pump, respectively. However, for the other fluids, the exergy losses of the evaporator, turbine, and condenser account for 19%
31%, 19%
23%, and 25%
26%, respectively. As for the pump, the exergy loss is obviously small and it accounts for 1%
5% for all fluids. At the optimization state the location of

Organic Rankine cycle driven by geothermal heat source Chapter | 3

89

FIGURE 3.7 Total exergy loss and exergy efficiency of the basic ORC with various working fluids.

pinch-point temperature in the evaporator is at the bubble point, as presented in Fig. 3.2, while the condenser pinch-point temperature difference is located in the dew point. The corresponding pinch-point temperatures in the evaporator are 5 C, 6.19 C, 5.02 C, 5 C, 5 C, and 5.07 C for R1270, R1234yf, R290, R134a, R1234ze(E), and R600a, respectively. On the other hand, the optimal condenser pinch-point temperature is 5 C for all fluids.

3.4.2

Effects of design parameters on economic performance

The optimization working conditions, thermal performance, and areas of heat exchangers of the basic ORC with various working fluids based on the optimal EPC are listed in Table 3.12. According to the economic analysis, the lowest evaporating pressure is approximately 1283 kPa for the ORC with R600a. The range of evaporating pressure for the other fluids is from 2000 to 3900 kPa. The optimized pinch-point temperatures in the evaporator and condenser are nearly 5 C and 7 C for fluids R290, R134a, R1234ze(E), and R600a, and those of R1234yf are 5.87 C and 8.35 C, respectively. Moreover, the results of the thermal performance reveal that R1234yf has the highest net power output, thermal efficiency, and exergy efficiency,

90

PART | I Technologies

FIGURE 3.8 Exergy loss of different components in the basic ORC with various working fluids as well as the corresponding proportions.

which are 164.80 kW, 9.80%, and 51.04%, respectively. It also should be noted that R600a has the worst economic performance. The Wnet, ηth, and ηex are 155.73 kW, 9.26%, and 48.23%, respectively. With respect to the heat exchanger areas, for R1234yf, the preheater has a larger area than the evaporator while for other fluids, the evaporator area is larger. This is because the evaporating pressure of R1234yf is close to the critical pressure and its twophase region is much smaller. The condenser areas for fluids R1270, R1234yf, R290, R134a, R1234ze(E), and R600a are 939.74, 843.72, 958.94, 972.05, 978.79, and 951.40 m2, respectively. Fig. 3.9 illustrates the economic performance of the basic ORC with various working fluids. The basic ORC with R290 displays the highest total cost and EPC while the ORC with R1234yf shows the smallest ones. This indicates that under an inlet geothermal water of 110 C, the ORC with R1234yf has the best economic performance among the fluids; its EPC and total cost are 0.19 $/kWh and 2.00 M$, respectively. As shown in Fig. 3.10, the primary source of capital cost for the components of the basic ORC is the turbine cost, which accounts for 50.97%
58.34% in all fluids. With respect to

TABLE 3.12 Optimization working conditions of the basic ORC based on the optimal electricity production cost. Working fluid Evaporating pressure (kPa) 

Pinch point temperature in evaporator ( C)

R1270

R1234yf

R290

R134a

R1234ze(E)

R600a

3930.45

3382.00

3279.11

2703.68

2064.39

1283.51

5.01

5.87

5.00

5.00

5.03

5.03

1375.89

854.62

1134.18

813.35

609.99

426.46

Pinch point temperature in condenser ( C)

7.32

8.35

7.14

7.01

6.95

7.17

Mass flow rate of working fluid (kg/s)

4.65

10.70

4.62

8.72

9.21

4.46

Net power output (kW)

158.66

164.80

157.03

158.00

157.76

155.73

Thermal efficiency (%)

9.43

9.80

9.34

9.40

9.38

9.26

49.14

51.04

48.63

48.93

48.86

48.23

Condensing pressure (kPa)

Exergy efficiency (%) 2

Area of preheater (m )

18.60

33.06

19.47

19.63

20.12

16.27

Area of evaporator (m2)

94.15

15.70

117.00

136.29

138.75

129.41

Area of condenser (m2)

939.74

843.72

958.94

972.05

978.79

951.40

92

PART | I Technologies

FIGURE 3.9 Total cost and electricity production cost of the basic ORC with different working fluids.

fluids R1270, R134a, R290, R1234ze(E), and R600a, the cost of condenser of approximately 20.02%
21.32% ranks the second, followed by those of the evaporator, preheater and pump. However, with respect to R1234yf, the cost rank of the different components in decreasing order is as follows: turbine, condenser, preheater, evaporator, and pump.

3.4.3 Effects of design parameters on exergoeconomic performance Table 3.13 presents the system parameters, exergy flow rates, unit exergy costs, and costs flow rates for the basic ORC system with various working fluids. As shown, the unit exergy costs of the turbine are 14.90, 12.53, 14.97, 14.14, 13.74, and 12.81 $/GJ for the ORC with R1270, R1234yf, R290, R134a, R1234ze(E), and R600a, respectively. This indicates that for different working fluids and various pressure ratios, the unit exergy cost of the turbine differs. The exergy and exergoeconomic parameters for the various components of the basic ORC with different working fluids are provided in Table 3.14. According to the exergoeconomic analysis, the evaporator in the basic ORC

Organic Rankine cycle driven by geothermal heat source Chapter | 3

93

FIGURE 3.10 Costs of components in the basic ORC with various working fluids as well as the corresponding proportions.

has the highest ED. However, the highest value of Z_ 1 C_ D belongs to the turbine. Thus it is important to pay more attention to this component. With respect to R1270, the Z_ 1 C_ D of the turbine is 20.42 $/h. The preheater has the highest exergoeconomic factor of 96.91%, which means that the investment cost is the main source of the cost rate. The condenser has the second highest value of Z_ 1 C_ D , and its exergoeconomic factor is 80.47%, indicating that 19.53% of its cost is related to exergy destruction. Similarly the condenser also shows the second highest Z_ 1 C_ D for the ORCs with other fluids. Moreover the lowest exergoeconomic factor belongs to the pump in the basic ORC, whose values are 42.05%, 42.46%, 42.69%, 47.07%, 48.18%, and 53.15% for R1270, R1234yf, R290, R134a, R1234ze(E), and R600a, respectively. A low exergoeconomic factor means that the exergy destruction cost in this component is higher than the investment cost, and it would be suggested to add capital investments in order to improve the exergoeconomic performance. Fig. 3.11 displays the cost per exergy unit of the net power produced by the basic ORC with various working fluids. It can be noted that the ORC

94

PART | I Technologies

TABLE 3.13 System parameters, exergy flow rates, and costs for the basic ORC with different working fluids. Working fluid

State

T ( C)

p (kPa)

m (kg/s)

E (kW)

c ($/GJ)

C ($/h)

1

95.46

3972.01

4.55

856.77

14.90

45.94

2

33.79

3

30.00

1304.98

4.55

624.10

14.90

33.47

1304.98

4.55

573.91

14.90

30.78

4

32.87

3972.01

4.55

598.52

17.04

36.71

5

83.55

3972.01

4.55

692.87

16.10

40.15

1

98.52

3382.00

10.44

663.33

12.53

29.93

2

33.43

788.20

10.44

422.30

12.53

19.054

3

30.22

788.20

10.44

371.28

12.53

16.75

4

32.63

3382.00

10.44

396.71

15.73

22.47

5

94.70

3382.00

10.44

584.34

12.75

26.83

1

87.57

3207.55

4.55

771.73

14.97

41.59

2

33.49

1079.00

4.55

552.58

14.97

29.78

3

30.00

1079.00

4.55

502.23

14.97

27.07

4

32.18

3207.55

4.55

522.41

17.04

32.04

5

81.28

3207.55

4.55

605.68

16.23

35.39

1

84.91

2589.98

8.61

594.92

14.14

30.28

2

33.44

770.20

8.61

387.81

14.14

19.74

3

30.00

770.20

8.61

337.43

14.14

17.17

4

31.43

2589.98

8.61

350.75

16.25

20.52

5

79.22

2589.98

8.61

423.42

15.61

23.80

1

78.72

1953.24

9.10

519.11

13.74

25.68

2

34.32

578.33

9.10

317.46

13.74

15.70

3

30.00

578.33

9.10

266.93

13.74

13.20

R1270

R1234yf

R290

R134a

R1234ze(E)

(Continued )

Organic Rankine cycle driven by geothermal heat source Chapter | 3

95

TABLE 3.13 (Continued) Working fluid

State

T ( C)

p (kPa)

m (kg/s)

4

31.06

1953.24

5

78.72

1

E (kW)

c ($/GJ)

C ($/h)

9.10

277.96

15.97

15.98

1953.24

9.10

348.82

15.33

19.26

76.83

1258.30

4.42

442.81

12.81

20.42

2

41.40

404.72

4.42

246.44

12.81

11.37

3

30.00

404.72

4.42

194.56

12.81

8.97

4

30.67

1258.30

4.42

201.57

14.85

10.77

5

76.83

1258.30

4.42

257.47

14.97

13.87

R600a

with R600a generates the lowest cP,sys, followed by ORCs with R1234yf, R1234ze(E), R134a, R1270, and R290, and the corresponding values are 83.16, 85.95, 88.23, 90.12, 95.42, and 95.44 $/GJ, respectively. It must be pointed out that fluid R600a is the best candidate for the basic ORC under the given conditions based on the exergoeconomic objective. This is different from the one discussed earlier. When Wnet and EPC are selected as the optimization objectives, fluid R1234yf is the most suitable candidate. However, when the optimization is changed, the calculation results also change and the conclusion differs. This also demonstrates the limitation of the singleobjective optimization, and the selection of candidate fluid in the basic ORC is constrained by the choice of the objectives.

3.4.4 Sensitivity analysis on the economic performance and inlet temperature of geothermal source The method of control variates is adopted to conduct the sensitivity analysis of economic performance on the basic ORC with various working fluids. Fig. 3.12 depicts the sensitivity analysis of annual loan interest, life-cycle time, and annual operating time on economic indicator EPC. Taking fluid R1270 as an example as shown in Fig. 3.12A, the EPCs are 0.20, 0.21, and 0.22 $/kWh with respect to the annual loan interest rates of 5%, 6%, and 7%, respectively. For the other fluids the same increase trend is indicated, and the EPC shows an average increase rate of 6.02% with an increase in loan interest rate. On the contrary, the EPC presents a decreasing trend with the increases in life-cycle time and annual operating time. Taking fluid R290 as an example, the EPCs are 0.22, 0.18, and 0.17 $/kWh with respect to the

TABLE 3.14 Exergy and exergoeconomic parameters of the basic ORC. Working fluid

E_ D (kW)

ε (%)

_ D ($/h) C

94.35

36.36

72.18

0.17

2.83

3.00

94.32

192.17

163.90

28.27

85.29

0.13

4.89

5.02

97.37

Component

E_ F (kW)

E_ P (kW)

Preheater

130.71

Evaporator Turbine

Z_ ($/h)

_ D ($/h) Z_ 1 C

f (%)

R1270

232.68

198.94

33.74

85.50

1.81

18.62

20.42

91.14

Condenser

50.19

12.78

37.41

25.46

2.01

8.27

10.27

80.47

Pump

32.37

24.61

7.76

76.03

1.21

0.88

2.10

42.05

Preheater

235.86

187.69

48.17

79.58

0.23

3.26

3.48

93.54

Evaporator

87.03

79.05

7.98

90.83

0.04

2.69

2.73

98.63

240.83

205.91

34.92

85.50

1.58

19.02

20.59

92.35

Condenser

51.34

12.73

38.62

24.79

1.74

7.98

9.72

82.07

Pump

33.46

25.44

8.03

76.01

1.16

0.86

2.02

42.46

Preheater

120.91

83.26

37.64

68.87

0.18

2.79

2.97

94.06

Evaporator

201.98

166.05

35.93

82.21

0.17

5.25

5.42

96.90

Turbine

219.14

187.35

31.79

85.49

1.71

17.92

19.64

91.27

Condenser

50.35

12.83

37.53

25.47

2.02

8.28

10.30

80.36

Pump

26.56

20.18

6.38

75.98

1.01

0.75

1.77

42.69

R1234yf

Turbine

R290

R134a Preheater

109.56

72.68

36.89

66.33

0.17

2.76

2.93

94.12

Evaporator

213.32

171.50

41.82

80.39

0.20

5.48

5.68

96.55

Turbine

207.11

177.07

30.05

85.49

1.53

17.29

18.82

91.87

Condenser

50.38

12.84

37.54

25.48

1.91

8.26

10.17

81.20

Pump

17.54

13.32

4.22

75.92

0.66

0.59

1.25

47.07

Preheater

108.94

70.86

38.08

65.04

0.19

2.76

2.94

93.94

Evaporator

213.94

170.29

43.65

79.60

0.20

5.42

5.62

96.37

Turbine

201.65

172.47

29.18

85.53

1.44

17.00

18.44

92.17

Condenser

50.53

12.85

37.68

25.43

1.86

8.23

10.09

81.53

Pump

14.53

11.02

3.51

75.90

0.55

0.51

1.06

48.18

Preheater

87.19

55.91

31.29

64.12

0.15

2.69

2.84

94.84

Evaporator

235.69

185.33

50.36

78.63

0.24

5.45

5.68

95.85

Turbine

196.37

168.50

27.87

85.81

1.29

16.74

18.03

92.87

51.87

12.84

39.04

24.75

1.80

8.04

9.84

81.70

9.23

7.00

2.23

75.87

0.34

0.39

0.73

53.15

R1234ze(E)

R600a

Condenser Pump

98

PART | I Technologies

FIGURE 3.11 Costs per exergy unit for the basic ORC with various working fluids.

life-cycle time of 15, 20, and 25 years (Fig. 3.12B), which exhibits an average decreasing rate of 13.86%. Similarly the EPCs are 0.21, 0.19, and 0.18 $/kWh with respect to annual operating time of 7000, 7500, and 8000 h, showing an average reduction rate of 6.90% for fluid R600a (Fig. 3.12C). Therefore the extensions of life-cycle time and annual operating time could improve the economic performance of the basic ORC. At the same time, a suitable loan interest by the government would greatly encourage the development of ORC system. Studies have demonstrated that the inlet temperature of the heat source has a great influence on the system performance of ORCs [6,36]. As shown in Fig. 3.13 the maximum net power output, minimum EPC, and cost per exergy unit are greatly affected by the inlet temperature of the geothermal source of the basic ORC. With the increase in inlet temperature of the geothermal source, the net power output is monotonically increasing for all fluids. Taking fluid R1234ze(E) as an example (Fig. 3.13A), the maximum net power outputs are 164.24, 222.89, 308.64, and 360.72 kW with the inlet temperature of the geothermal source increasing from 110 C to 120 C, 130 C, and 140 C, respectively, with corresponding increase rates of 35.71%, 38.47%, and

Organic Rankine cycle driven by geothermal heat source Chapter | 3

99

FIGURE 3.12 Sensitivity analysis of (A) loan interest rate, (B) life cycle time, and (C) annual operation time on electricity production cost.

16.88%. On the contrary, the minimum EPC is monotonically decreasing with the increase in inlet temperature of the geothermal source as illustrated in Fig. 3.13B. Taking R134a as an example, the minimum EPCs are 0.21, 0.18, 0.14, and 0.13 $/kWh with respect to the geothermal source inlet temperatures of 110 C, 120 C, 130 C, and 140 C. The average reduction rate is 17.91% for EPCs of inlet temperature from 110 C to 140 C. The influence of geothermal source inlet temperature on the cost per exergy unit is the same as that on the EPC as exhibited in Fig. 3.13C. That is, a higher inlet temperature of the geothermal source can improve the thermodynamic, economic and exergoeconomic performance of the basic ORC.

3.5 Life-cycle and carbon footprint analysis of the organic Rankine cycle 3.5.1

Environmental evaluation of life cycle

The optimal parameters of the basic ORC with various working fluids based on the maximum emission reductions are listed in Table 3.15. The lowest

100

PART | I Technologies

FIGURE 3.13 (A) Maximum net power output, (B) minimum electricity production cost, and (C) minimum cost per exergy unit of various working fluids at different inlet temperature of geothermal source.

and highest evaporating pressures are 1284.42 and 3967.04 kPa for fluids R600a and R1270, respectively. The range of net power output is from 158 to 173 kW for all selected six fluids. With respect to heat exchanger area, the area of the condenser is the largest, which ranges from 1093 to 1226 m2. It should be noted that for R1234yf, the preheater has a larger area than the evaporator because the evaporating pressure is very close to the critical pressure. On the other hand, the condensing temperature is approximately 31 C for all fluids, while the pinch point temperature in the condenser is from 5 C to 5.92 C. Fig. 3.14 exhibits the CO2,eq emissions at the different phases and the total electricity produced by the basic ORC with various working fluids. The annual leak rate and end loss rate of the working fluid are assumed as 5% and 15% [30,31], respectively, and the fluid in the system is recharged every year to keep the ORC running steadily. It can be observed that the CO2,eq emissions from the construction phase account for the highest proportion, which ranges from 78.86% to 94.63% for all fluids except R134a. For R134a, the CO2,eq emissions from the operation and decommissioning phases

TABLE 3.15 Optimal parameters of the basic ORC with various working fluids based on the maximum emission reductions. Working fluid Evaporating pressure (kPa) 

Condensing temperature ( C) 

Pinch point temperature in condenser ( C) Net power output (kW) 2

Area of preheater (m )

R1270

R1234yf

R290

R134a

R1234ze(E)

R600a

3967.04

3382.00

3244.99

2715.71

2073.98

1284.42

30.21

30.00

30.87

31.10

31.01

30.97

5.00

5.00

5.00

5.70

5.92

5.77

165.80

173.19

159.42

160.85

160.62

158.61

18.58

31.48

18.46

19.66

20.16

16.13

2

Area of evaporator (m )

91.72

14.92

111.50

135.77

138.16

127.15

Area of condenser (m2)

1226.29

1214.35

1190.79

1107.77

1093.88

1093.54

102

PART | I Technologies

FIGURE 3.14 Total emissions of CO2 equivalent (right) and total electricity production (left) of the basic ORC with various working fluids during life-cycle time.

are larger than those from the construction phase because it has the highest GWP of 1370 among all the selected working fluids. The corresponding CO2,eq emissions are 90.69, 1031.19, and 199.20 tons for the construction, operation and decommissioning phases of the ORC with R134a. In this regard the environmental restriction on the GWP of organic fluids has profound implications, and it could effectively reduce the CO2,eq emissions during the operation and decommissioning phases if the leak phenomenon occurs in the lifetime of the ORC system. With respect to the total electricity production of the basic ORC using different working fluids, the system with R1234yf produces the highest electricity of 18 094.43 MWh, followed by R1270, R134a, R1234ze(E), R290, and R600a, with values of 17 409.36, 16 889.30, 16 865.47, 16 739.34, and 16 654.23 MWh. As discussed earlier, imposing restrictions on the GWP of the working fluid could reduce the CO2,eq emissions during the operation and decommissioning phases. Meanwhile the CO2,eq emissions of the basic ORC can be classified by another viewpoint into three parts, as shown in Fig. 3.14: emissions from components, emissions from fluids and emissions from transportation. With respect to the emissions from transportation, the ORC with R1270 produces the highest CO2,eq emissions of approximately 1.32 tons during the life-cycle time, which accounts for approximately 1.33% of the

Organic Rankine cycle driven by geothermal heat source Chapter | 3

103

FIGURE 3.15 Emissions of CO2 equivalent and the corresponding proportions of components, working fluids, and transportation during life-cycle time.

total CO2,eq emissions. It can be observed that the emissions from transportation are much smaller than those from the components and fluids. As illustrated in Fig. 3.15, it also can be found that the emissions from the components range from 83.79 to 95.30 tons, which are the main sources of pollution except for the ORC with R134a. Moreover with respect to R134a, the main CO2,eq emissions are from the fluid, and the proportions for the components, fluid, and transportation are 6.60%, 93.31%, and 0.09%, respectively. The CO2,eq emissions from the fluid are almost 1233 tons during its lifetime. It is clearly observed that the usage of high GWP fluids would highly contaminate the environment.

3.5.2

Environmental evaluation of components

Fig. 3.16 displays the CO2,eq emissions and proportions from various components in the ORC with different working fluids during its lifetime. The main CO2,eq emissions are from the condenser, followed by those from the turbine. The condenser emissions range from 54.18 to 60.76 tons, which correspond

104

PART | I Technologies

FIGURE 3.16 Emissions of CO2 equivalent of components during life-cycle time and the proportions of preheater, evaporator, turbine, condenser, and pump.

to proportion of 62.98%
64.66%. For the turbine, the ranges of emissions and proportions are 24.70
30.25 tons and 29.49%
32.18%, respectively. The emissions from the pump, whose proportions range from 0.47% to 0.70%, are the least among the emissions from various components.

3.5.3

Environmental evaluation of working fluids

The CO2,eq emissions of the working fluids during their lifetime and the proportions of charging, leak, and disposal processes are shown in Fig. 3.17. It should be noted that for hydrofluoroolefins, including R1234yf and R1234ze (E), the main source of emissions is from the charging process. Taking R1234yf as an example the CO2,eq emissions for charging, leak and disposal processes are 27.48, 4.09, and 1.99 tons, respectively, and the corresponding proportions are 81.88%, 12.19%, and 5.92%. With respect to the low GWP fluids, the emissions from the leak process are the highest, followed by the charging and disposal processes. However, there is a different phenomenon on the emissions from R134a. It can be observed that most of the CO2,eq emissions are from the leak process, which account for approximately

Organic Rankine cycle driven by geothermal heat source Chapter | 3

105

FIGURE 3.17 Emissions of CO2 equivalent of working fluids during life-cycle time and the proportions of charging, leak, and disposal processes.

99.18%, and this is caused by the high GWP of R134a. Moreover the values of the emissions from R134a fluid are much higher than those from other fluids.

3.5.4

Analysis of emission reductions

The earlier discussion is focused on the various emission sources of the basic ORC. However, as an effective technology for waste heat power generation, more attention should be focused on the emission reduction to comprehensively evaluate the environmental benefits of the ORC. Therefore the emission reductions of the basic ORC with various working fluids are evaluated during its life-cycle time as shown in Fig. 3.18. The emission reductions of various working fluids are different. Among the selected six working fluids, the ORC with R1234yf demonstrates the highest emission reduction while the ORC using R134a presents the lowest one. Their corresponding values are 2252.01 and 891.42 tons, respectively. In contrast, the range of emission reductions for the other fluids is from 2076.16 to 2181.01 tons. It can be observed that ORCs with high GWP fluids result in a lower environmental

106

PART | I Technologies

FIGURE 3.18 Emissions reductions of CO2 equivalent of different working fluids during lifecycle time.

benefit than ORCs with low GWP fluids. However, many organic fluids with high GWP including R134a are safer, particularly in terms of nonflammability. In contrast, HFOs including R1234yf and R1234ze(E) and hydrocarbons including R600a and R290 are organic fluids that are inflammable. Thus the requirement for system tightness is relatively higher. This is the reason that high GWP fluids are not completely replaced by low GWP organic fluids. Therefore the fluid selection for system design depends on whether the system safety or environmental benefits is the top priority.

3.5.5

Sensitivity analysis

The sensitivity analysis of the life-cycle time and annual operating time on the emission reduction of CO2,eq is depicted in Fig. 3.19. Taking R1234ze(E) as an example the emission reductions are 2092.52, 2824.14 and 3555.76 tons corresponding to the life-cycle times of 15, 20, and 25 years as

Organic Rankine cycle driven by geothermal heat source Chapter | 3

107

FIGURE 3.19 Sensitivity analysis of (A) life-cycle time and (B) annual operation time on emission reductions of CO2,eq.

shown in Fig. 3.19A. The average increase rate is 30.43% with the increase in life-cycle time. Similarly the increase in annual operating time brings more environmental benefits. The emission reductions are 2092.52, 2250.33, and 2408.14 tons with respect to the annual operating time of 7000, 7500, and 8000 h, showing an average increase rate of 7.28% for fluid R1234ze(E) as exhibited in Fig. 3.18B. Thus a suitably longer life cycle time and annual operating time can improve the environmental benefits of the basic ORC. The influence of geothermal source inlet temperature on emission reductions is shown in Fig. 3.20. For the ORCs with low GWP fluids, the maximum emission reduction is greatly affected by the inlet temperature of the heat source. As an example for R1270, the maximum emission reductions are 2181.01, 2947.76, 3582.96, and 4187.87 tons with respect to the geothermal source inlet temperatures of 110 C, 120 C, 130 C, and 140 C, which show increase rates of 35.16%, 21.55%, and 16.88%, respectively. It can also be observed that the increase trend is higher when the inlet temperature of the geothermal source is low. For the ORCs with high GWP fluids, there is also an increase trend. However, this trend is relatively lower than for the other fluids. For R134a, the emissions reductions are 891.42, 1210.28, 1578.77, and 1848.20 tons when the inlet temperatures are 110 C, 120 C, 130 C, and 140 C, respectively. This indicates that a higher temperature of geothermal source would enhance the environmental benefits of the basic ORC. On the other hand, the optimal fluid with the maximum emission reduction is changed as the inlet temperature increases from 110 C to 140 C. When the temperature is 110 C, R1234yf is the optimal. However, when the temperature increases to 120 C, the optimal fluid is R290. At 130 C and 140 C, the optimal fluid is R1234ze (E). It can be observed that there is a correlation between the heat source

108

PART | I Technologies

FIGURE 3.20 Sensitivity analysis of geothermal source inlet temperature on emission reductions of CO2,eq.

temperature and organic fluids, and this had been pointed out in related works [6,36].

3.6 Comparison between different layouts of organic Rankine cycle systems To compare various layouts of ORCs, the cycles have been optimized from the viewpoints of minimum EPC, minimum unit cost of power produced, and maximum emission reduction. Working fluid R1234yf is selected as a representative. The obtained results from the three optimized ORCs are presented in Tables 3.16
3.18. With respect to the specific ORC system from different viewpoints, the optimized working conditions under various optimal objects have different details. Taking the basic ORC as an example, the value of the optimum evaporating pressures is 3382 kPa for all objectives. However, the pinch-point temperatures in the evaporator are 5.87 C, 6.16 C,

Organic Rankine cycle driven by geothermal heat source Chapter | 3

109

TABLE 3.16 Optimum value for the basic ORC with R1234yf. System indicators

Minimum electricity production cost

Evaporating pressure (kPa)

Minimum unit cost of power produced

Maximum emission reduction

3382.00

3382.00

3382.00

Pinch point temperature in evaporator ( C)

5.87

6.16

6.19

Condensing temperature ( C)

33.24

30.29

30.00

8.35

5.41

5.00

164.80

172.44

173.19

Pinch point temperature in condenser ( C) Net power output (kW) EPC ($/kWh)
 C_ D;overall $=h
 Z_ overall $=h foverall (%)




C_ D;overall 1 Z_ overall $=h Cost per exergy unit ($/GJ) Total emissions (tons CO2,eq) Emission reductions (tons CO2,eq)



0.193

0.196

0.197

5.20

4.75

4.59

31.87

33.81

34.13

85.97

87.69

88.15

37.08

38.55

38.71

87.07

85.95

86.06

108,102.87

126,898.48

130,157.78

2158.73

2245.06

2252.01

and 6.19 C corresponding to the viewpoints of minimum EPC, minimum unit cost of power produced, and maximum emission reduction. There is a difference not only in the value of pinch point temperature in the evaporator but also in the condensing temperature and pinch point temperature in the condenser. With regard to the optimum values for various ORCs with the same objective, the minimum EPC is obtained for the basic ORC in which the EPC in the optimum state is 0.193 $/kWh, which is respectively 14.51% and 25.39% less than the IHX-ORC and dual-pressure ORC. On the other hand the net power output in dual-pressure is respectively 3.82% and 1.68% more than the corresponding value for the basic ORC and IHX-ORC. As it

110

PART | I Technologies

TABLE 3.17 Optimum value for the IHX-ORC with R1234yf. System indicators

Minimum EPC

Minimum unit cost of power produced

Maximum emission reduction

Evaporating pressure (kPa)

3382.00

3382.00

3382.00

Pinch point temperature in evaporator ( C)

5.94

6.14

6.14

Condensing temperature ( C)

32.09

30.00

30.00

Pinch point temperature in condenser ( C)

7.19

5.08

5.09

Net power output (kW)

168.47

174.10

174.11

Electricity production cost ($/kWh)
 C_ D;overall $=h
 Z_ overall $=h

0.221

0.222

0.222

6.70

6.16

6.16

37.18

38.71

38.71

foverall (%)

84.73

86.27

86.27

43.87

44.87

44.87

Cost per exergy unit ($/GJ)

103.79

102.55

102.55

Total emissions (tons CO2,eq)

116,819.47

132,333.38

132,295.85

Emission reductions (tons CO2,eq)

2200.44

2262.44

2262.47

C_ D;overall 1 Z_ overall


 $=h

can be observed, because of the additional components in the IHX-ORC and dual-pressure ORC, the net power output is improved compared to that of the basic ORC. However, the capital investment for unit electricity production is also increased due to the additional components. Among the three cycles, the minimum unit cost of power produced is also obtained from the basic ORC in which the optimum value is 85.95 $/GJ. This is 19.31% and 28.66% less than those of the IHX-ORC and dualpressure ORC, respectively. In addition, the value of C_ D 1 Z_ in the basic ORC is 38.55 $/h, which is 16.39% and 21.11% less than the corresponding values for the IHX-ORC and dual-pressure ORC. The value of foverall for the

Organic Rankine cycle driven by geothermal heat source Chapter | 3

111

TABLE 3.18 Optimum value for the dual-pressure ORC with R1234yf. System indicators

Minimum electricity production cost

Minimum unit cost of power produced

Maximum emission reduction

Evaporating pressure (kPa)

3373.19

2677.66

3373.00

Superheat degree in evaporator-H ( C)

7.16

21.12

7.50

Pinch point temperature in evaporator ( C)

5.00

5.01

5.00

Middle evaporating pressure (kPa)

2907.13

2634.69

2900.00

Condensing temperature ( C)

30.53

30.00

30.50

Pinch point temperature in condenser ( C)

5.81

5.85

5.80

Net power output (kW)

171.35

163.24

171.94

Electricity production cost ($/kWh)
 C_ D;overall $=h
 Z_ overall $=h

0.242

0.244

0.243

7.97

6.86

7.97

41.46

39.82

41.59

foverall (%)

83.87

85.30

83.92

49.43

46.69

49.56

Cost per exergy unit ($/GJ)

114.56

110.58

114.51

Total emissions (tons CO2,eq)

124,794.76

116,258.73

125,364.07

Emission reductions (tons CO2,eq)

2232.06

2129.16

2239.70

C_ D;overall 1 Z_ overall




 $=h

basic ORC is 87.69% which is higher than the values of 86.27% and 85.30% for the IHX-ORC and dual-pressure ORC, respectively. With regard to the maximum emission reduction, the IHX-ORC shows a higher amount compared to the basic ORC and dual-pressure ORC: the corresponding values are 2262.47, 2252.01 and 2239.70 tons CO2,eq. The rates of decrease are only

112

PART | I Technologies

0.46% and 1.01% for the basic ORC and dual-pressure ORC, respectively. It is also worth noting that the basic ORC exhibits a better thermoeconomic performance when the inlet and outlet temperatures of the geothermal heat source are 110 C and 70 C, respectively. On the other hand, the IHX-ORC has more environmental benefits under the given conditions.

3.7

Results of multifactor evaluation

In this study eight different cases are selected to evaluate the comprehensive feasibility of the subcritical ORC with various working fluids. The comparative analysis is based on the same heat load offered from the waste heat source but different weight allocations for the energetic, exergetic, economic and environmental criteria. For the ORC with various working fluids, the results based on the four criteria are presented in Table 3.19. First, all single indicators are normalized and transformed to nondimensional parameters as listed in Table 3.20. Then Table 3.21 displays the graded normalization indicators according to grading standards. Each value is represented by a level in the grading standards: 1 means extremely low, 2 means low, 3 means medium, 4 means high and 5 means extremely high. The grading values for the ORCs with different working fluids are summarized in Table 3.22. The results of multifactor indicators for the eight cases of ORCs with various working fluids are provided in Table 3.23. The weight allocations of energetic, exergetic, economic and environmental criteria are set as 100% in cases 1
4. The results reveal that the ORC with R1234yf has the maximal FLs, which are 4.96, 4.96, 4.63 and 4.96 for cases 1
4, and achieves the optimal thermoeconomic and environmental performances under the given working conditions. This conclusion is different from the previous multifactor analysis [14] because the type of ORC and heat source temperature is different. In terms of thermodynamic performances, R1234yf generates the maximum net power output and attains the highest thermal efficiency, followed by R1270. In terms of environmental impact, R1234yf still exhibits the best environmental performance, followed by R600a and R1270. Equal weight allocations are set for the four criteria in case 5. With regard to cases 6
8, different weight allocations are set and only the economic and environmental criteria are considered. It can be observed that R1234yf always shows the highest FL followed by R600a, R1234ze(E), R1270, R290 and R134a. This is different from those in cases 1 and 2. Therefore the multifactor analysis can be an effective way for multiobjective evaluation, and the feasibility of the adopted subcritical ORC can be easily evaluated quantitatively through the indicator FL based on the focus of decision makers.

TABLE 3.19 Results of energetic, exergetic, economic and environmental performances. Working fluid

Energetic criteria

Economic criteria

Environmental criteria

Itot (kW)

ηex (%)

EPC ($/kWh)

cp ($/kW)

R1270

166.57

9.91

143.24

51.59

0.21

95.42

43403.83

2181.01

R1234yf

173.19

10.30

136.66

53.64

0.19

85.95

49947.43

2241.49

R290

164.07

9.76

145.65

50.81

0.22

95.44

43647.68

2095.48

R134a

164.72

9.80

145.10

51.01

0.21

90.12

347912.75

891.42

R1234ze(E)

164.24

9.77

145.53

50.87

0.21

88.23

49686.32

2092.52

R600a

162.41

9.66

146.58

50.30

0.21

83.16

47273.25

2076.16

Wnet (kW)

ηth (%)

Exergetic criteria

ECE (tons CO2,eq)

ERCE (tons CO2,eq)

TABLE 3.20 Normalization of energetic, exergetic, economic and environmental performances. Working fluid

Energetic criteria

Exergetic criteria

Economic criteria

Environmental criteria

E1

E2

EX1

EX2

EC1

EC2

EN1

EN2

R1270

0.9618

0.9618

0.9541

0.9618

0.9116

0.8715

1.0000

0.9730

R1234yf

1.0000

1.0000

1.0000

1.0000

1.0000

0.9676

0.8690

1.0000

R290

0.9474

0.9474

0.9383

0.9474

0.8979

0.8713

0.9944

0.9349

R134a

0.9511

0.9511

0.9418

0.9511

0.9080

0.9228

0.1248

0.3977

R1234ze(E)

0.9483

0.9483

0.9390

0.9483

0.9118

0.9426

0.8736

0.9335

R600a

0.9378

0.9378

0.9323

0.9378

0.9322

1.0000

0.9181

0.9262

TABLE 3.21 Indicators grading standards. Indicator

E1

Indicator grading standards 1

2

3

4

5

0.9378
0.9502

0.9502
0.9627

0.9627
0.9751

0.9751
0.9876

0.9876
1.0000

E2

0.9378
0.9502

0.9502
0.9627

0.9627
0.9751

0.9751
0.9876

0.9876
1.0000

EX1

0.9323
0.9458

0.9458
0.9594

0.9594
0.9729

0.9729
0.9865

0.9865
1.0000

EX2

0.9378
0.9502

0.9502
0.9627

0.9627
0.9751

0.9751
0.9876

0.9876
1.0000

EC1

0.8979
0.9183

0.9183
0.9387

0.9387
0.9592

0.9592
0.9796

0.9796
1.0000

EC2

0.8913
0.8970

0.8970
0.9228

0.9228
0.9485

0.9485
0.9743

0.9743
1.0000

EN1

0.1248
0.2998

0.2998
0.4749

0.4749
0.6499

0.6499
0.8250

0.8250
1.0000

EN2

0.3977
0.5182

0.5182
0.6386

0.6386
0.7591

0.7591
0.8795

0.8795
1.0000

TABLE 3.22 Grading values of indicators for organic Rankine cycle with various working fluids. Working fluid

Energetic criteria

Exergetic criteria

Economic criteria

Environmental criteria

E1

E2

EX1

EX2

EC1

EC2

EN1

EN2

R1270

2.0000

2.0000

2.0000

2.0000

1.0000

1.0000

5.0000

5.0000

R1234yf

5.0000

5.0000

5.0000

5.0000

5.0000

4.0000

5.0000

5.0000

R290

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

5.0000

5.0000

R134a

2.0000

2.0000

1.0000

2.0000

1.0000

3.0000

1.0000

1.0000

R1234ze(E)

1.0000

1.0000

1.0000

1.0000

1.0000

3.0000

5.0000

5.0000

R600a

1.0000

1.0000

1.0000

1.0000

2.0000

5.0000

5.0000

5.0000

TABLE 3.23 Weights allocation of indicators for different cases in multifactor assessment and results of feasibility level with various working fluids. Criteria

Criteria weights for each case (%) Case 1

Case 2

Case 3

Case 4

Case 5

Case 6

Case 7

Case 8

Energy

100

0

0

0

25

0

0

0

Exergy

0

100

0

0

25

0

0

0

Economy

0

0

100

0

25

50

20

80

Environmental

0

0

0

100

25

50

80

20

Working fluids

Estimated feasibility level Case 1

Case 2

Case 3

Case 4

Case 5

Case 6

Case 7

Case 8

R1270

2.17

2.17

1.50

4.17

2.50

2.83

3.63

2.03

R1234yf

4.96

4.96

4.63

4.96

4.87

4.79

4.89

4.69

R290

1.33

1.33

1.33

3.96

1.99

2.65

3.44

1.86

R134a

1.88

1.54

1.88

1.27

1.64

1.57

1.39

1.75

R1234ze(E)

1.42

1.42

2.08

4.05

2.24

3.07

3.66

2.48

R600a

1.54

1.54

3.34

4.21

2.66

3.77

4.03

3.51

118

PART | I Technologies

3.8

Conclusions

An ORC for geothermal heat source utilization was modelled and evaluated through thermodynamic, thermoeconomic, exergoeconomic, and environmental approaches. A parametric study was conducted to analyze the effects of thermodynamic parameters on the system performance. The EPC, cost per exergy unit, and emission reduction during the life-cycle time were further optimized simultaneously. A comparative study between the basic ORC, IHX-ORC, and dual-pressure ORC was carried out. The main conclusions based on the study results are summarized as follows: G

G

G

G

G

G

G

For all selected working fluids, R1234yf is the optimal working fluid from the thermodynamic, thermoeconomic and environmental point of views for a low-grade geothermal source at an inlet temperature of 110 C. On the other hand, R600a is the best only from the exergoeconomic viewpoint. The exergy loss of the heat exchanger is the highest, followed by those of the turbine and pump. However, the capital cost of the turbine accounts for the greatest cost, followed by those of the heat exchanger and pump. For the ORC with a low GWP fluid, the CO2,eq emissions mainly come from the construction phase while the main source for the ORC with a high GWP fluid is from the operation phase. Moreover the emissions from equipment are mainly from the air-cooled condenser, followed by those from the turbine. Sensitivity analysis indicates that an appropriate extension of the lifecycle time and annual operation time would improve the thermoeconomic, exergoeconomic, and environmental performance of the ORC. The EPC and per exergy unit cost would gradually decrease as the inlet temperature of geothermal water increases. In addition, the amount of emission reduction increases. The optimal fluid also differs with the increase in the geothermal source temperature. Among the considered ORCs, the basic ORC has the optimal thermoeconomic and exergoeconomic performance, whereas the IHX-ORC is the best candidate in terms of emission reduction. The multifactor evaluation is an effective method for multiobjective analysis. R1234yf exhibits the best performance for the subcritical ORC system under the given conditions.

Appendix A The diagrams of the IHX-ORC system and dual-pressure ORC system are shown in Fig. 3.A1A and B, respectively. In the IHX-ORC (Fig. 3.A1A), a component IHX is added between the pump and evaporator (turbine and condenser). After the organic fluid is pressurized by the pump, it passes to the

Organic Rankine cycle driven by geothermal heat source Chapter | 3

119

FIGURE 3.A1 Schematic diagrams for (A) IHX-ORC and (B) dual-pressure ORC.

FIGURE 3.A2 T-s diagrams of (A) IHX-ORC and (B) dual-pressure ORC.

IHX and absorbs heat from the expanded vapour at the expander outlet. Fig. 3.A2A shows the T-s diagram of the IHX-ORC. In the dual-pressure ORC (Fig. 3.A1B), the system mainly consists of two evaporators, two turbines with generator, a condenser, two pumps and a preheater. The working fluid is pressurized by pump-L and flows into the preheater-L to absorb heat from the geothermal fluid. Then the working fluid is divided into two parts: a portion is pumped again and flows into preheater-H and evaporator-H, and another part is still heated by the geothermal fluid in evaporator-L. After passing through expander-H, the expanded vapour is mixed with the saturated fluid from evaporator-L and then flows into the condenser being cooled by the cooling source. The T-s diagram of the dual-pressure ORC is illustrated in Fig. 3.A2B. The energy balance and exergy for each component in the IHXORC and dual-pressure ORC are presented in Table 3.A1. Moreover the exergy cost rate balances and auxiliary equations for each component in the IHX-ORC and dual-pressure ORC are given in Table 3.A2.

TABLE 3.A1 Models of energy balance and exergy for each component in different ORC systems. Cycle type

Component

Energy balance model

Exergy of fuel (EF,k)

Exergy of product (EP,k)

IHX-ORC

Preheater

Qevap 5 mwf ðh5 2 h4a Þ 5 mhs ðhhs;pp 2 hhs;out Þ

E_ hs;pp 2 E_ hs;out

E_ 5 2 E_ 4a

Evaporator

Qevap 5 mwf ðh1 2 h5 Þ 5 mhs ðhhs;in 2 hhs;pp Þ

E_ hs;in 2 E_ hs;pp

E_ 1 2 E_ 5

Turbine

Wtur 5 mwf ðh1 2 h2 Þ ηtur 5 ðh1 2 h2 Þ=ðh1 2 h2s Þ

E_ 1 2 E_ 2

_ tur W

IHX

QIHX 5 mwf ðh2 2 h2a Þ 5 mwf ðh4a 2 h4 Þ

E_ 2 2 E_ 2a

E_ 4a 2 E_ 4

Condenser

Qcond 5 mwf ðh2 2 h3 Þ 5 mcs ðhcs;out 2 hcs;in Þ

E_ 2a 2 E_ 3

E_ cs;out 2 E_ cs;in

Pump

Wpump 5 mwf ðh4 2 h3 Þ ηpump 5 ðh4s 2 h3 Þ=ðh4 2 h3 Þ

_ pump W

E_ 4 2 E_ 3

Preheater-L

Qpre;L 5 ðmwf;L 1 mwf;H Þðh7 2 h6 Þ 5 mhs ðhhs;4 2 hhs;5 Þ

E_ hs;4 2 E_ hs;5

E_ 7 2 E_ 6

Evaporator-L

Qevap;L 5 mwf;L ðh8 2 h7L Þ 5 mhs ðhhs;3 2 hhs;4 Þ

E_ hs;1 2 E_ hs;2

E_ 8 2 E_ 7L

Preheater-H

Qpre;H 5 mwf;H ðh10 2 h9 Þ 5 mhs ðhhs;2 2 hhs;3 Þ

E_ hs;2 2 E_ hs;3

E_ 10 2 E_ 9

Evaporator-H

Qevap;H 5 mwf;H ðh1 2 h10 Þ 5 mhs ðhhs;1 2 hhs;2 Þ

E_ hs;1 2 E_ hs;2

E_ 1 2 E_ 10

Turbine-H

Wtur;H 5 mwf;H ðh1 2 h2 Þ ηtur 5 ðh1 2 h2 Þ=ðh1 2 h2s Þ

E_ 1 2 E_ 2

_ tur1 W

Turbine-L

Wtur;L 5 ðmwf;L 1 mwf;H Þðh3 2 h4 Þ ηtur 5 ðh3 2 h4 Þ=ðh3 2 h4s Þ

E_ 3 2 E_ 4

_ tur2 W

Condenser

Qcond 5 ðmwf;L 1 mwf;H Þðh4 2 h5 Þ 5 mcs ðhcs;out 2 hcs;in Þ

E_ 4 2 E_ 5

E_ cs;out 2 E_ cs;in

Pump-L

Wpump;L 5 ðmwf;L 1 mwf;H Þðh6 2 h5 Þ ηpump 5 ðh6s 2 h5 Þ=ðh6 2 h5 Þ

_ pump;L W

E_ 6 2 E_ 5

Pump-H

Wpump;H 5 mwf;H ðh9 2 h7H Þ ηpump 5 ðh9s 2 h7H Þ=ðh9 2 h7H Þ

_ pump;H W

E_ 9 2 E_ 7H

Dual-pressure ORC

IHX-ORC, ORC with an internal heat exchanger.

TABLE 3.A2 Exergy cost rate balances and auxiliary equations for each component in different ORC systems. Cycle type ORC with an internal heat exchanger (IHX)-ORC

Dual-pressure ORC

Component

Energy balance model

Investment model

Preheater

C_ 4a 1 C_ hs; pp 1 Z_ pre 5 C_ 5 1 C_ hs;out

chs;pp 5 chs;out

Evaporator

C_ 5 1 C_ hs;in 1 Z_ evap 5 C_ 1 1 C_ hs;pp

chs;in 5 chs;pp

Turbine

C_ 1 1 Z_ tur 5 C_ 2 1 C_ Wtur

c1 5 c2

IHX

C_ 2 1 C_ 4 1 Z_ IHX 5 C_ 2a 1 C_ 4a

c2 5 c2a

Condenser

C_ 2a 1 C_ cs;in 1 Z_ cond 5 C_ 3 1 C_ cs;out

c2a 5 c3 ; ccs;in 5 0

Pump

C_ 3 1 C_ Wpump 1 Z_ pump 5 C_ 4

cWpump 5 cWtur

Preheater-L

C_ 6 1 C_ hs;4 1 Z_ pre;L 5 C_ 7 1 C_ hs;5

chs;4 5 chs;5

Evaporator-L

C_ 7L 1 C_ hs;3 1 Z_ evap;L 5 C_ 8 1 C_ hs;4

chs;3 5 chs;4

Preheater-H

C_ 9 1 C_ hs;2 1 Z_ pre;H 5 C_ 10 1 C_ hs;3

chs;2 5 chs;3

Evaporator-H

C_ 10 1 C_ hs;1 1 Z_ evap;H 5 C_ 1 1 C_ hs;2

chs;1 5 chs;2

Turbine-H

C_ 1 1 Z_ tur;H 5 C_ 2 1 C_ Wtur;H

c1 5 c2

Turbine-L

C_ 3 1 Z_ tur;L 5 C_ 4 1 C_ Wtur;L

c3 5 c4

Condenser

C_ 4 1 C_ cs;in 1 Z_ cond 5 C_ 5 1 C_ cs;out

c4 5 c5 ,ccs;in 5 0

Pump-L

C_ 5 1 C_ Wpump;L 1 Z_ pump;L 5 C_ 6

cWpump;L 5 cWtur;L

Pump-H

C_ 7H 1 C_ Wpump;H 1 Z_ pump;H 5 C_ 9

cWpump;H 5 cWtur;H

122

PART | I Technologies

References [1] Rahbar K, Mahmoud S, Al-Dadah RK, Moazami N, Mirhadizadeh SA. Review of organic Rankine cycle for small-scale applications. Energy Convers Manag 2017;134:135
55. [2] Marty F, Serra S, Sochard S, Reneaume J-M. Simultaneous optimization of the district heating network topology and the organic Rankine cycle sizing of a geothermal plant. Energy 2018;159:1060
74. [3] Imran M, Usman M, Park B-S, Yang Y. Comparative assessment of organic Rankine cycle integration for low temperature geothermal heat source applications. Energy 2016;102:473
90. [4] Liu Q, Shen A, Duan Y. Parametric optimization and performance analyses of geothermal organic Rankine cycles using R600a/R601a mixtures as working fluids. Appl Energy 2015;148:410
20. [5] Walraven D, Laenen B, D’Haeseleer W. Economic system optimization of air-cooled organic Rankine cycles powered by low-temperature geothermal heat sources. Energy 2015;80:104
13. [6] Zhang C, Liu C, Wang S, Xu X, Li Q. Thermo-economic comparison of subcritical organic Rankine cycle based on different heat exchanger configurations. Energy 2017;123:728
41. [7] Quoilin S, Declaye S, Tchanche BF, Lemort V. Thermo-economic optimization of waste heat recovery organic Rankine cycles. Appl Therm Eng 2011;31(14
15):2885
93. [8] Shokati N, Ranjbar F, Yari M. Exergoeconomic analysis and optimization of basic, dualpressure and dual-fluid ORCs and Kalina geothermal power plants: a comparative study. Renew Energy 2015;83:527
42. [9] Fergani Z, Touil D, Morosuk T. Multi-criteria exergy based optimization of an organic Rankine cycle for waste heat recovery in the cement industry. Energy Convers Manag 2016;112:81
90. [10] Ashouri M, Ahmadi MH, Pourkiaei SM, Astaraei FR, Ghasempour R, Ming T, et al. Exergy and exergo-economic analysis and optimization of a solar double pressure organic Rankine cycle. Therm Sci Eng Prog 2018;6:72
86. [11] Liu C, He C, Gao H, Xie H, Li Y, Wu S, et al. The environmental impact of organic Rankine cycle for waste heat recovery through life-cycle assessment. Energy 2013;56:144
54. [12] Heberle F, Schifflechner C, Bru¨ggemann D. Life cycle assessment of organic Rankine cycles for geothermal power generation considering low-GWP working fluids. Geothermics 2016;64:392
400. [13] Nami H, Mahmoudi SMS, Nemati A. Exergy, economic and environmental impact assessment and optimization of a novel cogeneration system including a gas turbine, a supercritical CO 2 and an organic Rankine cycle (GT-HRSG/SCO 2). Appl Therm Eng 2017;110:1315
30. [14] Zhang C, Liu C, Xu X, Li Q, Wang S. Energetic, exergetic, economic and environmental (4E) analysis and multi-factor evaluation method of low GWP fluids in trans-critical organic Rankine cycles. Energy 2019;168:332
45. [15] Liu Q, Duan Y, Yang Z. Performance analyses of geothermal organic Rankine cycles with selected hydrocarbon working fluids. Energy 2013;63:123
32. [16] Hajabdollahi H, Ganjehkaviri A, Mohd Jaafar MN. Thermo-economic optimization of RSORC (regenerative solar organic Rankine cycle) considering hourly analysis. Energy 2015;87:369
80.

Organic Rankine cycle driven by geothermal heat source Chapter | 3

123

[17] Roy JP, Mishra MK, Misra A. Performance analysis of an Organic Rankine Cycle with superheating under different heat source temperature conditions. Appl Energy 2011;88 (9):2995
3004. [18] Tian H, Liu L, Shu G, Wei H, Liang X. Theoretical research on working fluid selection for a high-temperature regenerative transcritical dual-loop engine organic Rankine cycle. Energy Convers Manag 2014;86:764
73. [19] Shu G, Gao Y, Tian H, Wei H, Liang X. Study of mixtures based on hydrocarbons used in ORC (Organic Rankine Cycle) for engine waste heat recovery. Energy. 2014;74:428
38. [20] Franco A, Villani M. Optimal design of binary cycle power plants for water-dominated, medium-temperature geothermal fields. Geothermics 2009;38(4):379
91. [21] Guzovi´c Z, Raˇskovi´c P, Blatari´c Z. The comparision of a basic and a dual-pressure ORC (organic Rankine cycle): geothermal power plant Velika Ciglena case study. Energy 2014;76:175
86. [22] Zare V. A comparative exergoeconomic analysis of different ORC configurations for binary geothermal power plants. Energy Convers Manag 2015;105:127
38. [23] Li G. Organic Rankine cycle performance evaluation and thermoeconomic assessment with various applications part II: economic assessment aspect. Renew Sustain Energy Rev 2016;64:490
505. [24] Zhang C, Liu C, Xu X, Li Q, Wang S, Chen X. Effects of superheat and internal heat exchanger on thermo-economic performance of organic Rankine cycle based on fluid type and heat sources. Energy 2018;159:482
95. [25] Wang S, Liu C, Zhang C, Xu X, Li Q. Thermo-economic evaluations of dual pressure organic Rankine cycle (DPORC) driven by geothermal heat source. J Renew Sustain Energy 2018;10(6):063901. [26] Toffolo A, Lazzaretto A, Manente G, Paci M. A multi-criteria approach for the optimal selection of working fluid and design parameters in organic Rankine cycle systems. Appl Energy 2014;121:219
32. [27] Forooghi P, Hooman K. Experimental analysis of heat transfer of supercritical fluids in plate heat exchangers. Int J Heat Mass Transf 2014;74:448
59. [28] Garc´ıa-Cascales JR, Vera-Garc´ıa F, Corber´an-Salvador JM, Gonz´alvez-Maci´a J. Assessment of boiling and condensation heat transfer correlations in the modelling of plate heat exchangers. Int J Refrig 2007;30(6):1029
41. [29] Tian H, Chang L, Gao Y, Shu G, Zhao M, Yan N. Thermo-economic analysis of zeotropic mixtures based on siloxanes for engine waste heat recovery using a dual-loop organic Rankine cycle (DORC). Energy Convers Manag 2017;136:11
26. [30] Life Cycle Climate Performance Working Group. Guideline for life cycle climate performance. Paris, France: International Institute of Refrigeration; 2015. [31] Wang S, Liu C, Ren J, Liu L, Li Q, Huo E. Carbon footprint analysis of organic Rankine cycle system using zeotropic mixtures considering leak of fluid. J Clean Prod 2019;239:118095. [32] Sullivan JL, Clark C, Han J, Harto C, Wang M. Cumulative energy, emissions, and water consumption for geothermal electric power production. J Renew Sustain Energy 2013;5 (2):023127. [33] Wang S, Liu C, Liu L, Xu X, Zhang C. Ecological cumulative exergy consumption analysis of organic Rankine cycle for waste heat power generation. J Clean Prod 2019;218:543
54.

124

PART | I Technologies

[34] Aprea C, Greco A, Maiorino A. HFOs and their binary mixtures with HFC134a working as drop-in refrigerant in a household refrigerator: energy analysis and environmental impact assessment. Appl Therm Eng 2018;141:226
33. [35] Weckert M, Restrepo G, Gerstmann S, Frank H. Ranking of refrigerants by different assessment methods. Environ Inform Ind Ecol 2008;66
74. [36] Zhai H, An Q, Shi L. Analysis of the quantitative correlation between the heat source temperature and the critical temperature of the optimal pure working fluid for subcritical organic Rankine cycles. Appl Therm Eng 2016;99:383
91.

Chapter 4

Renewable energy based trigeneration systems— technologies, challenges and opportunities Deepesh Sonar Department of Mechanical Engineering, Government Polytechnic College, Ujjain, India

Chapter Outline 4.1 Introduction 126 4.2 Cogeneration and trigeneration 127 4.2.1 Trigeneration systems classification 128 4.2.2 Microgeneration 130 4.2.3 Polygeneration 130 4.2.4 Distributed/decentralized energy system 130 4.2.5 District energy systems and polygeneration microgrids 132 4.2.6 Combined cooling, heating and power operation strategies (modes) 133 4.2.7 Energy tools/software used in energy systems 133 4.3 Heat-recovery units 134 4.3.1 Types of heat-recovery units 134 4.3.2 Heat pumps 135 4.4 Cooling technologies 135 4.4.1 Types of cooling technologies 135 4.4.2 Cooling applications in trigeneration systems 135 4.5 Thermal energy storage 136 4.5.1 Storage concept 136

4.5.2 Storage mechanisms/types of thermal energy storage 136 4.5.3 Combined heat storage 137 4.5.4 Packed bed systems 137 4.5.5 Solar thermal energy storage 138 4.6 Renewable energy 138 4.6.1 Hybrid energy systems 139 4.6.2 Wind energy 139 4.6.3 Geothermal energy technologies 141 4.6.4 Biomass energy 142 4.6.5 Solar energy 144 4.6.6 Other renewable sources 148 4.7 Research trends in renewable energy integrated trigeneration technologies 149 4.8 Challenges and opportunities in renewable energy-based trigeneration systems 154 4.8.1 Challenges and barriers 154 4.8.2 Opportunities and prospects 156 4.9 Conclusions 158 Abbreviations 160 References 161 Further reading 168

Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00004-2 © 2021 Elsevier Inc. All rights reserved.

125

126

PART | I Technologies

4.1

Introduction

Cogeneration also called combined heat and power (CHP) systems simultaneously produce power and useful heat. Trigeneration goes a step ahead and the waste heat generated by the energy system is further utilized to produce cooling, so this system is also called as combined cooling, heating and power (CCHP) system. Polygeneration or multigeneration is the system integration for delivering multiple utilities from a single fuel source. Cogeneration (CHP), trigeneration (CCHP) or polygeneration technologies are applied in sectors such as industry, building and agriculture [1]. These energy systems can provide substantial improvements in energy efficiency with fuel savings. CHP/CCHP use can also lower emissions of greenhouse gases (especially CO2) and pollutants (especially SO2 and NOx) [2,3]. Thus these systems are more efficient, less polluting and more economical than conventional systems [4]. According to International Energy Agency (IEA), the building sector is the largest energy consuming sector in the world, accounting for about onethird of final energy use, but equally important source of CO2 emissions [5]. Buildings sector
residential, public or commercial, represents 32% of the total final energy consumption. In terms of primary energy consumption, they represent around 40% in most IEA countries [5,6]. It is important to integrate solutions for heat, cold, electricity and fuel production into buildings at the beginning of the planning process of construction. Buildings in future energy systems should be energy positive, so that they can generate more energy than they consume within a certain timeframe, usually over 1 year [7]. In buildings, the current trends refer to reducing specific consumption (kWh/m2) and increasing local energy production (in particular from renewable sources) [8]. Due to the growing need for extracting energy from renewable sources, which are generally intermittent in nature, much attention has also been directed towards energy storage in the last few years. Also, in residential building applications, the demand for heat and electricity is not synchronized. Therefore thermal energy storage (TES) is a demand management technique to bridge the gap between energy supply and energy demand, thus improving the energy efficiency of thermal energy systems. In future, cooling demand and hence energy consumption for climate control and refrigeration, will continue to grow. Moreover, common refrigerants in compression chillers, that is chlorofluorocarbons (CFCs), hydrochlorofluorocarbons (HCFCs) and hydrofluorocarbons (HFCs), are contributors to global warming, and have ozone-depleting potential as well. Alternatives for vapour compression cooling are already in use or are under development. Promising technologies include thermal energy driven cooling systems as used by trigeneration or CCHP systems.

Renewable energy based trigeneration systems—technologies Chapter | 4

127

The fast development of thermally activated cooling technologies together with reduction of their market price, and the commercial success of CHP/CCHP technologies in the past, have contributed to strengthen and spread the on-site application of CCHP technology. These types of cooling systems bring about a lower primary energy usage and lower CO2 emissions per unit cold produced. Renewable energy refers to energy produced from a natural resource having the characteristics of inexhaustibility over time. Renewable energy resources can be integrated into energy systems by considering economic, technical, social or operational objectives and different criteria sets [9]. Renewable energy sources include hydropower, wind, biomass, geothermal, tidal, wave and solar energy sources [10]. Amongst renewable technologies, solar and wind systems are the most favourable and mature ones. In Europe, for example, wind farms are dominating the north whilst solar systems (both thermal and electrical systems) are much penetrating in the energy systems of the south [11,12]. Unfortunately, renewable energy sources are intermittent in nature and there is a mismatch between the supply and demand [12,13]. Although fossil fuels remain the main resource used for co- and trigeneration purposes, renewable sources such as biogas, biomass and solar energies can also be used. The controllable sources of supply like biogas and biomass driven energy systems, as well as waste-incineration plants will continue as undeniable parts of energy systems with high penetration of renewables [1,12].

4.2

Cogeneration and trigeneration

In ‘separate production’ (SP), the electricity is delivered from the local electrical grid, the cooling from electrical chiller and the gas boiler provides thermal energy [14,15]. Cogeneration or CHP production is the simultaneous generation of electricity and useful heat, from a single fuel source [6,16]. Trigeneration or CCHP refers to the simultaneous generation of electricity, useful heating and cooling from a single fuel source [3,6], [6,16]. The level of specific emissions (i.e. emissions per unit of useful energy produced) from co- and trigeneration systems is lower than those with conventional systems [16]. In addition, trigeneration systems attain higher overall efficiencies than separate production, or cogeneration [6]. A typical trigeneration (CCHP) system consists of five main components: the prime mover, electricity generator, heat-recovery system, thermally activated equipment and the management and control system [4,17
19]. Trigeneration systems vary from site to site, with diverse prime movers, cooling options, connecting forms, rated size ranges, heat-to-power rates, user demand limitations and similar characteristics [17].

128

PART | I Technologies

Fig. 4.1 shows the comparison of typical (1) SP of energy; (2) cogeneration and (3) trigeneration. In SP, the total energy input for electricity, heat production and cold production is 150 units. It can be seen that the losses have been minimized in cogeneration and trigeneration. In trigeneration, the same outputs are produced at reduced input of only 100 units [16,17,20].

4.2.1

Trigeneration systems classification

4.2.1.1 Classification by size Trigeneration applications are categorized into micro, small-scale, medium and large-scale systems, whilst the size range of these categories are under 20 kW, from 20 to 1 MW, from 1 to 10 MW and above 10 MW, respectively [17]. The capacity of distributed CCHP systems ranges from less than 1 kW in domestic dwellings to more than 10 MW in hospitals or university campuses, and as much as 300 MW to supply energy to a district of a city [17,21,22]. 4.2.1.2 Classification by applications According to applications, the CCHP systems are classified broadly as: 1. Traditional large-scale predominantly CHP systems in centralized power plants or large industries; 2. Relatively small/micro capacity CCHP units in commercial, institutional, residential and small industrial sections [17,23].

4.2.1.3 Classification by type of prime-mover Prime-movers based on thermodynamic cycles concerns both internal and external combustion technologies. In micro and small-scale, residential, commercial and institutional applications, the main devices currently used are Internal Combustion Engines (ICEs) and microturbines [1]. Traditional large-scale centralized power plants or large industries have configurations based on either steam turbine, or combustion turbines [1,17]. Systems not based on thermodynamic cycles are under development, and mainly concerns stirling engines, fuel cells, organic rankine cycle (ORC) and renewable energy systems, such as biomass, solar and wind, with potential to achieve high efficiency and low emission levels [1,4,16]. Reciprocating ICEs, steam turbines and combustion turbines still make up most of the gross capacity being installed [17]. In addition, fuel cells, stirling engines and microturbines, mainly gas driven, present a promising future for prime movers [17].

FIGURE 4.1 Typical energy flow diagram in (A) separate production, (B) cogeneration and (C) trigeneration. Source: Adapted and modified from Onovwionaa HI, Ugursalb VI. Residential cogeneration systems: review of the current technology. Renew Sustain Energ Rev 2006;10:389
431; Wu DW, Wang RZ. Combined cooling, heating and power: a review. Progr Energ Combust Sci 2006;32:459
495; Small-scale cogeneration, why? In which case? A guide for decision makers. European Commission, Directorate General for Energy DGXVII; July; 1999 [16, 17, 20].

130

PART | I Technologies

4.2.1.4 Classification by sequence of energy These can be classified as either a topping or a bottoming cycle systems [6,24]. In a topping cycle, the supplied fuel is first used to produce electricity and then thermal energy which is the by-product of the cycle, is recovered. It is widely used method of co and trigeneration [6,25]. In a bottoming cycle, first high-temperature thermal energy is produced by the combustion of fuel. Then, the heat rejected from the process is recovered to generate electricity using a turbine [6]. 4.2.2

Microgeneration

A microCCHP in the residential sector may be composed of a CHP prime mover, an auxiliary boiler and a thermal storage unit [1]. MicroCHPs are especially interesting for market of small family houses and medium family houses, smaller buildings such as hospitals, schools, industrial premises, offices and small- and medium-scale enterprises because of their technical and performance features [26,27].

4.2.3

Polygeneration

In polygeneration or multigeneration systems, in addition to the generation of power, cooling and heating, a few more energetic processes, such as production of chemicals, hydrogen, ethanol, biodiesel, fertilizers, and drinking water, are possible [4,15,28,29]. Polygeneration system can be fuelled by renewable sources (geothermal, solar, biomass, wind and hydro), and fossil fuels (natural gas, coal, hydrogen, etc.) or by hybrid technologies [29]. Cogeneration, trigeneration and polygeneration systems are classified according to size, applications, types of prime movers, energy sequence, etc. A detailed classification diagram showing main system components of such efficient energy generation systems is given in Fig. 4.2.

4.2.4

Distributed/decentralized energy system

A distributed/decentralized energy resource is a novel technical concept in energy supply, where an electricity-generation system located near user facilities provides electrical and thermal energy simultaneously to industrial, commercial and residential users connected along a district energy network [17,30,31]. In distributed generation (DG) system, the synergy of cogeneration, trigeneration and energy storage is enhanced. Moreover, it can stimulate the diffusion of renewable energy technologies through integration [32].

FIGURE 4.2 Classification diagram of the main system components of co-, tri- and polygeneration systems. Source: Adapted and modified from Moussawi HA, Fardoun F, Gualous HL. Review of tri-generation technologies: design evaluation, optimization, decision-making, and selection approach. Energ Convers Manag 2016;120:157
96 [6].

132

4.2.5

PART | I Technologies

District energy systems and polygeneration microgrids

District heating (or cooling), in contrast to individual system, refers to the centralized production and network-based distribution of thermal energy, and constitutes a major area for CHP/CCHP applications. They use a network of underground pipes to pump steam/hot water/chilled water to multiple buildings in an area. District energy systems (DESs) provide an opportunity to exploit additional energy sources that are better suited for combustion in large-scale systems and that would otherwise remain unused (e.g. municipal solid waste) [3]. Microgrids are small DESs integrating different DG for satisfying local energy demands by utilizing local energy sources such as natural gas, biomass, solar and wind [5]. Low-temperature heat from renewable energy sources such as solar and geothermal energy, as well as industrial waste heat from industries and power plants can be used in ‘low energy’ buildings in conjunction with a DE network in district heating [30]. Fig. 4.3 depicts various distributed CCHP/polygeneration microgrids using prime-mover technology and renewable energy systems, whilst utilizing steam/hot water/chilled water through distributed network from the central facility, as well as waste heat from industries or power plants.

FIGURE 4.3 Distributed energy systems and typical polygeneration microgrids. Source: Adapted and modified from Wu DW, Wang RZ. Combined cooling, heating and power: a review. Progr Energ Combust Sci 2006;32:459
95 [17].

Renewable energy based trigeneration systems—technologies Chapter | 4

133

4.2.6 Combined cooling, heating and power operation strategies (modes) Two classical operation strategies for the CHP/CCHP systems are, following electrical load (FEL) and following thermal load (FTL), which can also be referred to as the electrical demand management (EDM) and the thermal demand management (TDM) [23]. In FEL (or, EDM), the generated electricity at any moment is equal to the electrical load. If the cogeneration heat is lower than the thermal load, the supplementary heat from auxiliary boiler is required; if it is higher, excess heat is rejected to the environment through coolers or exhaust gases [14,23]. In FTL (or, TDM), the useful thermal output at any moment is equal to the thermal load. If the generated electricity is higher than the load, excess electricity is sold to the grid or stored; if it is lower, supplementary electricity is purchased from the grid [14,23,33].

4.2.7

Energy tools/software used in energy systems

The software tools are useful in planning and implementation of DG systems of a district/city. These are used to analyse energetic, economic and environmental performances of energy generation systems, equipments and buildings in a community [34]. These are also used to analyse the integration of renewable energy into various energy-systems. Seven different tool types, which can be used exclusively or collectively have been defined [35]: 1. A simulation tool simulates the operation of a given energy system. 2. A scenario tool usually combines a series of years into a long-term scenario, typically 20
50 years. 3. An equilibrium tool seeks to explain the behaviour of supply, demand and prices in a whole economy or part of an economy with many markets. 4. A top
down tool is a macroeconomic tool to determine growth in energy prices and demands. 5. A bottom
up tool identifies the specific energy technologies and hence investment options and alternatives. 6. Operation optimization tools optimise the operation of a given energy system. 7. Investment optimization tools optimise the investments in an energy system. In CCHP systems, various decision-making tools may be selected using data gathered from evaluation process, design analysis and system

134

PART | I Technologies

optimization. These include cost
benefit analysis, elementary methods, multicriteria decision-making, sensitivity and risk analysis, etc. [6]. Some examples of software tools are: ARCGIS, Raster Cities, HOMER, RETScreen, LEAP, Energy PLAN, Energy Plus, Hybrid2, TRNSYS, PLACE3S, Life Cycle Cost Program, etc. [34,36].

4.3

Heat-recovery units

Heat-recovery equipments, mainly composed of heat exchangers are used in CCHP systems to capture thermal energy rejected from exhaust gas streams, liquid coolant circuits, etc. [6]. Heat exchanger contains the heat transfer area, which is in direct contact with fluids and through which heat energy is transferred. Area and flow arrangements have direct impact on heat exchanger effectiveness and heat transfer efficiency. Performance and efficiency of heat or energy recovery system are also related to temperature and relative humidity, effects of airflows, pressure drop and fan power [37].

4.3.1

Types of heat-recovery units

Two general types of heat-recovery units can be distinguished; simple unfired units whose role is to exchange heat without any additional generation, where more complex are those having supplementary firing by combustion of additional fuel [6].

4.3.1.1 Unfired units Unfired heat-recovery units are used in CCHP applications to extract waste heat rejected from reciprocating engines, gas turbines, microturbines and fuel cells. These units include heat exchangers, unfired heat-recovery steam generators (HRSGs), mufflers, regenerators and recuperators. Such units are not a source of air pollutants since they do not burn fuel to produce heat. Besides, many other heat exchanger types can be used in the heat-recovery process. Those can be shell and tube, plate, fins or even microchanneled heat exchangers [6,38]. 4.3.1.2 Fired units Fired heat-recovery equipments used in CCHP applications extract heat from a heat source and combust fuel to produce complementary thermal energy for a required process, for example, fired HRSGs and boilers. Such units can be source of air pollutants as a result of the fuel combustion process depending on the used fuel type [6].

Renewable energy based trigeneration systems—technologies Chapter | 4

4.3.2

135

Heat pumps

Heat pumps are a very well known solution for heating and cooling purposes. They are composed of two heat exchangers: in winter, the heat exchanger located outdoors will absorb heat from the environmental air, transferring it to the indoor exchanger to heat the indoor environment; and, in summer, the role of each part is inverted [39].

4.4 4.4.1

Cooling technologies Types of cooling technologies

Heat-driven cooling technologies include mainly closed cycles (absorption and adsorption) and open cycles (desiccant systems). The waste heat, in the form of steam, hot water or exhaust gases, generated by prime movers can be used to produce a cooling rather than a heating effect, or to produce both [6,40].

4.4.1.1 Sorption technology Sorption technology is utilized in thermal refrigeration techniques; the cooling effect is obtained from the chemical or physical changes between the sorbent and the refrigerant [10,41]. Absorption is the process in which a substance in one phase is incorporated into another substance of a different phase (e.g. gases being absorbed by a liquid). Adsorption refers to the use of a solid for adhering or bonding ions and molecules of another substance onto its surface. In addition to the benefit of using recovered heat to save energy, sorption systems are environmentally benign: they do not use ozone-depleting CFCs (due to the use of water, ammonia and other natural refrigerants); they are noise and vibration free, and long lasting. But, high cost, heavy weight, large size and possible need for backup cooling systems are their limitations. [41]. 4.4.1.2 Desiccant technology Desiccant cooling works on the principle of incorporating desiccant dehumidification and the cooling unit, and its unique merit is that the sensible and latent heat can be processed separately. The cooling unit deals with the sensible load, whilst the desiccant removes the latent load [41]. 4.4.2

Cooling applications in trigeneration systems

In a CCHP system, thermally driven cooling technologies must be a good match for the prime movers, since each system has its own suitable working temperature [6,41].

136

PART | I Technologies

Whilst lithium bromide
water absorption systems have been widely adopted for air conditioning applications, the water
ammonia machines are most commonly used for refrigeration applications in CCHP systems [41].

4.5

Thermal energy storage

TES avoids the discontinuous operation of the CHP/CCHP system by correcting the mismatch between the supply and demand of energy, increases the number of operating hours, allows high operating efficiencies and allows sizing of energy system to a lower peak demand reducing thereby the plant investment costs [6,42
45]. TES can be short-term or long-term storage depending on the specific storage durations [6]. Aquifers and solar ponds are best suited for economical large-scale storage applications for centralized urban heating systems [13,46,47]. TES can be classified by storage mechanism (sensible, latent or chemical) and by storage concept (active or passive) [45,47].

4.5.1

Storage concept

4.5.1.1 Active system Active thermal storage systems are characterized by forced convection in the storage material: the storage medium itself circulates [42,45,47]. 4.5.1.2 Passive system In a passive storage system the heat transfer fluid (HTF) passes through the storage only for charging or discharging the system. The storage medium itself does not circulate. Passive systems are generally dual medium storage systems (these systems are also called regenerators). Passive storage systems are mainly solid storage systems [concrete and castable materials, and phase change material (PCM)] [45,47]. 4.5.2

Storage mechanisms/types of thermal energy storage

TES systems are classified as sensible heat storage, latent heat storage and thermochemical storage [42,44,48].

4.5.2.1 Sensible heat storage Sensible heat storage materials undergo no change in phase over the temperature range encountered in the storage process, and store thermal energy by sensible heat in solid or liquid materials [42,45]. Solid materials, like concrete and castable ceramics have low price and good thermal conductivities. Concerning liquid materials, a variety of fluids

Renewable energy based trigeneration systems—technologies Chapter | 4

137

have been tested to transport the heat, including water, air, oil, and sodium, before molten salts were selected as best [45]. Water remains the most widely used material in sensible heat storage systems with best compromise between cost, heat storage capacity, density and environmental impact [13].

4.5.2.2 Latent heat storage Latent heat storage involves the phase change (generally solid
liquid) of the storage material. It is a nearly isothermal process that can provide significantly enhanced storage quantities as compared to sensible storage systems. But it has poor heat transfer properties, leading to slow charging and discharging rates. Other disadvantages are solid deposits on the heat transfer surfaces, phase segregation, subcooling and finally cost [13,42,45,48,49]. There are mainly three types of PCMs suitable for different thermal applications such as organic, inorganic and eutectic mixture of PCMs [50]. 4.5.2.3 Chemical storage The need for long-term, cross-seasonal storage is made possible by thermochemical storage processes. Potentially high energy densities can be stored using chemical storage [48]. Reformation of methane and CO2, metal
oxide/metal conversions and ammonia synthesis/dissociation are a few examples of such heat-assisted chemical reactions that can store thermal energy in their endothermic reaction products and release it at a later time/place by the reverse process, which is exothermic reaction [44,48]. The chemical storage technology is promising, but is in developing phase [45]. 4.5.3

Combined heat storage

Combined storage system is best suited for intermittent applications, and eliminates the difficulties experienced in the sensible heat storage and latent heat storage systems to some extent and possesses the advantages of both the systems [51]. An example is encapsulating PCMs to increase their heat exchange area and circulating a heat transport fluid through a bed of these capsules. This results in combined heat storage [47].

4.5.4

Packed bed systems

In a packed bed system, the bed consists of storage materials such as rocks, ores or encapsulated PCMs, a container and the flow of HTF through the voids in the bed. Since the HTF is in direct contact with the storage material,

138

PART | I Technologies

heat transfer coefficients can be large. Most packed bed systems are single tank systems that act as a thermocline [42].

4.5.5

Solar thermal energy storage

Molten salts are mostly used in solar power tower systems. These have desired characteristics like high density, low vapour pressure, moderate specific heat, low chemical reactivity, and low cost, and are nonflammable and nontoxic. One of the disadvantages of molten salts is their cost [45,52,53]. Research on heat transfer enhancement techniques using fins and encapsulations, utilizing nano-based technologies, with low cost, higher temperatures and higher exergetic efficiencies in the design of storage systems is underway [13,42]. It is seen that PCMs impregnated or encapsulated with nanoparticles (1
100 nm) exhibit much improved thermo-physical properties than in their pure state [6]. Thus it can be concluded that various types of energy storage system technologies could be the best technological advancement so far for energy conversion and sustainability [50,54,55].

4.6

Renewable energy

Renewable energy is the energy obtained from energy sources which are naturally and constantly replenished, and are virtually inexhaustible in human time-scale. Renewable energy sources include hydropower, wind, biomass, geothermal, tidal, wave and solar energy sources [10]. Renewable energy systems can address a variety of sustainability related issues, like reliable energy supply, increase in energy security, energy equity problems, poverty reduction, clean water protection, development of transport, agriculture, infrastructure, and industry, new employment opportunities, improvement in human health, environment protection, and reduction of green house gas (GHG) and other harmful emissions [9,56,57]. Renewable energy systems are environmentally friendly and produce much less emissions than conventional energy resources, but they still pose major issues, like high costs, differences with the existing energy structure, and low efficiencies [57]. As regards environment and ecology considerations, hydropower conversion technologies have negative impact on fish and wildlife, along with possibility of floods. Geothermal energy produces virtually no emissions, but can have some local effects like land erosion. Wind turbines produce noise that disrupts the wildlife habitats, and collision mortality rates of birds and bats are very high. Crops for biomass production can alter the land and wildlife ecology. Also, incineration of biomass can produce some harmful emissions and toxic substances. On the other hand, solar energy has a very low impact on environment and ecology [58
63].

Renewable energy based trigeneration systems—technologies Chapter | 4

4.6.1

139

Hybrid energy systems

Hybrid energy systems use two or more different kinds of renewable energy sources together and provide energy supply in a more economic, environment friendly and reliable manner [9,64]. They take advantage of the complementarity of the energy subsystems to increase the overall system efficiency, and reduce capital and/or operating costs [65]. But, hybrid systems require optimal sizing and advanced control tools [66]. In majority of the cases, the solar energy is hybridized with other forms of renewable energy like the biomass energy and the fuel cells to produce utility outputs [58,67]. Renewable and fossil fuel hybrid energy systems can improve the overall efficiency and provide a more reliable supply of electricity [66].

4.6.1.1 Zero energy building Promotion of high energy efficiency and sustainability in building sectors leads to the concept of zero energy buildings (ZEBs). Total annual energy consumption of ZEB is not larger than the on-site exported RES on the primary energy source basis. Integrating polygeneration with renewable energy sources is promising to implement ZEB balance [5]. 4.6.2

Wind energy

Amongst the renewable alternatives, wind energy is becoming popular for a variety of functions ranging from electricity generation to pumping water, sailing boats and driving pumps [65,68]. Electricity generation from wind turbine depends on factors such as the wind speed, wind turbine availability, swept area of the rotating turbine blade and arrangement of the turbines [65,69].

4.6.2.1 Wind power meteorology and wind modelling Wind shows great temporal and spatial variability. Spatial variations of wind and other meteorological parameters can be expressed in the form of maps. European wind atlas, ‘Wind Atlas Analysis and Application Program’ (WASP) is the most complete mapping methodology in wind-power meteorology searches. It is used for primary siting of wind turbines singly or in farms [68,70]. Wind turbines produce electricity between 3 and 25 m/s, and high productions are evaluated after 10
15 m/s values. Wind turbines range in capacity from several kilowatts to Megawatts. The crucial parameter is the diameter of the turbine and the longer the blades, the larger the areas swept by the rotor, leading to higher energy outputs [68].

140

PART | I Technologies

Wind energy is largely free of environmental impacts and they do not emit CO2 [58,68].

4.6.2.2 Turbine technology The wind turbine converts wind kinetic energy into mechanical energy and the latter into electrical energy by means of an electrical generator [68]. A typical wind turbine design is made up of rotor blades, a drive shaft, a gear box, a speed shaft, a generator, and support cables and casing [65,71]. Wind turbines can be horizontal-axis or vertical-axis turbine types. Twoor three-bladed turbines are usually used for electricity generation, whereas 20 or more blades are used for water pumping. Currently three-bladed wind turbines with horizontal-axis dominate the market [58,68]. Wind farms are groups of wind turbines generating electricity on a significant scale. In rural and isolated areas, stand-alone power systems are used [68]. 4.6.2.3 Wind hybrid systems and applications 4.6.2.3.1 Wind
diesel system One of the suggested energy systems for isolated communities is a wind
diesel system. Wind turbines, if used with diesel generators, significantly reduce the cost of fuel consumption and greenhouse gases emission. The basic requirement is to achieve fuel saving and provide a reliable power supply [68]. 4.6.2.3.2 Wind
photovoltaic-hydrogen system Wind and PV systems rely on highly transient energy sources and exhibit strong short-term and seasonal variations in their energy outputs. They, thus need to store the energy produced in periods of low demand, to stabilize the output, when the demand is high [68]. 4.6.2.3.3 Seawater desalination Desalination systems driven by renewable energies are scarce and they tend to have a limited capacity. Coastal areas have a high availability of wind power resources. Therefore wind power desalination is a promising alternative at these places [68].

4.6.2.4 Wind power development Wind turbines today have matured. These operate with one fifth of the downtime; achieve about one-third more energy per unit of installed power or swept surface, and with much reduced prices. Already over 20,000 turbines are producing electricity worldwide. In the Netherlands, 90% of the power used in industry is based on wind energy [68].

Renewable energy based trigeneration systems—technologies Chapter | 4

141

The biggest producers of wind power in the world are Germany, Spain and USA [58,72]. United Kingdom uses significant amount of wind energy; for example, the large-scale wind farm in Clyde in South Lanarkshire, Scotland has 152 turbines and generates 350 MW [65,73]. Several countries, like United Kingdom, United States, India, and so on have combined market-stimulation incentives with utility and national targets (for a better environment and/or an increase in wind power capacity) resulting in a domestic market for wind energy [68]. The goal for future development of wind turbines is to increase wind capture and at the same time decrease, or eliminate loads (i.e. mechanical stress). The development of new wind turbines will also involve material development and preparation, in addition to the development of wind turbine manufacturing methods [68]. But, there are technological challenges and issues related to development of wind power. Main issue is optimization of cost-effective wind power plants for widely different and complex terrains. Second, structural aspects along with system dynamics of very large-scale wind turbine pose challenges. Grid reliability issues also remain to be addressed. Until now, benefits and cost analysis has been clearly in favour of standalone CHP power plant, rather than the standalone wind power plant. [27].

4.6.3

Geothermal energy technologies

Geothermal energy is the thermal energy generated and stored in earth. Geothermal energy is the second most abundant natural source of heat on earth after solar energy and is continuously available. Compared with other renewable energy sources, geothermal power generation is a clean and reliable source of energy with minimal environmental impacts [58]. There are three technologies that can be used for energy production: geothermal heat pumps, direct-use applications, and electric power plants. Today there are about 250 electric power plant installations operating worldwide. These plants include the following four main types: dry steam power plant, single flash steam power plant, binary cycle power plant and double flash power plant [58,60]. A ground source heat pump, or geothermal heat pump taps into heat at earth’s sub-surface to get hot water to produce heating/cooling of buildings. Ground source heat pump typically has a coefficient of performance (COP) of about 4 [30]. Geothermal wells that produce dry steam can be used directly in a turbine to generate power. Flash systems use hot water at high pressure to generate steam in a flash chamber that can drive the turbine, and are the most common type of geothermal power plants. The third type, the binary geothermal

142

PART | I Technologies

plants running on ORC, uses the low-temperature geothermal resource to vapourize an organic fluid [74
76]. The use of geothermal energy directly for district heating has increased notably; geothermal sites contribute 49% of the installed capacity of heating systems in Europe, 29% in Asia and 17% in the Americas [30,77]. Use of geothermal district heating systems has increased by 10% over the past 30 years [30,78].

4.6.4

Biomass energy

Biomass is defined as a decomposable matter derived from plants or animals. It includes different sources such as wood, herbaceous, agricultural crops or municipal organic wastes [1,79]. Biomass energy, a traditional renewable energy source, currently ranks fourth worldwide providing approximately 14% of the world’s primary energy supply. In the developing countries, biomass accounts for approximately 35% or higher of the primary energy supply [27]. The use of biomass brings economic, social and environmental benefits, and being an indigenous fuel, it provides local employment and support to the rural economy. Biomass results in no net release of carbon dioxide (CO2) if the cycle of growth and harvest is sustained [80].

4.6.4.1 Biomass energy technologies Biomass energy production and conversion methods are divided into four groups [58,62]: 1. Thermal conversion processes: includes combustion, pyrolysis and gasification. 2. Biological conversion processes: includes ethanol fermentation with yeast or bacteria, and methane gas production under anaerobic conditions, better known as an anaerobic digestion (AD). 3. Chemical conversion processes: includes esterification. 4. Nonthermal conversion processes: Refuse-derived fuel (RDF) technology and landfill gas (LFG) recovering system.

4.6.4.2 Biofuels According to IEA, biofuels are liquid and gaseous fuels (i.e. biogas) produced from biomass. Examples of conventional biofuels are straight vegetable oil, biodiesel and sugar-based ethanol. Biofuels are gaining popularity as they are immediate substitutes to the existing petro fuels [81].

Renewable energy based trigeneration systems—technologies Chapter | 4

4.6.4.2.1

143

Straight vegetable oils

Vegetable oils can be successfully used in diesel engine through minor engine modifications and fuel modifications [82,83]. The main advantages of nonedible oil are their liquid nature, portability, ready availability, renewability, higher heat content, lower sulfur content, lower aromatic content and biodegradability [84]. High viscosity is the main barrier that prevents the use of direct vegetable oils in conventional diesel engines, as it interferes with the injection process and leads to poor fuel atomization, leading to heavy smoke, lube oil dilution and high carbon deposits. These problems can be solved, if the vegetable oils are chemically modified to biodiesel, which is similar in characteristics to diesel fuel [84,85].

4.6.4.2.2

Biodiesel

Biodiesel is the mono-alkyl esters of long chain fatty acids derived from vegetable oils or animal fats by using pyrolysis, dilution, microemulsion and transesterification [84,86]. Almost all the important properties of biodiesel are in very close agreement with the mineral diesel [82]. Biodiesel is biodegradable, nonflammable, renewable and nontoxic as well as environment friendly, with minimum engine modifications requirement [84
86]. Disadvantages of biodiesel include slight loss in fuel economy and high cost of production [85]. Many researchers have reported that CO, CO2 and UBHC emissions are reduced in biodiesel and its blends. NOx emissions are slightly increased. Biodiesel is considered carbon-neutral because all the CO2 released during consumption had been sequestered from the atmosphere for the growth of vegetable oil crops [85].

4.6.4.2.3

Bioethanol

Bioethanol, a petrol additive/substitute fuel is derived typically from plants, straw, and wood, and is fermented from sugars, starches or from cellulosic biomass and is also the most widely used liquid biofuels [82,87].

4.6.4.2.4

Biomethanol

Methanol, or ‘wood alcohol’, is produced from synthetic gas or biogas and evaluated as an engine fuel. But, the production of methanol is a costintensive chemical process [87].

144

PART | I Technologies

4.6.4.2.5

Biogas

AD converts organic material directly to a gas called biogas. Methane and CO2 are the major components of biogas. Different biodegradable wastes, like pure cellulose, mixed waste feedstocks and fats/oils can yield biogas with 40%, 50% and even 70% methane content, respectively [79]. Fig. 4.4 shows various technologies for converting biomass, crops, vegetable oils and waste to various biofuels.

4.6.4.3 Biomass-fuelled combined cooling, heating and power systems Biomass-fuelled CCHP systems have the potential to solve the energy trilemma, that is security of supply, affordability of energy and environmental protection. Biomass energy has its advantage of continuity over the intermittence of solar energy and wind energy [27]. Several studies indicate significant potential for biomass-fired CCHP systems, but so far only medium-scale (1
10 MW) and large-scale ( . 10 MW) systems have been commercialized successfully, whilst microscale (,20 kW) and small-scale (20
1 MW) systems are still in an experimental phase [89].

4.6.5

Solar energy

Solar energy is inexhaustible, and represents a considerable amount of energy if it is efficiently captured and stored. Main technologies using solar as primary energy for CHP applications are solar photovoltaic systems and solar thermal electricity systems [1] Efficient design of solar power generation, conversion, storage and distribution systems are required due to sporadic nature of solar insolation, and its diurnal, seasonal, annual and topographic variations [80]. Solar thermal power cycles are classified as low (up to 100 C), medium (up to 400 C) and high (above 400 C) temperature cycles [90].

4.6.5.1 Solar collectors The basic function of a solar collector is to absorb incident solar radiation and convert it into heat, which is then carried away by a HTF, like air, water, oil, and so on flowing through the collector, to the power generation system, to a central steam generator or to a thermal storage system as it circulates [48,49,91,92]. Solar collectors are classified into two categories according to concentration ratios: nonconcentrating collectors and concentrating collectors [49,92].

FIGURE 4.4 Conversion technologies-biomass to biofuels. Source: Adapted and modified from Hamelinck C, Broek RVD, Rice B, Gilbert A, Ragwitz M, Toro F. Liquid biofuels strategy study for Ireland. A report of sustainable energy Ireland (report no. 04-RERDD-015-R-01), 2004. p. 1
105 [88].

146

PART | I Technologies

4.6.5.1.1

Nonconcentrating solar collectors

Nonconcentrating solar collectors include a flat-plate collector, in which a black flat surface collects heat, and then the energy is transferred to water, air, or other fluids for further use, like space heating and cooling, water heating, and desalination in a low/medium temperature range [92]. 4.6.5.1.2

Concentrating solar collectors

Sun-tracking, concentrating solar collectors utilize optical elements to focus large amounts of radiation onto a small receiving area and follow the sun throughout its daily course to maintain the maximum solar flux at their focus for high-temperature applications such as electricity generation [48,92]. Concentrating collectors are generally associated with higher operation temperatures and greater efficiencies. The four main types of concentrating solar collectors are [48]: (1) (2) (3) (4)

Parabolic trough collectors; Heliostat field collectors; Linear Fresnel reflectors and Parabolic dish collectors.

4.6.5.2 Solar photovoltaic systems A solar cell converts solar energy into electricity by means of the photoelectric phenomenon found in semiconductor materials such as silicon and selenium. Presently, efficiency of PV cells is about 12
19% at the most promising conditions [91,93]. Common applications for stand-alone systems are: solar cars, vans and boats; remote cabins and homes; traffic lights; solar pumps, etc. [93]. 4.6.5.3 Hybrid photovoltaic-thermal systems Concentrating PV/T systems produces both electricity and heat simultaneously and makes it more efficient and cost-effective as compared to the current PV systems [92]. In such hybrid configuration, a PV panel is also equipped with a heatrecovery system consisting in a bottom layer of the receiver tube to absorb the concentrated solar flux as heat [48,49,91]. 4.6.5.4 Solar thermal applications Solar thermal collectors have more applications than solar PV collectors in polygeneration systems. The fuels cell hybridized polygeneration generally produces electricity with cooling, heating and potable water as the utilities [28]. Typical solar thermal applications are classified as below: [48,91,93]

Renewable energy based trigeneration systems—technologies Chapter | 4

147

District/building heating and cooling using solar technologies; Solar hot water/steam production for industrial processes; Solar water heating; Solar drying and dehydration processes; Solar desalination/distillation of sea or brackish water; Solar reforming of fuel; Pasteurization, sterilization; Washing, cleaning; Chemical production; etc.

4.6.5.5 Solar-renewable hybrids Integration of renewable sources such as solar, geothermal, biogas and biomass, with microCHP systems could be an efficient way to introduce renewables in several areas of applications such as residential and industrial environments to provide a stable energy supply [25]. The hybrid solar technologies have been categorized into high, medium, and low-renewable hybrids based on their renewable energy component [74]. (a) High—hybrids of concentrating solar plant (CSP) with wind, biomass and geothermal energy resources [74,90]. (b) Medium—this category includes solar plants that use supplementary firing of fossil fuels such as natural gas to enhance the plant output and capacity factor [74,90]. (c) Low—this category represents conventional fossil fuel plants that incorporate solar energy for auxiliary functions such as preheating and evaporation. This category includes ISCC plants, solar-aided coal power plants, and solar-Brayton cycles [74,90]. The high-renewable hybrids report the least specific CO2 emissions (,100 kg/MW h), followed by medium (,200 kg/MW h) and lowrenewable hybrids ( . 200 kg/MW h). The solar share in high-renewable hybrids is 100%, followed by medium (70%) and low-renewable hybrids (,20%) [74]. 4.6.5.5.1

High-renewable hybrids

Concetrating solar plant-biomass hybrids CSP-biomass hybrids are restricted below 50 MWe to ensure a continuous supply of biomass and to reduce logistical costs. Lower capacity plants in the range of 2
10 MWe are possible in the case of trigeneration systems [74]. Concetrating solar plant-geothermal hybrids CSP-geothermal hybridization can overcome several challenges faced by standalone geothermal plants such as cost, low efficiency, and reduction in output over time [74]. The two most common integration schemes are [74,90]:

148

PART | I Technologies

(a) Preheating of the geothermal brine using solar energy, and (b) Superheating of the steam by solar energy before entering the turbine Concetrating solar plant-wind hybrids Solar energy naturally complements wind energy in generating power more uniformly as wind speed is lower during the day and summer compared to nights and winter.The excess electricity produced (at night, or in winter) by the wind farm can be utilized to charge the TES system of the CSP plant. Currently, there are no CSPwind hybrid plants operational in the world [74]. 4.6.5.5.2

Medium-renewable hybrids

Medium-renewable hybrids are the most common solar plants as they can operate continuously. Currently they are not competitive with conventional power plants or with PVs [74,90]. The capacity of such plants can range from 20 to 500 MWe and is suitable both for distributed and centralized power systems. Currently efficiency and cost are the primary constraints that prevent further penetration of these systems into the market and should be the focus of future research. Several operating solar plants are using auxiliary natural gas burners for heating purposes [74]. 4.6.5.5.3 Low renewable hybrids Solar-Brayton cycles Solar energy is used either to preheat the compressed air before entering the combustion chamber or to generate steam that is injected into the combustion chamber as the working fluid. In both the cases, the increased inlet temperature to the combustion chamber reduces the fuel consumption rate whilst increasing the cycle efficiency. There are currently no operational CSP-Brayton hybrid plants in the world [74]. Solar-aided coal power plants (Rankine cycle) These plants utilize solar energy for preheating and boiling. Both direct steam generation and heat transfer fluid technologies can be used [74,90]. Integrated solar combined cycles Integration of solar energy in topping cycle as well as the bottoming cycle, called ISCC offers higher efficiency. Several ISCC plants (bottoming cycle) are currently operational or under construction around the world [74].

4.6.6

Other renewable sources

Conventional dam and reservoir-based hydroelectricity is often viewed as a low-cost, immediately available zero-carbon resource. Within the energy

Renewable energy based trigeneration systems—technologies Chapter | 4

149

community, that could facilitate more intermittent renewable electricity integration, seasonal storage, and other grid benefits [94]. Hydroelectric power plants derive power from potential or kinetic head of flowing water. It can be used for irrigation, mills, cranes, etc., or for generation of electricity. They help expand renewable energy generation. This is particularly true for small hydroelectric plants. But, large dams are detrimental to river ecosystem, cover large area for dams, and affect thousands of people and their livelihood. The development of hydroelectric power plants depends on the economic and financial feasibility, and social factors [95]. The reported research of integration of CHP/CCHP with hydropower is scarce. Ocean energy sources have limited applicability in CCHP in the present and in the future. According to Johansson et al. [62], the four resources of ocean-energy (tidal energy, wave energy, ocean-thermal energy and saltgradient energy) are spread over a considerable geographical expanse far from centers of consumption, thus creating low-energy densities and raising the costs of collecting the energy. Currently there is renewed interest in harnessing the vast tidal resource to combat the twin challenges of climate change and energy security. However, high costs involved limits its further development [96]. Tidal energy is one of the most predictable forms of renewable energy. There has been much commercial and R&D progress in tidal stream energy, and now is a more mature technology with tidal range power plants. Two main research areas are: the present and future tidal range resource, and the optimization of tidal range power plants [97]. But, integration of tidal energy with CHP/CCHP has not been fully developed yet. Fig. 4.5 depicts typical average normalized rankings of various energy sources in terms of emissions, efficiencies, renewability, and the multigeneration capabilities, on a scale of 0
1. The ideal system is shown to be having a ranking of 1 each for multigeneration capabilities with renewability, and maximum efficiency with minimum emission.

4.7 Research trends in renewable energy integrated trigeneration technologies A household size trigeneration based on a small-scale diesel engine generator set was designed and realized in laboratory. Experimental tests were carried out to find that the total thermal efficiencies of trigeneration are upto 438% higher, and CO2 emission per unit (kW h) of trigeneration output were upto 81.4% lower compared to those of single generation [98]. Effect of varying injection opening pressure (IOP) using diesel
mahua oil (preheated and raw) blends and biodiesel as fuel on the characteristics of a small agricultural engine were reported. At optimum IOP of 226 bar, maximum brake thermal efficiency was obtained with significant reduction in carbon monoxide and

FIGURE 4.5 Typical average normalized rankings (on a scale of 0
1) of various energy sources in terms of efficiencies, emissions, multigeneration capabilities and renewability. Source: Adapted and modified from Dincer I, Acar C. Smart energy systems for a sustainable future, Appl Energy 2017;194:225
35 [57].

Renewable energy based trigeneration systems—technologies Chapter | 4

151

hydrocarbon emission were observed. Nitrogen oxides emissions increased marginally [99]. A 5 kW bioethanol system based on fluidized bed reactor and proton-exchange membrane fuel cells has also been presented [100]. Operation of natural refrigerants such as CO2 and trigeneration systems was integrated. It was shown that this system produces 30% energy savings and over 40% GHG emissions savings over conventional refrigeration in supermarkets [101]. In another work comprising of ORC-cascade refrigeration, the electricity consumption in cascade refrigeration system reduced by 61% and COP of compression section improved by 155% compared to the conventional equivalent vapour compression refrigeration system [102]. A selection of working fluids adapted to ORC small-scale domestic applications was presented. Cycle efficiency for twenty different fluids were evaluated and it was concluded that R123 and R141b are the best suited fluids for microCHP applications [103]. The energy performance and thermo-economic assessment of a smallscale (100 kWe) CCHP plant serving a tertiary/residential energy demand fired by natural gas and solid biomass was assessed. Global energy efficiency in the range of 25%
45%, and IRR in the range of 15%
20% assuming the Italian subsidy framework could be attained [104]. A novel renewable biomass energy-assisted multigeneration system was introduced for useful commodities such as cooling, hydrogen, heating, drying, hot water and power generation. It included biomass gasifier unit, gas turbine cycle, Kalina cycle, reverse osmosis unit, proton exchange membrane electrolyzer, and absorption cooling cycle, dryer and heat pump subsystem. Results demonstrated that the whole energy and exergy efficiencies of the examined plant are 63.84% and 59.26%, respectively [105]. A biomass-fuelled SE with different kinds of biomass was tested such as bagasse, sawdust, switchgrass and pruned wood. Maximum power was obtained using sawdust [106]. Cogeneration coupling with TES offered a simple and economical approach for maximizing the utilization of cogenerated chilled-water, and showed 23% reduction in peak demand, 21% savings in energy consumption, with a higher IRR, greater than 25% [107]. A novel microtrigeneration system based on the compressed air and TES was proposed and assessed, in a small office building located in Chicago. The average comprehensive efficiency reached very high values of 50% and 35% in winter and summer, respectively [108]. A proposed conventional waste-driven CCHP improved the plant performance in terms of energy and sustainability indices. For the energy market of Denmark, the results showed that the thermal and electrical efficiencies of the proposed hybrid system are better than the conventional configuration by 12% and 1.3%, respectively. In addition, the exergy efficiency, sustainability index and emission reduction of 28.58%, 1.4 and 445.935 kg-CO2/GJ were obtained for the system operating with a third-generation district heating system [12]. A scenario analysis using optimization models to perform an

152

PART | I Technologies

economic, energetic and environmental assessment of a new polygeneration system in Spain in the framework of the Polycity project was conducted. The results showed that the polygeneration plant is an efficient way to reduce the primary energy consumption and CO2 emissions (up to 24%) [109]. Performance comparison of solar tower central receiver system with that of parabolic troughs in Spain revealed that parabolic troughs offer better performance as compared to towers in the summer months with thermal efficiencies up to 60%. However, the yearly performance analysis shows more energy at a higher efficiency for the tower technology [110]. A hybrid solarbiomass energy multigeneration system consisted of an ORC, a gas turbine cycle, and an absorption chiller was assessed. The values of energy and exergy efficiencies of the overall system were determined as 91.0% and 34.9%, respectively [111]. The analysis of CSP-geothermal hybrids showed that the power output of the standalone geothermal plant increased from 6.3 to 8.4 and 9.9 MWe, and plant thermal efficiency increased from 10.2% to 12.5% and 14.1%, for a parabolic trough hybrid and solar tower hybrid, respectively [112]. A solar
geothermal hybrid system for multigeneration, for power, cooling, heating and drying was designed and developed. It was observed that the energy efficiency of the single and multigeneration systems were 7% and 37%, respectively [113]. The performance of a 100 MWe solar plant using parabolic trough with three backup technologies and two cooling technologies was evaluated. The life cycle CO2 emission of plants without backup was 35 kg/MW h, which was 1.5
3 times lower than with storage (60
73 kg/MW h). For plants with natural gas backup, the emission (127
317 kg/MW h) was 4
9 times higher than without backup [114]. A 50 MWe parabolic trough CSP plant with backup was investigated. With 12% of the output energy being supplied by the backup, the maximum CO2 emission was obtained for coal (187 kg/ MW h) and the minimum for wheat straw (34 kg/MW h). The solar-only plant reported an emission of 26.9 kg/MW h [115]. In a novel arrangement, the saturated steam generated from the solar collectors was superheated by burning natural gas with pure oxygen before injection into the gas turbine. Under design conditions, the system offers high solar-to-electricity efficiency (30.8%) and solar share (59.85%) at an average collector temperature of 272.8 C [116]. A comparative study of ISCC, SEGS and CCGT has been performed. The study suggests that ISCC has the highest efficiency of 68.6% in comparison to 34.7% for a SEGS. However, the SEGS offers lower GHG emissions [117]. An integrated solar combined cooling and power (SCCP) was presented and evaluated. Combined cycle gas turbine SGT-800, a parabolic trough collector (PTC) solar field and double-effect steam driven absorption chillers, were modelled in details. Plant performance was compared against that of a conventional pure-fossil solution based on a combined cycle gas

Renewable energy based trigeneration systems—technologies Chapter | 4

153

turbine SGT-1000F and compression chillers. The SCCP plant shows the advantages over the conventional pure-fossil CC: (1) smaller GT power: SGT-800 instead of SGT-1000F; (2) higher overall efficiency both in summer and winter (3) significant fossil fuel savings in summer (33%) and winter (26%) [66]. An energy system optimization model to minimize the costs and the GHG emission was proposed for a solar CCHP system. The model was applied to 360 apartments connected through district heating network. It was found that the best solution comes from the system optimization, and not from the single device optimization. The best costs reduction obtained is 35% and the best GHG reduction obtained is 60% [32]. A large-scale solar trigeneration system with solar assisted desiccant cooling, heating and hot water generation installed in an institute building has been reported. Under suitable ambient conditions, approximately 35% of total building cooling load was met by the solar-driven desiccant cooling system [118]. A reciprocating engine, fed by rapeseed oil, was coupled to concentrating PTC to produce thermal energy and a double-stage LiBr/H2O absorption chiller to produce cooling energy. The whole trigeneration system was modelled and achieved a primary energy saving higher than 93% [119]. A system was investigated, and could produce inlet room air at 19 C when the ambient temperature was as about 31 C. The solar energy used for the regeneration of the sorbent was close to 76% of the total input energy, and the COP of the system was 0.6 [120]. A multigeneration system consisting of a solardriven ORC integrated with a triple effect LiBr
H2O absorption chiller, and a new molten salt, NaCl
MgCl2, utilized in the solar block was investigated. The exergetic utilization factor of 0.39 based on generating electricity of 428 kW could be obtained. Furthermore, the energetic COP of absorption chiller was determined to be 1.34. In the proposed system, the energetic and exergetic efficiencies of ORC were 14.4% and 26%, respectively [121]. A solar energy system capable of heating, cooling, natural ventilation and hot water supply has been built in Shanghai Research Institute of Building Science. It is used for heating in winter, cooling in summer, natural ventilation in spring and autumn, hot water supply in all the year for 460 m2 building area. The system mainly contains 150 m2 solar collector arrays, two adsorption chillers, floor radiation heating pipes, finned tube heat exchangers and a hot water storage tank of 2.5 m3 in volume, circulating pumps and a cooling tower. All components are connected by tubes and valves to form the whole circulating system. The whole system is controlled automatically by an industrial control computer. Effective operation of the adsorption cooling-based air-conditioning system was validated. Yearlong operation has confirmed that the solar system contributes 70% total energy of the involved space for the weather conditions of Shanghai [122]. A novel micro solar CCHP cycle was integrated with ORC. A thermal storage tank was installed to correct the mismatch between the supply of the solar energy and the

154

PART | I Technologies

demand of thermal source consumed by the CCHP subsystem, for stable operation. The results indicated that in summer, thermal efficiency, energy efficiency and total product cost rate in optimum case are improved to 28%, 27% and 17%, respectively, whilst in winter, these values are 4%, 13% and 4% [123]. Novel hybrid wind-solar-compressed air energy storage (WS-CAES) system was also proposed. Combined with ORC, the cascade utilization of energy with different qualities was achieved in the WS-CAES system. The results show that the electric energy storage efficiency, round trip efficiency and exergy efficiency can reach 87.7%, 61.2% and 65.4%, respectively [124].

4.8 Challenges and opportunities in renewable energybased trigeneration systems 4.8.1

Challenges and barriers

Challenges in promotion of CHP/CCHP technologies [3,5,17,23, 26,27,29,125
128]: G

G

G

G

G

G

G

G

G

G

G G

Micro and small-scale systems are prohibitively expensive for building applications mainly due to varying load profiles in buildings. Interdependence between different energy products makes system operations more complex. Optimizing smart polygeneration microgrid can be different from those for optimizing traditional polygeneration systems. The sizing, design and operation management of polygeneration systems requires a multidisciplinary approach and analysis, mainly due to high complexity of such systems, since they are mostly based on the integration of different technologies. Differences in the scale of the systems ranging from microlevel for a single home energy supply to large-level for residential district energy supply pose problems. The loads are characterized by high diversity, stochastic nature and small quantity. Dissemination of knowledge and awareness about CCHP benefits and savings is required at a larger scale. Adoption of an agreed methodology to recognise energy saving and environmental benefits is required. Promotion of research to cut down high initial and upfront costs and long payback periods is required. Research and training in the field of CHP/CCHP technologies and infrastructure is also lacking and must be promoted. Incentive policy for CHP/CCHP efficiency must be implemented. Promotion of retail electricity sale must be facilitated.

Renewable energy based trigeneration systems—technologies Chapter | 4 G

G G

G

G

G

155

A better valorisation of local resources and their more efficient use, along with renewable integration will be required in future. Development of smart metering and control solutions will be required. Establishment and maintenance of fully functioning markets has to be actively pursued by policy-makers and regulators alike. The market introduction of microCHP has to be supported actively with the help of energy utilities, research institutes, producers and networks. The installation and service networks are supposed to take a key function in designing a microCHP market launch. The technological obstacles have still remained against the development of residential microCHP systems, and its large-scale diffusion, as low price and ease of operation still a large barrier till date. Therefore future introduction of Micro CHP for domestic applications would subject to available technology, matching of electrical and thermal loads, and the gas and electricity prices. Challenges in promotion of DG [17,125,126,128,129]:

G G G G G G

G

G

G G

Planning and promotion of integrated urban heating / cooling supply; Framing of electricity grid access and interconnection regulations; Liberalization of electricity market; Reduction of gas price volatility; Standardization of grid access and interconnection requirements; Adoption of uniform technical standards for interconnecting distributed power with the grid; Accelerated development of distributed power control technology and systems; Adoption of regulatory tariffs and utility incentives to fit the new distributed power model; Modernization of regulatory laws impacting energy; Diffusion of monopoly of energy utilities. Challenges in promotion of DESs [3,30,36]:

G G

G

G

G

G

Knowledge of know-how and technical skills for DES is limited. The main contentious issue is that a long-term investment in heat distribution network is needed to dispense energy from the source to the endusers. Installations can be expensive and difficult to retrofit into existing homes/buildings. District energy demands a substantial front-end investment, often requiring extensive negotiations with investors for funding. Finding appropriate sites for DESs, so as to have the source of heat near users, can be challenging, especially in populated areas. When fossil fuel taxation is not applicable, especially in sparse areas, DE systems may not be competitive with local heating systems

156

G

G

G

PART | I Technologies

financially. In such instance, DE systems require government support to be viable. Potential monopoly provided to the owner of the thermal network, as a non- competitive market is usually not beneficial for consumers. Appropriate government regulations can avoid this problem. Modifying taxation to encourage efficiency and including energy efficiency in all federal activity and funding is required. Modernization of regulatory laws impacting energy and implementing regulations to promote energy conservation is required. Challenges in promotion of Renewable Energy Sector [5,65,130
133]:

G

G

G

G

G

G

G

G G G

G G

Whilst utilization of RES, the intermittent nature of solar energy and wind power need special attention. RES is not so easy to integrate in the urban environment because of its low-energy density (at least compared to fossil fuels). RES technologies often require connection and access to the existing local energy networks that are often congested, or simply missing. Encouragement of hybrid power generation systems based on renewable energy technologies. Promoting research and development activities in the field of renewable energy technologies. Promoting awareness campaign about renewable energy technologies, along with widespread training and formation of a pool of experts. Integration and streamlining the policies on renewable energy technologies. Technological and financial limitations to be overcome. Effective strategies for international collaborations to be enhanced. Promoting incentive schemes for best practices in feed-in-tariffs, public
private partnership, etc. Deployments of renewable energy technologies to rural areas. Challenges arising due to lack of investment guidelines, and sluggish government support must be addressed properly.

4.8.2

Opportunities and prospects

Opportunities in CHP/CCHP [3,6,15,26,27,128,134,135]: G

G

In the longer term, brighter prospects for CHP/CCHP is expected due to free access of independent power producers (who may exploit market niches) and higher fuel prices (e.g. due to increasing scarcity). Moreover, coupled with modern data communication technologies, a significant number of ‘virtual power plants’ might be commercially established within a few years.

Renewable energy based trigeneration systems—technologies Chapter | 4 G

G

G

G

G G G

G

G

G

G

157

Supply side is promising due to the fact that the microCHP installations serves and binds the customer to a long-term usage. Legal regulations for power supply support an accelerated market implementation for CHP technologies. Promotion of CCHP is particularly interesting in countries having large electricity blackout periods. New revenue streams may be generated by selling excess electricity and providing saleable steam and heat or other industry-specific products. Increased energy reliability due to ability to run on multiple fuels. Greater flexibility in transmission and distribution planning. Integration of renewables such as solar energy in hybrid energy systems, and industry-specific benefits such as waste management, reduction, and productive sustainable use, will be obtained. Enhanced electricity network stability through reduction in congestion and ‘peak-shaving’. Beneficial use of local and renewable energy resources, like the use of waste, biomass, and geothermal resources in district heating/cooling systems. Reduced reliance on imported fossil fuels as well as very low specific GHG and other emissions. Primary energy saving; reduction of fuel costs; investment saving, all can lead to shorter payback period for trigeneration systems. Opportunities in DG [29,136]:

G G G

G G G

G

Self-sufficiency; Reduction of the dependence on the external energy supply; Local production, low transmission losses, by avoiding transportation costs; No need for expansion of costly grid interconnection; Lower capital costs with respect to centralized plants; Reduction of the fossil fuels dependency and the mitigation of environmental impact; Reduction of breakdown effects of a single unit on the whole energy network. Opportunities in DES [30]:

G

G

For society: flexibility in choosing heat sources (permitting more costeffective operation), independence of a sole heat source, reduced fuel consumption, enhanced environmental quality, emissions reductions (due to higher efficiency and lower fuel consumption), and reduced CFC utilization via district cooling. For communities: enhanced community energy management and employment, increased opportunities to use local energy resources, greater

158

G

PART | I Technologies

ability for controlling environmental emissions, retention of energy capital in the local economy, and reduced fuel costs. For building owners and tenants: reduced heating costs, reduced operation cost and complexity, safer operation, reduced space requirements, improved comfort, and increased reliability. Opportunities in Renewable Energy Sector [9,57,90,93,137,138]:

G G G

G G G

G

G

G

G

Decentralization of heat and electricity supply; Reduced transmission line losses; Higher overall primary energy utilization efficiency by exploiting waste energy to supply heat Increased flexibility and use for remote locations; Reliable energy supply and increase in energy security; Environment protection and reduction of GHG and other harmful emissions; A comparative study on the world energy consumption released by International Energy Agency (IEA) shows that in 2050, solar array installations will supply around 45% of energy demand in the world; Prediction of unit electricity costs for new technologies are subject to many uncertainties that include system reliability, equipment efficiencies and lifetimes, organizational learning, manufacturing capability, and technological improvements; Cost reductions will come from technical improvements, larger plant sizes, and large volume production, along with market acceptance; Organic Rankine power cycles are becoming a leading technology for energy conversion, especially to convert low-temperature heat source. Available heat sources are solar energy, geothermal energy, biomass, and waste heat from various thermal processes (industrial energy intensive process, gas turbines and ICEs.

4.9

Conclusions

Sustainability is the comprehensive paradigm which governs the evolution of all future anthropogenic activities. Sustainable energy systems must address technical, economic and environmental issues [139]. Efficiency and sustainability are key concepts in today’s energy systems. Sustainable energy systems should have at least the following features: (1) maximum possible resource utilization, (2) reduced GHG emission and (3) economical feasibility. Polygeneration system addresses all these issues [28,139]. According to this outlook, future energy systems will be based on renewable energy sources (RES), buildings as positive power plants, smart grids and electric vehicles, and energy storage [7].

Renewable energy based trigeneration systems—technologies Chapter | 4

159

Building energy regulations and energy certification programmes are basic tools for the improvement of energy efficiency in this sector, and aim at setting minimum energy efficiency requirements in new buildings [140]. The level of CHP development in a country depends on heating and cooling demand in the industrial, commercial and residential sectors. IEA has gathered data regarding share of CHP electricity generation in total national electricity generation from over 40 countries. Two challenges have confronted this task: G G

Not all countries systematically collect CHP data. Where countries do collect data, they tend to use similar methodologies [128].

There is no internationally comparable standard CHP data except the EU, where there is a standard methodology across all its member states [128]. Diffusion of CHP into liberalized energy markets is likely to be modest, at least in the early stages of market opening. The development of Micro CCHP systems in competitive markets is rather disenchanting. Main reasons are seen in the rather slow technology development, in the assessment of the economic opportunities, in the institutional and regulatory framework, in lacking innovation policies, and consumer acceptance [3,27]. Better understanding of user demands, careful selection of technologies and full consideration of revenue are the keystones to a successful CCHP application. It is apparent that government policies, liberation of the electricity market and price of electricity and fuels are critical in the development of CCHP [17]. Legislative initiatives play a basic role to support these technologies. For example, EU has introduced directives on emission trading, on electricity and gas and on the energy performance of building. Further incentives, such as low tax rates on gas, carbon tax exemption, dispatch priority in the transmission grid and economic instruments to support high energy efficiency systems, could be introduced by governments [15]. Proliferated utilization of renewable energy sources and renewable energy-based technologies could potentially tackle a comprehensive range of issues, such as energy security, energy equity problems, energy conversion and use related emissions. Additionally, renewable energy resources successfully address a variety of other sustainability related issues for instance poverty reduction, clean water protection, development of transport, agriculture, infrastructure, and industry, job creation, etc. [93]. Relatively small-capacity-distributed CCHP units are the trend in future applications. Novel technologies such as fuel cells, microturbines, stirling engines, adsorption chillers and dehumidifiers are emerging, which possess some promising characteristics, including low emission, high efficiency and low-grade thermal energy recovery [17,141].

160

PART | I Technologies

Abbreviations ACH AD CC CCGT CCHP CFC CHP CO2 COP CSP DE DER DES DSG EDM FEL FTL GHG GJ h HCFC HFC HRSG HTF ICE IEA ISCC kWe kWth LHS MFH MW NOx ORC PCM PTC PV PV/T SCCP SEGS SFH SHS SME SO2 TDM TES

absorption chiller anaerobic digestion combined cycle combined cycle gas turbine combined cooling, heating and power chlorofluorocarbon combined heat and power carbon dioxide coefficient of performance concentrating solar plant district energy decentralized energy resources district energy system direct steam generation electrical demand management following electrical load following thermal load greenhouse gases giga-joule hour hydrochlorofluorocarbon hydrofluorocarbon heat-recovery steam generator heat transfer fluid internal combustion engine International Energy Agency Integrated Solar Combined Cycle kilowatt-electric kilowatt-thermal latent heat storage medium family houses megawatt nitrogen oxides organic rankine cycle phase change material parabolic trough collector photovoltaic photovoltaic-thermal solar combined cooling and power solar electric generation system small family houses sensible heat storage small and medium enterprises sulphur dioxide thermal demand management thermal energy storage

Renewable energy based trigeneration systems—technologies Chapter | 4 VCRS WASP WS-CAES ZEB

161

vapour compression refrigeration system Wind Atlas Analysis and Application Programme wind-solar-compressed air energy storage zero energy building

References [1] Martinez S, Michaux G, Salagnac P, Bouvier JL. Micro-combined heat and power systems (micro-CHP) based on renewable energy sources. Energ Convers Manag 2017;154:262
85. [2] Greene N, Hammerschlag R. Small and clean is beautiful. Exploring the emissions of distributed generation and pollution prevention policies. Electr J 2000;13:50
60. [3] Madlener R, Schmid C. Combined heat and power generation in liberalised markets and a carbon-constrained world. Sustain Energy Provision, GAIA 2003;12:2. [4] Murugan S, Hor´ak B. Tri and polygeneration systems—a review. Renew Sustain Energ Rev 2016;60:1032
51. [5] Rong A, Su Y. Polygeneration systems in buildings: a survey on optimization approaches. Energy Build 2017;151:439
54. [6] Moussawi HA, Fardoun F, Gualous HL. Review of tri-generation technologies: design evaluation, optimization, decision-making, and selection approach. Energ Convers Manag 2016;120:157
96. [7] Krajaˇci´c G, Milan Vujanovi´c M, Neven D, Kılkı¸s S, Rosen MA, Al-Nimr MA. Integrated approach for sustainable development of energy, water and environment systems. Energ Convers Manag 2018;159:398
412. [8] Oztop HF, Hepbasli A. Cogeneration and trigeneration applications. Energy Sources Part A: Recovery, Utilization, Environ Eff 2006;28(8):743
50. [9] Yilmaz S, Selim H. A review on the methods for biomass to energy conversion systems design. Renew Sustain Energ Rev 2013;25:420
30. [10] Ullah KR, Saidur R, Ping HW, Akikur RK, Shuvo NH. A review of solar thermal refrigeration and cooling methods. Renew Sustain Energ Rev 2013;24:499
513. [11] Pacesila M, Burcea SG, Colesca SE. Analysis of renewable energies in European Union. Renew Sustain Energ Rev 2016;56:156
70. [12] Nami H, Arabkoohsar A, Moghaddam AA. Thermodynamic and sustainability analysis of a municipal waste-driven combined cooling, heating and power (CCHP) plant. Energ Convers Manag 2019;201:112
58. [13] Tatsidjodoung P, Pierre NL, Luo L. A review of potential materials for thermal energy storage in building applications. Renew Sustain Energ Rev 2013;18:327
49. [14] Wang JJ, Jing YY, Zhang CF, Zhai ZJ. Performance comparison of combined cooling heating and power system in different operation modes. Appl Energ 2011;88:4621
31. [15] Angrisani G, Roselli C, Sasso M. Distributed microtrigeneration systems. Prog Energ Combust Sci 2012;38:502
21. [16] Onovwionaa HI, Ugursalb VI. Residential cogeneration systems: review of the current technology. Renew Sustain Energ Rev 2006;10:389
431. [17] Wu DW, Wang RZ. Combined cooling, heating and power: a review. Progr Energ Combust Sci 2006;32:459
95. [18] Ge YT, Tassou SA, Chaer I, Suguartha N. Performance evaluation of a tri-generation system with simulation and experiment. Appl Energ 2009;86:2317
26.

162

PART | I Technologies

[19] Eisavi B, Khalilarya S, Chitsaz A, Rosen MA. Thermodynamic analysis of a novel combined cooling, heating and power system driven by solar energy. Appl Therm Eng 2018;129:1219
29. [20] Small-scale cogeneration, why? In which case? A guide for decision makers. European Commission, Directorate General for Energy DGXVII; July 1999. [21] Resource Dynamics Corporation. Assessment of distributed generation technology applications; February 2001. [22] Ackermann T, Andersson G, Soder L. Distributed generation: a definition. Electr Power Syst Res 2001;57:195
204. [23] Liu M, Shi Y, Fang F. Combined cooling, heating and power systems: a survey. Renew Sustain Energ Rev 2014;35:1
22. [24] Oland CB, et al. Guide to combined heat and power systems for boiler owners and operators. United States: Department of Energy; 2004. [25] Tora E. Integration and optimization of trigeneration systems with solar energy, biofuels, process heat, and fossil fuels, a PhD dissertation; 2010. [26] Kuhn V, Klemes J, Bulatov I. MicroCHP: overview of selected technologies, products and field test results. Appl Therm Eng 2008;28:2039
48. [27] Maghanki MM, Ghobadian B, Najafi G, Galogah RJ. Micro combined heat and power (MCHP) technologies and applications. Renew Sustain Energ Rev 2013;28:510
24. [28] Jana K, Ray A, Majoumerd MM, Assadi M, De S. Polygeneration as a future sustainable energy solution—a comprehensive review. Appl Energy 2017;202:88
111. [29] Calise F, Vastogirardi GN, Accadia MD, Vicidomini M. Simulation of polygeneration systems. Energy 2018;163:290
337. Available from: https://doi.org/10.1016/j. energy.2018.08.052. [30] Rezaie B, Rosen MA. District heating and cooling: review of technology and potential enhancements. Appl Energy 2012;93:2
10. [31] Rosen MA. Energy considerations in design for environment: appropriate energy selection and energy efficiency. Int J Green Energy 2004;1(1):21
45. [32] Anatone M, Panone V. A model for the optimal management of a CCHP plant. Energy Procedia 2015;81:399
411. [33] Jradi M, Riffat S. Tri-generation systems: energy policies, prime movers, cooling technologies, configurations and operation strategies. Renew Sustain Energ Rev 2014;32: 396
415. [34] Markovic D, Cvetkovic D, Masic B. Survey of software tools for energy efficiency in a community. Renew Sustain Energ Rev 2011;15:4897
903. [35] Connolly D, Lund H, Mathiesen BV, Leahy M. A review of computer tools for analysing the integration of renewable energy into various energy systems. Appl Energy 2010;87: 1059
82. [36] Swithenbank J, Finney KN, Chen Q, Yang YB, Nolan A, Sharifi VN. Waste heat usage. Appl Therm Eng 2013;60:430
40. [37] Mardiana A, Riffat SB. Review on physical and performance parameters of heat recovery systems for building applications. Renew Sustain Energ Rev 2013;28:174
90. [38] ASHRAE AH-H. Systems and equipment. Atlanta: Am Soc Heat Refrig Air- Cond Eng Inc.; 2000. [39] Fernandes EDO. Smart cities initiative: how to foster a quick transition towards local sustainable energy systems. Think project, 7th framework programme. European University Institute; 2010.

Renewable energy based trigeneration systems—technologies Chapter | 4

163

[40] Balaras CA, Grossman G, Henning HM, Ferreira CAI, Podesser E, Wang L, et al. Solar air conditioning in Europe—an overview. Renew Sustain Energ Rev 2007;11:299
314. [41] Deng J, Wang RZ, Han GY. A review of thermally activated cooling technologies for combined cooling, heating and power systems. Prog Energy Combust Sci 2011;37: 172
203. [42] Kuravi S, Trahan J, Goswami DY, Rahman MM, Stefanakos EK. Thermal energy storage technologies and systems for concentrating solar power plants. Prog Energ Combust Sci 2013;39:285
319. [43] Al-Abidi AA, Mat SB, Sopian K, Sulaiman MY, Lim CH, Th A. Review of thermal energy storage for air conditioning systems. Renew Sustain Energ Rev 2012;16:5802
19. [44] Abedin AH, Rosen MA. Assessment of a closed thermochemical energy storage using energy and exergy methods. Appl Energy 2012;93:18
23. [45] Gil A, Medrano M, Martorell I, Lazaro A, Dolado P, Zalba B, et al. State of the art on high temperature thermal energy storage for power generation. Part 1—concepts, materials and modellization. Renew Sustain Energ Rev 2010;14:31
55. [46] Hasnain SM. Review on sustainable thermal energy storage technologies, Part I: heat storage materials and techniques. Energ Convers Manag 1998;39:1127
38. [47] Pinel P, Cruickshank CA, Beausoleil-Morrison I, Wills A. A review of available methods for seasonal storage of solar thermal energy in residential applications. Renew Sustain Energ Rev 2011;15:3341
59. [48] Barlev D, Vidu R, Stroeve P. Innovation in concentrated solar power. Sol Energ Mater Sol Cell 2011;95:2703
25. [49] Tian Y, Zhao CY. A review of solar collectors and thermal energy storage in solar thermal applications. Appl Energy 2013;104:538
53. [50] Anisur MR, Mahfuz MH, Kibria MA, Saidur R, Metselaar IHSC, Mahlia TMI. Curbing global warming with phase change materials for energy storage. Renew Sustain Energ Rev 2013;18:23
30. [51] Nallusamy N, Sampath S, Velraj R. Experimental investigation on a combined sensible and latent heat storage system integrated with constant/varying (solar) heat sources. Renew Energy 2007;32:1206
27. [52] Moens L, Blake DM. Advanced heat transfer and thermal storage fluids. In: Proceedings international solar energy conference; 2005. p. 791
793. [53] Moens L, Blake DM, Rudnicki DL, Hale MJ. Advanced thermal storage fluids for solar parabolic trough systems. J Sol Energy Eng Trans ASME 2003;125:112
16. [54] Dincer I, Rosen MA. Energy storage systems, in thermal energy storage. John Wiley & Sons Ltd. 2010;51
82. [55] Sharma A, Tyagi VV, Chen CR, Buddhi D. Review on thermal energy storage with phase change materials and applications. Renew Sustain Energ Rev 2009;13(2):318
45. [56] Knox-Hayes J. Towards a moral socio-environmental economy: a reconsideration of values. Geoforum 2015;65:297
300. [57] Dincer I, Acar C. Smart energy systems for a sustainable future. Appl Energy 2017;194: 225
35. [58] Nakata T, Silva D, Rodionov M. Application of energy system models for designing a low-carbon society. Prog Energ Combust Sci 2011;37:462
502. [59] Demirbas A. Potential applications of renewable energy sources, biomass combustion problems in boiler power systems and combustion related envi- ronmental issues. Prog Energ Combust Sci 2005;31(1):71
92.

164

PART | I Technologies

[60] Boyle G. Renewable energy: power for a sustainable future. Oxford: Oxford University Press; 2000. [61] Berinstein P. Alternative energy: facts, statistics, and issues. Westport: Oryx Press; 2001. [62] Johansson TB, Kelly H, Reddy AKN, Williams RH. Renewable energy: sources for fuels and electricity. Washington, DC: Island Press; 1993. [63] Sims REH. The brilliance of bioenergy in business and in practice. London: James & James; 2002. [64] Ozlu S, Dincer I. Development and analysis of a solar and wind energy based multigeneration system. Sol Energy 2015;122:1279
95. [65] Ogbonnaya C, Abeykoon C, Damo UM, Turan A. The current and emerging renewable energy technologies for power generation in Nigeria: a review. Therm Sci Eng Prog 2019;13:100390. [66] Perdichizzi A, Barigozzi G, Franchini G, Ravelli S. Peak shaving strategy through a solar combined cooling and power system in remote hot climate areas. Appl Energy 2015;143: 154
63. [67] Suleman F, Dincer I, Angelin-Chaab M. Development of an integrated renewable energy system for multigeneration. Energy 2014;78:196
204. [68] Sahin AD. Progress and recent trends in wind energy. Prog Energ Combust Sci 2004;30: 501
43. [69] T.B.W.E.A. BWEA, wind turbine technology. ,https://www.nottingham.ac.uk/renewableenergyproject/documents/windturbinetechnology.pdf.; 2005. [70] Petersen EL, Troen I, Frandsen S, Hedegaard K. Wind Atlas for Denmark. A rational method for wind energy sitting, Riso-R-428. Roskilde: Riso National Laboratory;; 1981. [71] Alternative Energy Tutorials, Wind Turbine Design for Wind Power. ,http://www.alternative-energy-tutorials.com/wind-energy/wind-turbine-design.html.; 2018. [72] IEA. Energy technology perspectives 2008: scenarios and strategies to 2050. Paris: OECD/IEA; 2008. [73] DECC. UK renewable energy roadmap. Carbon N.Y. 2011;5:293
98. [74] Pramanik S, Ravikrishna RV. A review of concentrated solar power hybrid technologies. Appl Therm Eng 2017;127:602
37. [75] Zarrouk SJ, Moon H. Efficiency of geothermal power plants: a worldwide review. Geothermics 2014;51:142
53. [76] Franco A, Villani M. Optimal design of binary cycle power plants for water- dominated, medium-temperature geothermal fields. Geothermics 2009;38:379
91. [77] Marinova M, Beaudry C, Taoussi A, Trepanier M, Paris J. Economic assessment of rural district heating by bio-steam supplied by a paper mill in Canada. Bull Sci Technol Soc 2008;28(2):159
73. [78] Lund JW. Geothermal taping focus: tapping the earth’s natural heat. Refocus 2006;7: 48
51. [79] Demirbas A. Waste management, waste resource facilities and waste conversion processes. Energ Convers Manag 2011;52:1280
7. [80] Ramachandra TV, Jain R, Krishnadas G. Hotspots of solar potential in India. Renew Sustain Energ Rev 2011;15:3178
86. [81] Misra RD, Murthy MS. Straight vegetable oils usage in a compression ignition engine—a review. Renew Sustain Energ Rev 2010;14:3005
13. [82] Agarwal AK. Biofuels (alcohols and biodiesel) applications as fuels for internal combustion engines. Prog Energ Combust Sci 2007;33:233
71.

Renewable energy based trigeneration systems—technologies Chapter | 4

165

[83] Hossain AK, Davies PA. Plant oils as fuels for compression ignition engines: a technical review and life-cycle analysis. Renew Energy 2010;35:1
13. [84] Atabani AE, Silitonga AS, Ong HC, Mahlia TMI, Masjuki HH, Badruddin IA, et al. Nonedible vegetable oils: a critical evaluation of oil extraction, fatty acid compositions, biodiesel production, characteristics, engine performance and emissions production. Renew Sustain Energ Rev 2013;18:211
45. [85] Murugesan A, Umarani C, Subramanian R, Nedunchezhian N. Bio-diesel as an alternative fuel for diesel engines—a review. Renew Sustain Energ Rev 2009;13:653
62. [86] Mofijur M, Atabani AE, Masjuki HH, Kalam MA, Masum BM. A study on the effects of promising edible and non-edible biodiesel feedstocks on engine performance and emissions production: a comparative evaluation. Renew Sustain Energ Rev 2013;23:391
404. [87] Demirbas A. Progress and recent trends in biofuels. Prog Energ Combust Sci 2007;33: 1
18. [88] Hamelinck C, Broek RVD, Rice B, Gilbert A, Ragwitz M, Toro F. Liquid biofuels strategy study for Ireland. A report of sustainable energy Ireland (report no. 04-RERDD-015R-01); 2004. p. 1
105. [89] Wegener M, Malmquist A, Isalgue A, Martin A. Biomass-fired combined cooling, heating and power for small scale applications—a review. Renew Sustain Energ Rev 2018; 96:392
410. [90] Reddy VS, Kaushik SC, Ranjan KR, Tyagi SK. State-of-the-art of solar thermal power plants—a review. Renew Sustain Energ Rev 2013;27:258
73. [91] Nastasi B, Matteo UD. Solar energy technologies in sustainable energy action plans of Italian big cities. Energy Procedia 2016;101:1064
71. [92] Kasaeian A, Nouri G, Ranjbaran P, Wen D. Solar collectors and photovoltaics as combined heat and power systems: a critical review. Energ Convers Manag 2018;156: 688
705. [93] Mekhilef S, Saidur R, Safari A. A review on solar energy use in industries. Renew Sustain Energ Rev 2011;15:1777
90. [94] Grubert G. Conventional hydroelectricity and the future of energy: linking national inventory of dams and energy information administration data to facilitate analysis of hydroelectricity. Electr J 2020;33(1):106692. [95] Filho GLT, Santos IFSD, Barros RM. Cost estimate of small hydroelectric power plants based on the aspect factor. Renew Sustain Energ Rev 2017;77:229
38. [96] Petley S, Starr D, Parish L, Underwood Z, Aggidis GA. Opportunities for tidal range projects beyond energy generation: using mersey barrage as a case study. Front Architectural Res 2019;8(4):620
33. [97] Neill SP, Angeloudis A, Robins PE, Walkington I, Ward SL, masters I, et al. Tidal range energy resource and optimization—past perspectives and future challenges. Renew Energy 2018;127:763
78. [98] Lin L, Wang Y, Shemmeri TA, Ruxton T, Turner S, Zeng S, et al. An experimental investigation of a household size trigeneration. Appl Therm Eng 2007;27:576
85. [99] Sonar D, Soni SL, Sharma D, Shrivastava A, Goyal R. Performance and emission characteristics of a diesel engine with varying injection pressure and fuelled with raw mahua oil (preheated and blends) and mahua oil methyl ester. Clean Technol Environ Policy 2015;17:1499
511. [100] Foresti S, Manzolini G. Performances of a micro-CHP system fed with bio-ethanol based on fluidized bed membrane reactor and PEM fuel cells. Int J Hydrog Energy 2016;41: 9004
21.

166

PART | I Technologies

[101] Suamir IN, Tassou SA. Performance evaluation of integrated trigeneration and CO2 refrigeration systems. Appl Therm Eng 2013;50(2):1487
95. [102] Patel B, Desai NB, Kachhwaha SS. Optimization of waste heat based organic Rankine cycle powered cascaded vapor compression-absorption refrigeration system. Energ Convers Manag 2017;154:576
90. [103] Larjola J. 9—Organic Rankine cycle (ORC) based waste heat/waste fuel recovery systems for small combined heat and power (CHP) applications A2—. In: Beith R, editor. Small and micro combined heat and power (CHP) systems. Woodhead Publishing; 2011. p. 206
32. [104] Pantaleo AM, Camporeale SM, Markides CN, Mugnozza GS, Shah N. Energy performance and thermo-economic assessment of a microturbine-based dual-fuel gas-biomass trigeneration system. Energy Procedia 2017;105:764
72. [105] Yilmaz F, Ozturk M, Selbas R. Development and techno-economic assessment of a new biomass-assisted integrated plant for multigeneration. Energ Convers Manag 2019;202: 112154. [106] Damirchi H, Najafi G, Alizadehnia S, Mamat R, Nor Azwadi CS, Azmi WH, et al. Micro combined heat and power to provide heat and electrical power using bio- mass and Gamma-type Stirling engine. Appl Therm Eng 2016;103:1460
9. [107] Khan KH, Rasul MG, Khan MM. Energy conservation in buildings: cogeneration and cogeneration coupled with thermal-energy storage. Appl Energ y 2004;77:15
34. [108] Li Y, Wang X, Li D, Ding Y. A trigeneration system based on compressed air and thermal energy storage. Appl Energy 2012;99:316
23. [109] Ortiga J, Bruno JC, Coronas A. Operational optimisation of a complex trigeneration system connected to a district heating and cooling network. Appl Therm Eng 2013;50 (2):1536
42. [110] Franchini G, Perdichizzi A, Ravelli S, Barigozzi G. A comparative study between parabolic trough and solar tower technologies in solar Rankine cycle and integrated solar combined cycle plants. Sol Energy 2013;98:302
14. [111] Khalid F, Dincer I, Rosen MA. Thermoeconomic analysis of a solar-biomass integrated multigeneration system for a community. Appl Therm Eng 2017;120:645
53. [112] Peterseim JH, White S, Tadros A, Hellwig U. Concentrating solar power hybrid plants— enabling cost effective synergies. Renew Energy 2014;67:178
85. [113] Panchal S, Dincer I, Chaab MA. Analysis and evaluation of a new renewable energy based integrated system for residential applications. Energy Build 2016;128:900
10. [114] Klein SJW, Rubin ES. Life cycle assessment of greenhouse gas emissions, water and land use for concentrated solar power plants with different energy backup systems. Energy Policy 2013;63:935
50. [115] Corona B, Miguel GS. Environmental analysis of a concentrated solar power (CSP) plant hybridised with different fossil and renewable fuels. Fuel 2015;145:63
9. [116] Gou C, Cai R, Hong H. A novel hybrid oxy-fuel power cycle utilizing solar thermal energy. Energy 2007;32:1707
14. [117] Dersch J, Geyer M, Herrmann U, Jones SA, Kelly B, Kistner R, et al. Trough integration into power plants-a study on the performance and economy of integrated solar combined cycle systems. Energy 2004;29:947
59. [118] Hands S, Sethuvenkatraman S, Peristy M, Rowe D, White S. Performance analysis & energy benefits of a desiccant based solar assisted trigeneration system in a building. Renew Energy 2016;85:865
79.

Renewable energy based trigeneration systems—technologies Chapter | 4

167

[119] Calise F, d’Accadia MD, Vanoli L. Design and dynamic simulation of a novel solar trigeneration system based on hybrid photovoltaic/thermal collectors (PVT). Energ Convers Manag 2012;60:214
25. [120] Wang RZ, Oliveira RG. Adsorption refrigeration—an efficient way to make good use of waste heat and solar energy. Prog Energy Combust Sci 2006;32:424
58. [121] Sharifishourabi M, Chadegani EA. Performance assessment of a new organic Rankine cycle based multigeneration system integrated with a triple effect absorption system. Energ Convers Manag 2017;150:787
99. [122] Zhai XQ, Wang RZ, Dai YJ, Wu JY, Xu YX, Ma Q. Solar integrated energy system for a green building. Energy Build 2007;39:985
93. [123] Boyaghchi FA, Heidarnejad P. Thermoeconomic assessment and multi objective optimization of a solar micro CCHP based on organic Rankine cycle for domestic application. Energ Convers Manag 2015;97:224
34. [124] Ji W, Zhou Y, Sun Y, Zhang W, An B, Wang J. Thermodynamic analysis of a novel hybrid wind-solar-compressed air energy storage system. Energ Convers Manag 2017; 142:176
87. [125] Andrepont JS. Distributed generation: benefits and barriers. Cogener Compet Power J 2000;15(4):24
40. [126] Casten TR. Turning off the heat. Amherst, New York: Prometheus Books; 1998. [127] Comodi G, Lorenzetti M, Salvi D, Arteconi A. Criticalities of district heating in Southern Europe: lesson learned from a CHP-DH in Central Italy. Appl Therm Eng 2017;112:649
59. [128] IEA. Combined Heat and Power—Evaluating the benefits of greater global investment, International Energy Agency (IEA); 2008. [129] Greenberg S, Cooley C. Development, demonstration, and field testing of enterprisewide distributed generation energy management system, final report NREL/SR-56036951. U.S. Department of Energy; 2005. [130] Mohammed YS, Mustafa MW, Bashir N, Ibrahem IS. Existing and recommended renewable and sustainable energy development in Nigeria based on autonomous energy and microgrid technologies. Renew Sustain Energ Rev 2017;75:820
38. [131] Wojuola RN, Alant BP. Public perceptions about renewable energy technologies in Nigeria, African. J Sci Technol Innov Dev 2017;9:399
409. [132] Sambo AS. Electricity from renewable energy resources for remote locations: a case for solar and wind energy; 2006. [133] Fernandes EO. Smart cities initiative: how to foster a quick transition towards local sustainable energy systems, EU’s 7th framework programme, final report; January 2011. [134] Santos M, Andre J, Mendes R, Ribeiro JB. Design and modelling of a small scale biomass-fueled CHP system based on Rankine technology. Energy Procedia 2017;129: 676
83. [135] Hawkes AD, Leach MA. Cost-effective operating strategy for residential microcombined heat and power. Energy 2007;32(-5):711
23. [136] Barsali S, De Marco A, Giglioli R, Ludovici G, Possenti A. Dynamic modelling of biomass power plant using micro gas turbine. Renew Energy 2015;80:806
18. [137] Yagoub W, Doherty P, Riffat SB. Solar energy-gas driven micro-CHP system for an office building. Appl Therm Eng 2006;26:1604
10. [138] Tempesti D, Manfrida G, Fiaschi D. Thermodynamic analysis of two micro CHP systems operating with geothermal and solar energy. Appl Energy 2012;97:609
17.

168

PART | I Technologies

[139] Chicco G. Sustainability challenges for future energy systems. J Sustain Energ 2010;1 (1):6
26. [140] Lombard LP, Ortiz J, Coronel JF, Maestre IR. A review of HVAC systems requirements in building energy regulations. Energy Build 2011;43:255
68. [141] Sonar D, Soni SL, Sharma D. Micro-trigeneration for energy sustainability: technologies, tools and trends. Appl Therm Eng 2014;71:790
6.

Further reading IEA EBC, 2014IEA EBC. In: Entchev E, Tzscheutschler P, editors. Integration of microgeneration and related technologies in building; 2014. J Swithenbank, 2013J Swithenbank, K. N. Finney, Q. Chen, Y. B. Yang, A Nolan, V. N. Sharifi, Waste heat usage, Applied Thermal Engineering 60 (2013) 430-440. MacRae, 1992MacRae M. Realizing the benefits of community integrated energy system. Canadian Energy Research Institute; 1992.

Chapter 5

Integrated power transmission and distribution systems Abolhassan Mohammadi Fathabad1, Jianqiang Cheng1 and Kai Pan2 1

Department of Systems and Industrial Engineering, University of Arizona, Tucson, AZ, United States, 2Department of Logistics and Maritime Studies, Hong Kong Polytechnic University, Kowloon, Hong Kong SAR, China

Chapter Outline 5.1 Introduction 5.2 Mathematical model 5.2.1 First-stage unit commitment model 5.2.2 Second-stage economic dispatch model 5.2.3 Distributed energy resource management problem 5.2.4 Tighter formulations

5.1

169 174 175 177

178 183

5.3 Numerical results 5.3.1 Isolated unit commitment problem 5.3.2 Isolated distributed energy resource management problem 5.3.3 Integrated transmission and distribution systems 5.3.4 IEEE 118-bus network results 5.4 Conclusions References

184 185

190 192 195 197 198

Introduction

The interactions between the transmission system (TS) and distribution system (DS) of the power systems are increasing as a result of integrating large amounts of on-site generations in the DS through devices referred to as distributed energy resources (DERs). These DER generations, mostly from intermittent renewable resources, are driven by economic factors of on-site generations, clean means of renewable energy production, and recent advances in DER technology. However, the expansion of these technologies is limited by the high operational costs of stabilizing the power system. In this chapter, we introduce an integrated transmission and distribution system (InTDS) so as to minimize the operational costs of the TS and DS, respectively, while respecting the technical constraints of both systems. Our proposed InTDS model can address several challenges of Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00005-4 © 2021 Elsevier Inc. All rights reserved.

169

170

PART | I Technologies

FIGURE 5.1 Conventional transmission and distribution systems.

the modern power systems (e.g. mitigating renewable production uncertainty and stabilizing energy flow). The TS and DS have transitioned from a conventionally centralized paradigm to a currently decentralized one. The conventional power systems in Fig. 5.1 have one-way energy flows from generation to consumption, and they include large-scale central power stations (such as coal-, gas- and nuclear-powered plants) where electricity is generated and then delivered to consumers by conventional transmission and distribution networks. In contrast, a decentralized power system can be divided into many modern DSs, which nowadays are turning to localized grids that could be independent of

Integrated power transmission and distribution systems Chapter | 5

171

FIGURE 5.2 Modern transmission and distribution systems.

the transmission system (TS). In particular, the modern DSs can harness onsite generations to serve local loads through near-demand generation and storage components, as shown in Fig. 5.2. They usually represent lowvoltage AC power grids and use such DERs as solar panels, wind turbines and stationary batteries, which can help reduce carbon emission significantly. The DS is connected to the TS through the point of common coupling (PCC), but it could be isolated from the TS. Thus the electricity generated from both the TS and DS can be easily delivered to the consumers, leading to a highly reliable service. Currently most of these modern DSs are short of electricity generation and need to purchase large amounts of electricity from the TS via PCC at peak hours (leading to the so-called ‘needy DSs’). However, the increasing penetration of renewable energy resources (RES) in the DS is set to change this notion. Indeed, investments in renewable generation developments were higher than investments in the fossil fuel energy sector for the first time in

172

PART | I Technologies

2017 [1]. As a result, the DSs are growing more independent over time, and they will be able to satisfy more demands internally (leading to the so-called ‘balanced DSs’). Furthermore, they may generate extra electricity to share with other sections of the transmission network (leading to the so-called ‘rich DSs’). Thus the shift of paradigm towards a decentralized power system is going to intensify for the years to come, and it will be centred around the high penetration of renewable generations. The major challenge in the TS is that electricity must be generated at the same rate as it is consumed. A sophisticated optimization and control system is required to ensure exact match between the generation and consumption at any moment. The TS is usually operated by a transmission system operator (TSO) who balances the supply and demand through scheduling and controlling physical devices such as electrical power generators, switches and circuit breakers. More specifically TSOs solve a family of optimization problems called unit commitment (UC) problem to coordinate a set of generators that are geographically located at different locations to match the network’s demand. Solving the UC problem is very important because an imbalance between the generation and consumption may lead to major blackouts and shutdowns. However, solving the UC problem is very challenging because (1) on the generation side, a large number of generators are available to be online/offline in the TS (hundreds to thousands) and are geographically located at different regions, and (2) on the demand side, it is very challenging to accurately forecast future demands of the system, especially when different types of wind and solar generations are integrated in the system because they are highly uncertain and difficult to predict. As a consequence, the UC problem is usually very difficult to solve. Especially note that some units require hours to start up or shut down. It follows that the UC decisions should be made in advance, which implies that the UC problem needs to be solved within several minutes to a few hours. Therefore, for the purpose of simplification, most of the UC problems in the literature are solved by assuming that the DSs connected to the transmission network are simple load points to represent the whole complex DSs. For the DS, it generates electricity from three sources to satisfy demands: (1) electricity purchased from the TS, (2) internal generations by DERs (nondispatchable and dispatchable) and (3) control tools (such as storage units). Optimizing the operation of distributed generations and storage facilities under uncertain renewable generations and loads leads to a specific mathematical optimization problem, and we call it DER management problem. Such problem minimizes the operating cost of distributed generations, enhances the efficiency of the DER’s utilization and stabilizes the DS [2]. Solving the DER management problem helps a distribution system operator (DSO) to optimally balance the generation and consumption at the distribution level. However, the uncertain loads, the large amount of uncertain renewable generations, and the nonlinear characteristics of the underlying problem make this optimization problem extremely complex.

Integrated power transmission and distribution systems Chapter | 5

173

Traditionally the UC problem in the TS and DER management problem in the DS are solved separately. However, today’s power system is facing a lot of challenges (e.g. high pressure on the TS and reverse power flow from the DS to TS) mainly caused by DER generations in the DS, and solving them requires a synchronal attention from TSOs and DSOs. More specifically based on the projected RES growth [3], the DER generations in the DS are gradually reaching the levels in that their effects on the transmission network can no longer be neglected. Indeed, Ref. [4] shows that increasing the penetration of renewable distributed generations (RDGs) in the DS will lead to high pressure on the TS because the variance of the loads that the DS will impose on the TS will increase. Therefore, it will be demanding for the UC problem to additionally consider the impulsive variations of DS loads, rather than consider the DSs as fixed load points. In addition, DSs are growing to be more independent from the TS, and in some cases even produce more electricity than their local demands. From a technical perspective, the surplus electricity can be curtailed (e.g. by adjusting the blade angles in a wind farm) [5] or stored using energy storage technologies such as battery storage [6], elevated water [7] and compressed air [8]. From an operational perspective, renewable curtailment has a negative value because it wastes the investments [9] and may cause system disturbance. A more robust approach to handle the extra electricity in the DS is to sell it to the transmission network. Using an InTDS, the whole power system can coordinate the DSs by buying extra electricity from some of them and selling electricity to the others, which eventually may support the TS to reduce and stabilize the electrical generation and transmission. In an ideal scenario, this will lead to cost savings due to potential shut-down of some generators in the TS. Nevertheless, existing studies on such important coordination [10] lack in the literature and clearly a thorough research is demanded. There have been some attempts in the literature to integrate the transmission-level problems (e.g. unit commitment and economic dispatch) with the distribution-level problems [e.g. optimal power flow (OPF) and RDG planning]. Most of these attempts, however, simply assume that the online/offline status of each generator in the TS is given, which reduces the generation flexibility. For example, Refs. [11,12] focus only on the economic dispatch of the TS. A similar example happens in Ref. [13], which focuses on the distribution level’s dynamic pricing of electricity services. Simulation tools are also used to coordinate the two systems. For example, Ref. [14] demonstrates the need for and benefits of studying an InTDS through simulating such systems. Moreover, Ref. [15] emphasizes the increasing penetration of DERs in the distribution network, and proposes a combined simulation platform for studying the penetration of DERs in DSs. More recently Ref. [16] proposes a test system under high penetration of solar photovoltaic (PV) generations and such system can help to study existing problems like reverse power flows, while Ref. [17] proposes a model to operate

174

PART | I Technologies

the TS and DS in a coordinated manner. The proposed model, however, only considers the DC power flow of the DS and is solved using a Surrogate Lagrangian Relaxation approach. In this chapter, we propose mathematical formulations for an integrated UC and DER management problem that considers both the combinatorial UC decisions and the AC power flow characteristics of the DER management problem. The main contributions of this chapter can be summarized as follows: G

G

G

A two-stage stochastic programming formulation for the InTDS problem is proposed. This formulation includes uncertainties in both the transmission and distribution levels. The objective is to minimize the total cost of the two systems. We use sample average approximation to solve the problem with an empirical distribution corresponding to historical data to approximate the joint probability distribution of uncertainties. We strengthen our formulation by adding strong valid inequalities to the original InTDS problem. Note that the original InTDS problem leads to a complex mixed-integer linear programming (MILP) formulation that includes the complexities of both the TS and DS. These inequalities can speed up the solution process and make our method adaptable for largescale power system problems. Through extensive numerical results, we show that the InTDS problem will lead to significant cost savings in both the TS and DS as compared to the isolated ones. Moreover, we show that representing a DS as a fixed load point is a weak approximation of the loads in the DS, as our proposed InTDS model not only describes the distribution loads on the TS with superb accuracy but also enables the power system operators to tackle state-of-the-art problems in power system such as reverse power flow.

The remainder of this chapter is organized as follows. The mathematical model of the InTDS problem as well as its strengthened formulation are presented in Section 5.2. The numerical results are reported in Section 5.3. Finally we conclude this chapter in Section 5.4.

5.2

Mathematical model

Traditionally the UC problem and DER management problem are solved separately. In that case, the DS is referred to as ‘isolated distribution systems’ (IsDS), and the UC decisions of the TS is usually made by treating each IsDS as a net load point (in Fig. 5.3, DS1 and DS2 are traditional IsDSs). In this chapter, we explicitly represent the physical characteristics of each DS and integrate them with those of the TS when UC decisions are made. For such case, we call these types of DSs ‘integrated distribution system’ (InDS), and this type of power system, an InTDS. For example, in

Integrated power transmission and distribution systems Chapter | 5

175

FIGURE 5.3 A power system with DS1 and DS2 as IsDSs and DS3 as an InDS. DS, Distribution system; InDS, Integrated distribution system; IsDSs, Isolated distribution systems.

Fig. 5.3, DS3 is an InDS. In this section, we will introduce a two-stage stochastic InTDS model that includes both IsDSs and InDSs. In our model, we consider a power system layout similar to the US power sector, in which a TS is connected to multiple DSs as shown Fig. 5.3. The transmission and distribution networks are given by graphs (B, E) and ^ ; EÞ, ^ respectively. We use a mesh grid to represent the TS, and a radial ðN tree to represent the DS. Note that, throughout this chapter, we use the hat symbol, i.e. U^ , to represent the parameters and variables involved in the distribution level in order to distinguish them from those in the transmission level. In the following, we first introduce the two-stage integrated UC and economic dispatch model for the transmission level in Sections 5.2.1 and 5.2.2, then describe the details of the DS considered in the integrated model in Section 5.2.3 and finally present the strengthened MILP formulation in Section 5.2.4.

5.2.1

First-stage unit commitment model
 We let G Gb ; B ðBs Þ and E represent the set of generators (the set of generators at bus b), the set of buses (the set of buses with InDSs) and the set of transmission lines, respectively. The number of time intervals in the scheduling horizon is denoted by T. We use index k to represent a generator (also called ‘unit’) in G, index b to represent a bus in B and index e to represent a transmission line in E. The following table presents the parameters, random variables and decision variables of the unit commitment problem (Table 5.1). Thus the first-stage model of the UC problem can be represented as follows: ! T X X
 k k k k k k min f 5 SU ut 1 SD yt21 2 yt 1 ut 1 E½Qðy; uÞ; ð5:1aÞ y;u

kAG

t52

176

PART | I Technologies

TABLE 5.1 Notation for the UC problem. Parameters

The start-up cost of unit k.

k

The shut-down cost of unit k.

SU SD k

L /‘

k

The minimum-up/-down time limit of unit k.

k

G =G R

Description

k

k

k

The generation upper/lower bound when unit k is online. The start-up and shut-down ramp rate of unit k.

Rk

The regular ramp rate in the stable generation region of unit k.

F

Capacity of the transmission line (m, n) A E.

rt

The system reserve factor at time t.

Xe

Reactance of transmission line e.

τ

Binary indicator, 1 if an InDS is connected at bus b, 0 otherwise.

mn

b

Random variables

Description

b D1t

Uncertain net load at bus b A ℬ at time t when the DS connected to bus b is considered as a fixed load point.

Db

b

2tðξ^ 2t Þ

Uncertain load at bus b A ℬs at time t when the DS connected to bus b is explicitly represented.

b ξ^ 2t

Vector of uncertainty in compact form at DS b A ℬs at time t.

Decision variables

Description

ytk

Binary variable indicating if unit k is online (i.e. yt 5 1) or offline (i.e. yt 5 0) status.

utk

Binary variable indicating if unit k starts up (i.e. ut 5 1) or not (i.e. ut 5 0) at time t.

xtk

The amount of active power generation by unit k at time t.

θbt

The voltage phase angle at bus b at time t.

fte

The power flow on transmission line e at time t.

T X

s:t:

uki # ykt ;

’tA½Lk 11; TZ ; ’kAG;

ð5:1bÞ

i5t2Lk 11 T X i5t2‘ 11 k

uki # 1 2 ykt2‘k ;

’tA½‘k 11; TZ ; ’kAG;

ð5:1cÞ

Integrated power transmission and distribution systems Chapter | 5

2ykt21 1 ykt 2 ukt # 0 ykt21 Af0; 1g; ukt Af0; 1g;

’tA½2; TZ ; ’kAG;

’tA½1; TZ ; ’kAG; ’tA½2; TZ ; ’kAG;

177

ð5:1dÞ ð5:1eÞ ð5:1fÞ

where the objective function (5.1a) is to minimize the start-up and shutdown cost and the expected second-stage cost Q(y, u), which is described in detail in Section 5.2.2. Constraints (5.1b) and (5.1c) impose the minimumup/-down time limits for generator k, respectively. For instance, if generator k starts up at time t 2 Lk 1 1, then it has to stay online at least in Lk consecutive time periods till time t; if generator k shuts down at time t 2 ‘k 1 1, then it has to stay offline at least in ‘k consecutive time periods till time t. Constraints (5.1d) describe the relationship between binary variables y and u. Note that we use ½a; bZ to denote the set of integer numbers between a and b where a , b.

5.2.2

Second-stage economic dispatch model

After making the UC decisions in the first-stage (the day-ahead planning), each generator can be slightly adjusted to produce electricity to satisfy demand after the realization of the uncertainty (here-and-now decision). In the second-stage problem Q(y, u), we aim to minimize the operating costs of the TS that serves both IsDSs and InDSs: Qðy; uÞ 5 min

t XX


 hk xkt

ð5:2aÞ

kAG t51 k

s:t: G k ykt # xkt # G ykt ;  k
xkt 2 xkt21 # Rk ykt21 1 R 1 2 ykt21 ;

’tA½1; TZ ; ’kAG; ’tA½2; TZ ; ’kAG;

ð5:2bÞ ð5:2cÞ

k

xkt21 2 xkt # Rk ykt 1 R ð1 2 ykt Þ; ’tA½2; TZ ; ’kAG; ð5:2dÞ X X X xkt 1 fte 2 fte 5 Db1t 1 τ b Db ^ b ; ’bAB; ’tA½1; TZ ;

kAGb

eAE1 ðbÞ

eAE2 ðbÞ

2tðξ2t Þ

ð5:2eÞ n θm t 2 θt 1 fte 5 0; Xe

2F mn # fte # F mn ;

’e 5 ðm; nÞAE; ’tA½1; TZ ;

ð5:2fÞ

’eAE; ’tA½1; TZ ;

ð5:2gÞ

178

PART | I Technologies

X

k

G ykt $ ð1 1 rt Þ

kAG

X

ðDb1t 1 Db

bAB

Db

2tðξ^ 2t Þ b

b : 5 argmin Jðξ^ 2t Þ

’tA½1; TZ ;

ð5:2hÞ

’bABs ; ’tA½1; TZ ;

ð5:2iÞ

2tðξ^ 2t Þ b

Þ;

The operational cost for generator k at period t is denoted by hk ðxkt Þ, which is usually a quadratic nondecreasing function, i.e. hk ðxkt Þ 5 ck ðxkt Þ2 1 bk ðxkt Þ 1 ak ykt , that can be approximated by a piecewise linear function [18]. Constraints (5.2b) describe the generation upper/lower bounds of each unit, constraints (5.2c) [resp. (5.2d)] limit the ramping-up (resp. ramping-down) capacity of each unit, constraints (5.2e) enforce demand balance and transmission line capacity is restricted by constraints (5.2f) and (5.2g). Constraints (5.2h) restrict the system spinning reserve requirement. Note that, there are two types of loads involved in constraints (5.2e): 1. Db1t represents the IsDS load at period t at bus b, which is commonly considered in the literature. In this case, the ‘net load’ (Db1t ) is the raw electricity demand at each bus minus the renewable generations at this bus. 2. Db ^ b represents the InDS load connected at bus bABs at period t. It is 2tðξ 2t Þ

obtained by solving a DS’s DER management problem in constraints (5.2i). Therefore, constraints (5.2i) indicate that the amount of energy that each InDS purchased from the TS must be determined based on the economics and characteristics of the DS’s DER management problem J that we will describe in the next subsection.

5.2.3

Distributed energy resource management problem

We use a radial network (i.e. a common topology of distribution networks) ^ ^ ^ 5 ðN to model the DS. The radial network is modelled as a tree G:   ; EÞ, as ^ ^  shown in Fig. 5.4. Here, N denotes the set of subbuses and E  is the

FIGURE 5.4 A radial network.

Integrated power transmission and distribution systems Chapter | 5

179

^ The radial network is connected to cardinality of the set of power lines E. the TS via a distribution substation (i.e. PCC) at which the DS connects to one of the buses of the transmission network. In our model, DERs including ^ DN ^ ), (1) dispatchable active power generation units (collected in set N 1 such as microturbines (MTs), (2) reactive power generation units (collected ^ DN ^ ), and (3) RDG units (collected in set N ^ DN ^ ), such as wind in set N 2 r farms and solar farms, are connected to the distribution network. The loads in the distribution network are first satisfied by the power internally generated from the DERs, and when the internal power supply is not enough, the system will buy power from the TS via the PCC. For each InDS located at bus b of the TS, the objective of DER management problem is to minimize the costs of power generated by the dispatchable DERs and power purchased from the TS, while respecting physical constraints such as OPF constraints. Without loss of generality we omit the subscript b from our DER management formulation. We use random variable s^nt to represent the active power output of the RDG located at subbus n at pn qn period t, and random variable d^t (resp. d^t ) to represent the real (resp. reactive) power demand of subbus n at period t. Moreover, a short description of the other parameters and decision variables is presented in the following table (Table 5.2):  Thus given the actual realization of uncertainty ξ^ 2t : 5 s^1t . . .; s^nt ;  p1 pn q1 qn d^t ; . . .; d^t ; d^t ; . . .; d^t at period t, the minimum operating cost of the DS is obtained by solving the following second-order cone programming (SOCP) problem: X X J 5 min c^pt p^0t 1 c^qt q^0t 1 ð5:3aÞ c^nf p^nt 1 c^ne ω^ p^nt ^ nAN 1

s:t: p^nt # p^nt # p^ t ; n

^ nAN 1

^ ’nAN 1

n ^ qt ; ’nAN q^ nt # q^nt # b 2

t z^t AZ s^t ; d^ ; p^t ;

ð5:3bÞ ð5:3cÞ ð5:3dÞ

In the objective function (5.3a), c^pt p^0t 1 c^qt p^0t is the payment to the TS under the net metering rule. The active power purchased from the TS, i.e. p^0t , is the active power exchange between the TS and InDS (its optimal
n value  is n b b ^ assigned to the TS problem as: 5 D ). Moreover, the pairs p ; p and t ^
n n t 2tðξ2t Þ b qt ; q^t , in (5.3b) and (5.3c), are the upper and lower bounds on the dispatchable DERs and reactive DERs, respectively. We use z^t to aggregate all t the decision variables at period t and use Zð^st ; d^ ; p^t Þ in (5.3d) to denote the set of all the OPF constraints. The OPF problem (5.3d) is nonconvex in general, but recently Ref. [19] shows that when the network is radial, the OPF problem can be exactly convexified to a SOCP problem in a branch flow

180

PART | I Technologies

TABLE 5.2 Notation for the DER management problem. Parameters

Description

p q c^ t = c^ t

Electricity price of purchasing active/reactive power from the TS at period t.

c^ nf = c^ ne

Fuel/emission cost for the dispatchable DGs at subbus n.

ω^

Emission factor of the dispatchable DGs (kg/kWh).

^ ℜ

mn

C^

^ =X

mn

mn

Electrical resistance/reactance of distribution line (m, n) of DS. Electrical capacity of distribution line (m, n) of DS.

n δ^ r

Binary indicator if subbus n has a RDG unit.

n δ^ 1

Binary indicator if subbus n has a dispatchable unit.

n δ^ 2

Binary indicator if subbus n has a reactive power source.

^ u

Number of pieces in the polyhedral e-approximation of the second-order cone programming (SOCP) constraint.

^ v^ v=

The upper/lower bound of voltage magnitude distribution level of DS.

Decision variables

Description

p^ 0t =q^ 0t

Active/reactive power purchased from the TS to the DS at subbus 0 at period t.

mn ^ mn P^ t = Q t

Active/reactive power flow of the DS from subbuses m to n at period t.

n V^ t =v^ nt

Complex voltage of the DS at subbus n at period t/its magnitude.

^I mn =‘^ mn t t

Complex current of the DS from subbuses m to n at period t/its magnitude.

p^ nt

Active power output of the dispatchable DG unit at subbus n at period t.

q^ nt

Reactive power output of the reactive DERs at subbus n at period t.

model. Hence, we use the branch flow model to construct the constraint set t Zð^st ; d^ ; p^t Þ as follows: G

The real power balance equation at PCC or subbus 0 at period t is p^0t 5

X ^ nAN 0

0n P^t ;

ð5:4aÞ

Integrated power transmission and distribution systems Chapter | 5

181

^ is the set of subbuses connected to PCC (or subbus 0). Similarly where N 0 reactive power balance equation at PCC or subbus 0 at period t is X 0n Q^ t : q^0t 5 ð5:4bÞ ^ nAN 0 G

From the Kirchhoff current law (active and reactive nodal balance), we have the following two sets of constraints: ^ \f0g at period t are 1. The real power balance equations for all nAN mn ^ mn ‘^ mn 5 d^pn 2 δ^ n sn 2 δ^ n p^n 1 Σ ^ P^nl ; P^t 2 ℜ r t 1 t t t t lAN

ð5:4cÞ

n

^ \f0g at period t are 2. The reactive power balance equations for all nAN mn ^ mn ‘^ mn 5 d^pn 2 δ^ n q^n 1 Σ ^ Q^ nl ; Q^ t 2 X t t 2 t t lAN

ð5:4dÞ

n

G

The voltage drop on all lines ðm; nÞAE^ at period t is n ^ mn Þ2 Þ‘^ mn ; ^ mn P^mn 1 2X^ mn Q^ mn 2 ððℜ ^ mn Þ2 1 ðX v^m t t t 2 v^t 5 2ℜ t

^m 2 v^m t : 5 jV t j

G

represents the voltage of subbus n and where represents the current on distribution line (m, n). The branch power flow constraint shown in Ref. [19] is h i  ^mn ^ mn m ^ mn  ^ mn  2Pt ; 2Qt ; v^t 2 ‘t  # v^m t 1 ‘t ;

ð5:4eÞ

mn ‘^ t : 5 jItmn j2

ð5:4fÞ

2

for all lines ðm; nÞAE^ and all periods tA½1; TT . As solving the SOCP problems is computationally expensive, in order to solve the problem efficiently, we linearize the SOCP constraints. Base on the method introduced in Ref. [20], we build a polyhedral e-approximation of the above cone. We first represent a conic quadratic constraint (4f) by a system of conic quadratic constraints. To do so, we introduce an auxiliary vari^ mn , which is set as zero, to have an even number of variables inside able Y 3t the L2-norm function as follows: h i  ^mn ^ mn m ^ mn ^ mn  ^ mn  2Pt ; 2Qt ; v^t 2 ‘t ; Y3t  # v^m t 1 ‘t : 2

^ mn , we represent the above ^ mn and Y By further introducing variables Y 1t 2t conic quadratic constraint in the form of a system of conic quadratic constraints: rhffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi i h mn i2ffi mn 2 ^ mn ^ 2Pt #Y 1 2Q^ t ð5:4gaÞ 1t rhffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi i h mn i2ffi mn 2 ^ mn ^ #Y 1 Y v^m 2 ‘^ t

t

3t

2t

ð5:4gbÞ

182

PART | I Technologies

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi h mn i2 h mn i2 mn ^ ^ # v^m 2 ‘^ 1 Y Y 1t

2t

t

t

ð5:4gcÞ

We next reformulate the constraints above to a system of linear constraints. According to Ref. [20], the constraints (5.4ga), (5.4gb) and (5.4gc) can be approximated by the following set of linear constraints:   8  mn  0 2P^ ; η^ 0 $ v^m 2 ‘^ mn ; > ^ $ ε > t 1 t 1 t > <  mn   mn  ^    ð5:4haÞ η^ 01 $ 2Q^ t ; ε^ 02 $ Y 3t ; >     > mn mn > : ε^ 0 $ Y ^ ; η^ 0 $ Y ^ ; 3 1t 2t 3

8 π π j j21 > > ε^ 1 5 cos j11 ε^ 1 1 sin j11 η^ 1j21 ; > > 2 2 > > >  

> >   > π π j21 j21 j21  >  > ^ ^ ^ η $ 2 sin 1 cos ε η > 1 1 1 ;  j11 j11 > > 2 2 > > >

> > π π > j j21 j21 > > > ε^ 2 5 cos 2j11 ε^ 2 1 sin 2j11 η^ 2 ; <   j 5 1; 2; . . .; u^ ;

  > j21 >  2 sin π ε^ j21 1 cos π η^ j21 ; > ^ η $ > 2 2 2   j11 j11 > 2 2 > > >

> > > π π > > ε^ 3j 5 cos j11 ε^ 3j21 1 sin j11 η^ 3j21 ; > > 2 2 > > >  

> >   > π π > j21 j21 j21   > ^ ^ ^ η $ 2 sin 1 cos ε η > 3 3 ;  j11 j11 : 3 2 2

8 > u^ ^ mn ; η^ u^ # tan π ε^ u^ ; > ^ ε # Y > 1 1t 1 > > 2v11 1 > > >

> < u^ ^ mn ; η^ u^ # tan π ε^ u^ ; ^ε2 # Y 2t 2 2v11 2 > > >

> > π mn > u^ u^ u^ m > ^ > > ε^ 3 # v^t 1 ‘t ; η^ 3 # tan v11 ε^ 3 ; : 2 G

ð5:4hbÞ

ð5:4hcÞ

^ \f0gÞ is The voltage limit on bus nðnAN ^ v^ # v^nt # v;

ð5:4iÞ ^ 2 2 ^ 5 jVj and v: ^ 5 jVj are the upper and lower bounds on the where v: voltage magnitude of the distribution network.

Integrated power transmission and distribution systems Chapter | 5 G

183

The line capacity on all lines ðm; nÞAE^ is limited by mn mn mn ðP^t Þ2 1 ðQ^ t Þ2 # ðC^ t Þ2 :

ð5:4jÞ

Using the quadratic constraint linearization method in Ref. [21], the circle shape constraint (5.4j) is replaced with a number of linear constraints (intersection of two squares) as follows: 2 C^ 2 C^

mn

mn mn # P^t # C^ ;

ð5:4kaÞ

mn

mn mn # Q^ t # C^ ;

ð5:4kbÞ

pffiffiffi mn pffiffiffi mn mn mn 2 2C^ # P^t 1 Q^ t # 2C^ ;

ð5:4kcÞ

pffiffiffi mn pffiffiffi mn mn mn 2 2C^ # P^t 2 Q^ t # 2C^ :

ð5:4kdÞ

As mentioned before, the isolated UC problem is already very difficult to solve. Integrating the DER management problem with the UC problem leads to even more challenges as the size of the problem increases dramatically. Therefore, we introduce a tighter reformulation of the UC problem to speed up the solution process.

5.2.4

Tighter formulations

We introduce a strengthened MILP formulation for the InTDS problem. By investigating the structure of constraints (5.1b)
(5.1d) and (5.2b)
(5.2d), we derive strong valid inequalities that can help reduce the computational time of the InTDS problem. We present three of such inequalities that show a general overview of these types of inequalities. Similar strong inequalities can be derived to shorten the computational time of solving difficult UC problems. Without loss of generality, we drop the index k for the generators, as we investigate each individual generator. For any two consecutive time periods of the InTDS problem, when L 5 ‘ 5 1, and G 2 G 2 R $ 0, the generation amount for each time period can be tightened by the following constraint considering the generation upper bounds and ramp rate limits. xt # Ryt 1 ðG 2 RÞðyt11 2 ut11 Þ; ’tA½1; T 21Z :

ð5:5Þ

The validity of (5.5) can be verified easily. If yt 5 0 then xt 5 0, so (5.5) becomes yt11 $ ut11, which is valid because of (5.1b). If yt 5 1, then ut11 5 0 due to (5.1c) and one of the following two cases happens: (Case 1) xt # G if yt11 5 1, or (Case 2) xt # R if yt11 5 0. Both cases are valid due to constraints (5.2b) and (5.2c).

184

PART | I Technologies

Moreover, under the same set of conditions, that is, when L 5 ‘ 5 1, and G 2 G 2 R $ 0, the following set of tighter ramping constraints can provide a tighter MILP formulation for the whole problem. xt11 2 xt # ðG 1 RÞyt11 2 Gyt 2 ðG 1 R 2 RÞut11 ; ’tA½1; T 22Z ;

ð5:6aÞ

xt 2 xt11 # Ryt 2 ðR 2 RÞyt11 2 ðG 1 R 2 RÞut11 ; ’tA½1; T 22Z :

ð5:6bÞ

To verify the validity of the above inequalities, there are a few scenarios that can happen: G

G

G

G

if yt 5 0 and yt11 5 0, then clearly ut11 5 0; therefore, xt and xt11 should be zero. if yt 5 0 and yt11 5 1, then clearly xt 5 0 and ut11 5 1; therefore, xt11 # R and xt11 $ G. if yt 5 1 and yt11 5 1, then clearly ut11 5 0; therefore, xt11 2 xt # R and xt 2 xt11 # R. if yt 5 1 and yt11 5 0, then clearly xt11 5 0, ut11 5 0; therefore, xt # R and xt $ G.

For any three consecutive time periods, when L 5 ‘ 5 1, and G 2 G 2 2R $ 0, we use the following three-period ramping constraint to strengthen the MILP formulation. xt 2 xt11 1 xt12 # Ryt 2 ðR 2 RÞyt11 1 Ryt12 1 ðG 2 RÞðyt12 2 ut12 2 ut11 Þ;

’tA½1; T 23Z :

ð5:7Þ

The above inequalities can be extended to include four or more continuous variables. Also similar inequalities can be obtained at other settings such as when L 5 ‘ 5 2 and G 2 G 2 R $ 0, as shown in [22]. Note that this tight formulation is necessary to solve the large-scale InTDS problem, especially when the parameter settings are in a way that the feasible region is too large or when there are highly intermittent renewable generations and highly uncertain DS loads.

5.3

Numerical results

In this section, we report the computational results for InTDS problem on an IEEE six-bus system, and on an IEEE 118-bus system, based on the one given online at http://motor.ece.iit.edu/data/. Each of them connects to multiple isolated distribution systems (IsDSs), which are simply considered as load points though. In our tests, we replace some of these IsDSs with InDSs. All of the InDSs are based on the IEEE 33-bus radial system available on the web at [23], and their components have been scaled to different degrees in order to be compatible with the loads on TS. The operation interval was set to 24 hours (i.e. T 5 24). The system spinning reserve amount was set to

Integrated power transmission and distribution systems Chapter | 5

185

5% of the total load for each time period. All numerical tests were implemented on a computer with an Intel Core i7
7700 CPU and 16 GB memory. All optimization problems were solved by MOSEK solver on CVX software. Before analyzing the InTDS results in Section 5.3.3, we describe the characteristics of the isolated UC problem and isolated DER management problem, and report their operational results in Sections 5.3.1 and 5.3.2, respectively. Note that we mention them (1) to show the sensitivity of UC decisions with respect to the loads, (2) to show the decline in the capacity of the transmission networks as a result of high penetration of renewables, and (3) to compare them with our proposed InTDS method. It follows that the importance of developing InTDS is emphasized. Finally we show the InTDS results for a 118-bus system in the presence of the valid inequalities that we introduced in Section 5.2.4 in order to show the effectiveness of the proposed model for large-scale problems.

5.3.1

Isolated unit commitment problem

The original IEEE six-bus system, as shown in Fig. 5.5, includes three generators (at buses 1, 2 and 6, respectively) and three IsDS load points (at buses 3, 4 and 5, respectively). The raw load data were obtained from http:// motor.ece.iit.edu/data/6bus_Hourly_Data.xls, which also includes hourly wind generation coefficients. In addition, the hourly solar generation coefficients were obtained from Ref. [24]. Moreover, the characteristics of the units and transmission lines are described in Tables 5.3 and 5.4, respectively. In Table 5.5, we report the results under three different types of loads. First, we explain each column of the table. The column ‘Net load’ shows

FIGURE 5.5 The IEEE six-bus system.

186

PART | I Technologies

TABLE 5.3 Units data. k

Gk (MW)

G (MW)

Ramp Lk ðhÞ ‘k ðhÞ (MW/h)

50

50

200

55

4

200

100

5

100

50

0

0

5

40

20

Unit k

SUk ($)

SDk ($)

G1

100

G2 G3

Cost coefficients k

a ($)

bk ($/MW)

ck ($/MW2)

4

177

13.5

0.00045

3

2

130

40

0.001

1

1

137

17.7

0.005

TABLE 5.4 Transmission line data. Line No.

From bus

To bus

X (pu)

Flow capacity (MW)

1

1

2

0.170

200

2

2

3

0.037

100

3

1

4

0.258

100

4

2

4

0.197

100

5

4

5

0.037

100

6

5

6

0.140

100

7

3

6

0.018

100

each type of loads that are used in the isolated UC problem: (1) ‘No Gen.’ describes raw loads with no renewable generation penetrated, as shown in Fig. 5.6; (2) ‘Wind Gen.’ describes net loads with wind generation penetrated, as shown in Fig. 5.7 and (3) ‘Solar Gen.’ describes net loads with solar generation penetrated, as shown in Fig. 5.8. The total net load during 24 hours is denoted by ‘TNL’. The column ‘Time(s)’ represents the CPU times for the UC problem in seconds and the column ‘Total cost($)’ represents the optimal total cost of the UC problem. The last column shows the online/offline status of the three generators. Next, we describe the details of the three types of loads. We start with the raw load data in Fig. 5.6, which shows different load scenarios highlighted in different line trends. The load ranges in the data set are [50
130 MW] at bus 3, [30
80 MW] at bus 4, and [20
50 MW] at bus 5, and the overall load pattern does not change much between the different scenarios. There are two peaks in the daily loads, one in the morning at around 8
10 a.m., and the other around 17
21 p.m. Therefore, the UC problem can

TABLE 5.5 The unit commitment decisions for isolated TS. Row

Net load

Time (s)

Total cost ($)

Unit commitment status t 5 1: 24

1

No Gen. TNL 5 4.5 GW/day

1.1

67,081.7

G1:

111111111111111111111111

G2:

000000000000000000000000

G3:

000000001111111111111110

G1:

111111111111111111111111

G2:

000000000000000000000000

G3:

000000000000000000010000

2

Wind Gen. TNL 5 3.5 GW/day

1.6

3

Wind Gen. TNL . 4.2 GW/day







4

Solar Gen. TNL 5 4.0 GW/day

1.7

62,117.8

G1:

111111111111111111111111

G2:

000000000000000000000000

G3:

000000001100000001111110

Solar Gen. TNL . 4.0 GW/day



5

TNL, Total net load; ‘   ’ indicates no feasible solution.

50,714.5





188

PART | I Technologies

FIGURE 5.6 Hourly raw loads of buses 3, 4 and 5 for five different scenarios in five line trends.

FIGURE 5.7 Net loads including wind generations.

be solved relatively easily over the predicted demands to obtain the generator status and generation amounts. The first row of the Table 5.5 shows results for the case of raw loads; the total cost is $67,081, and a total of 4.5 GW electricity is scheduled for the next 24 hours. Note that in Table 5.5, ‘ ’ indicates no feasible solution. Nowadays, with more renewable penetration, the net load that should be satisfied by traditional generators changes, as compared to the raw load

Integrated power transmission and distribution systems Chapter | 5

189

FIGURE 5.8 Net loads including solar generations.

shown in Fig. 5.6. This can be seen in Fig. 5.7 (resp. Fig. 5.8) where a largescale 50 MW wind farm (resp. solar farm) and a medium-scale 25 MW wind farm (resp. solar farm) are connected at buses 3 and 4. There is no renewable source at bus 5. These plots demonstrate how uncertain wind (or solar) generations at buses 3 and 4 would distort the smooth and predictable trend of raw load. Bus 3 has an even more disrupted load structure because it has a larger solar (or wind) generation capacity integrated, as compared with bus 4. From the results in Table 5.5, we can observe that following such complex trend of net load is very troublesome. In Table 5.5, the results in the fourth row show that the TSO can shut down G3 for majority of the day when there is sunlight. This is because there is a deep convex curve in the middle of the day in Fig. 5.8 as the solar generations reach their peak. By comparing the results in the second and fourth rows, we can observe that wind farms lead to more substantial cost savings than the same size solar farms because wind generation exists throughout the day. Furthermore, the third and fifth rows of Table 5.5 show that integration of solar and wind generation puts enormous pressure on the TS that is not designed to handle such variations. Indeed, when we scale up the total net load to more than 4.2 GW/day, the UC problem becomes infeasible, while in the case of raw demands the TS was able to supply 4.5 GW/day. This indicates a huge reduction (about 7%) of the ability of TS to satisfy demand as we integrate more renewable generations. Therefore, the load information is very conclusive for the optimal UC decisions, and having more accurate load information at each DS will be extremely valuable.

190

5.3.2

PART | I Technologies

Isolated distributed energy resource management problem

In this section, we briefly present the physical characteristics of the original IEEE 33-bus radial DS, and use a simple example to illustrate the dynamics and complexity of the decisions made by the DSOs. The IEEE 33-bus radial system has 33 buses which are referred to as subbuses in this chapter in order to avoid confusion with the buses of the TS. This DS uses electricity purchased from the TS, active power generated by MTs at subbuses 15 and 29, reactive power sources at subbuses 11, 13 and 32 and three RDGs located at subbuses 17, 24 and 32 to satisfy the uncertain active and reactive demands of the subbuses. The uncertain loads were obtained from historical load data available from the Pecan street project [25], which provides access to a large set of real electricity usage data for academic use. Meanwhile, we use historical data from Ref. [26] for the RDG data. For instance, the historical load data for subbuses 15, 18, 21 and 30 are depicted in Fig. 5.9, respectively, where each one dashed line represents a single scenario. In each scenario, for each hour of the day (T 5 24), there are 67 uncertain parameters, including 3 uncertain wind p1


15 29  p32 generation outputs s^t ; s^t , 32 uncertain active demands d^t ; . . .d^t q1

q32 and 32 reactive demands d^t ; . . .; d^t . Depending on the amount of uncertain generations inside the DS and the uncertain generations by the RDGs, the demand of InDS from the TS will be obtained by solving the DER management model presented in Section 5.2.3.

FIGURE 5.9 Hourly loads for four subbuses of the DS problem. Different line trends represent different scenarios. DS, Distribution system.

Integrated power transmission and distribution systems Chapter | 5

191

Table 5.6 shows the optimal DER management results for the IEEE 33-bus system described above at time t 5 20 by considering five possible scenarios. The optimal values for the active and reactive power purchase from the TS are labelled as ‘Power purchase’, optimal power production by the MTs at bus 29 and 15 are labelled as ‘MT generation’, and the optimal cost of such decisions for a small-size IsDS are labelled as ‘Optimal cost’. Furthermore, Fig. 5.10 depicts the active power purchased by the InDS from the TS during a 24-hour period, and it is considered as load for the TS. In this figure, each one dashed line represents the active power corresponding to a single scenario of the DS. Note that such load pattern is completely

TABLE 5.6 Optimal DER management decisions. Scenarios

Power purchase (MWh)

MT generation (MWh)

Optimal cost ($)

Active

Reactive

Bus 15

Bus 29

1

1.79

1.72

0.57

1.03

1246

2

2.09

1.61

1.51

0.85

1765

3

1.78

1.74

1.12

0.95

1538

4

1.72

1.64

0.30

0.66

835

5

1.65

1.63

0.49

0.28

705

DER, Distributed energy resource.

FIGURE 5.10 Active power purchased by the distribution system under five different scenarios.

192

PART | I Technologies

different from the simplified raw load when the DS is considered as a single load point, which follows the two-peak pattern and is predictable as shown in Figs 5.6
5.8. The reason is that there are a lot of factors involved in the optimal planning of the DS. Such factors include the amount and location of renewable generations, and the location and intensity of uncertain loads. For example, in a scenario where there is a high intensity of DS loads concentrated far away from the subbus 0, perhaps, it is more economical to generate more electricity by the MT closer to those loads than to purchase from the TS. Such factors make it clear that a lot of details in the DS are missed when the TSOs make decisions, and further show how valuable it is to develop the InTDS model that accounts for DER management problem while solving the UC problem.

5.3.3

Integrated transmission and distribution systems

In this section, we first show the results for an integrated transmission and distribution problem, and the advantages of combining the two levels. Next, by sensitivity analyses, we show the effect of exact InDS loads on UC decisions. Moreover, we demonstrate how the power system scheduling changes under high penetration of RDGs in an integrated scheme. To begin with, we report the results for three configurations of IEEE sixbus system in Table 5.7. The configurations are the same as that in Section 5.3.1 except that the load at bus 5 is replaced with an InDS. This InDS is based on the IEEE 33-Bus network and it is scaled up to have a load range of [40
80 MW]. We compare the results in two modes: the integrated mode and the isolated mode. For the integrated mode, the integrated system runs over five global scenarios of uncertainties that include uncertainties in both the TS and DS. Each scenario has 69 elements in the integrated mode. These scenarios include two transmission-level uncertain loads at buses 3 and

TABLE 5.7 Integrated mode versus isolated mode. Configuration

TS cost ($)

DS cost ($)

Total cost ($)

Isolated

70,509.62

30,973.08

101,482.7

Integrated

70,358.64

29,126.01

99,484.65

Wind Gen.

Isolated

54,661.76

32,702.94

87,364.7

Integrated

54,503.58

29,854.16

84,357.74

Solar Gen.

Isolated

65,913.51

32,702.94

98,616.46

Integrated

65,910.81

29,854.16

95,764.97

No Gen.

DS, Distribution system; TS, transmission system.

Integrated power transmission and distribution systems Chapter | 5

193


 4 μTS 5 D31t ; D41t , and 67 uncertain parameters of the InDS connected to bus

29 ^p1 ^p32 ^q1 ^q32 . In the isolated mode, we will 5 μDS 5 s^15 t ; s^t ; d t ; . . .; d t ; d t ; . . .; d t divide the uncertainty vector into two parts of the TS uncertainties ðμTS Þ and the DS uncertainties ðμDS Þ. We use the DS uncertainties ðμDS Þ to run the isolated DER management problem to obtain the loads that this DS will put on the TS under each scenario. Then, we run the isolated UC problem by using these loads to represent loads at bus 5, and μTS to represent loads at buses 3 and 4. We solve the isolated UC problem and isolated DER management problem over 5 scenarios with all 120 possible alternatives (note that there are 5! 5 120 possibilities of permutating the InDS load scenarios with TS load scenarios) and report the average results. As shown in Table 5.7, in the integrated mode, there are significant savings in both UC decisions of the TS (labelled as ‘TS cost’) and the DER management decisions of the DS (labelled as ‘DS cost’). Additionally the total costs of UC problem and DER management problem reduce by 2%, 2.9% and 3.5% for the configurations of No Gen., Solar Gen. and Wind Gen, respectively. Thus applying this model to a real-world power system could lead to millions of dollars in terms of savings. Note that here only one of the load buses (i.e. bus 5) is explicitly represented as an InDS that considers the DER management problem. To show more general insights, we also report the results for the case when all three load buses are replaced with the InDSs in Table 5.9. Here, the IsDS loads at buses 3, 4 and 5 are replaced with integrated distribution system 1 (InD-1), integrated distribution system 2 (InD-2) and integrated distribution system 3 (InD-3), respectively. Each of these InDSs are based on the IEEE 33-bus radial network, as described in Section 5.3.2, and are modified as shown in Table 5.8. From Table 5.9 the integrated mode leads to significantly lower total costs than the isolated mode does, which is because of the coordination between the TS and DS. Moreover, when the InDS has a larger capacity (e.g. InD-1), the savings are even more significant. Note that all the InDSs that we considered are the case when the DS needs to purchase large

TABLE 5.8 The characteristics of the three integrated distribution systems. Distribution network

RDG locations

MT locations

Load on TS (MW)

InD-1

14

28




9

29

60
120

InD-2

17

24

32

15

29

40
80

InD-3

17

24

30

15

5

30
70

InD, Integrated distribution system; RDG, Renewable distributed generation; TS, transmission system.

194

PART | I Technologies

TABLE 5.9 Integrated mode versus isolated mode. Configuration

TS cost ($)

InD-1 cost ($)

InD-2 cost ($)

InD-3 cost ($)

Total cost ($)

Integrated

62,752.2

31,670.04

23,914.82

21,844.51

140,181.6

Isolated

63,640.92

36,356.49

27,044.8

22,326.12

149,368.3

InD, Integrated distribution system; TS, transmission system.

FIGURE 5.11 Hourly loads given by the needy, balanced and rich InDS on the transmission system. InDS, Integrated distribution system.

amounts of electricity from the TS. In fact, in the modern TSs with high penetration of renewable generation, the DSs are growing to be more independent, and excessive renewable generation during certain hours is becoming a daily problem for some DSs. Instead of curtailing the extra energy or saving it using storage units, both of which are costly, a DS can sell the extra electricity to the TS and turn it into profit for the DS with efficient modelling tool such as our proposed InTDS model. We will demonstrate this advantage of our model in the coming Section 5.3.3.1.

5.3.3.1 Sensitivity analyses: types of the integrated distribution systems The trend toward higher penetration of RDGs in the DSs is posing new challenges on the operation of DSs, and further the operation of the whole power system. Fig. 5.11 shows the development of the DSs from the needy ones

Integrated power transmission and distribution systems Chapter | 5

195

towards the rich ones. A ‘Needy InDS’ needs to purchase electricity from the TS in bulk amounts, however, in a ‘Balanced InDS’ (resp. a ‘Rich InDS’), the DS will face the problem of excessive electricity in some (resp. most) of its operation time. Instead of dealing with excessive electricity with costly methods, our proposed model can provide operational decisions with the InTDS model at no additional cost. The integrated transmission and distribution model will allow the InDS to serve as a generator when it has excessive electricity and support the operations of both the TS and DS. The effect of a Needy, Balanced and Rich InDS on the unit commitment decisions is reported in Table 5.10. We can observe that ‘Balanced InDS’ and ‘Rich InDS’ incur lower unit commitment costs than the ‘Needy InDS’. The reason is that InTDS model allows the excessive electricity at bus 2 that occurs with ‘Balanced InDS’ and ‘Rich InDS’ to be utilized in the power system. Indeed, a Rich InDS allows the TSOs to shut down many of their generators in TS at some hours.

5.3.4

IEEE 118-bus network results

Here, we use a modified IEEE 118-bus TS, integrated with numerous InDSs based on the IEEE 33-bus system to study the large-scale InTDS problem and the effect of the valid inequalities (or called ‘cuts’) proposed in Section 5.2.4. The 118-bus TS, which is available at http://motor.ece.iit.edu/ data/SCUC_118/, includes 186 transmission lines, 54 generators and 91 load buses. The results for the IEEE 118-bus system with 81 IsDS, 10 InDS and 54 generators under five scenarios are provided in Table 5.11. The InDSs were obtained based on the IEEE 33-bus system, and scaled up proportionally to replace the loads at buses 18, 23, 31, 43, 49, 58, 74, 79, 86, and 101. We run the test for different ramp levels. We can see that the UC problem for a large power system can be challenging to solve under the case ‘No cuts’, but we can reduce the computational time dramatically by adding only some valid inequalities under the case ‘With cuts’. Moreover, we increase the number of scenarios to be 20 and reports the results in Table 5.12. This table also includes the number of variables and constraints for the InTDS with/without the valid inequalities. Observe that the number of constraints and the number of variables slightly increase when we add the inequalities, while contrarily the problem is solved faster with cuts as they lead to a tighter MILP formulation. From Tables 5.11 and 5.12, the integrated model performs essentially superb when it was combined with the valid inequalities. Note that, we used only three valid inequalities to demonstrate the efficiency of solving such problems for the real power system setting, and it is possible to derive many other valid inequalities that could lead to even more significant reduction of the computational time.

TABLE 5.10 The effect of the InDS on the TS. InDS type

CPU (s)

UC cost ($)

Unit commitment status t 5 1: 24

Needy InDS

1.1

72,839.1

G1:

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

G2:

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

1

1

1

0

0

0

Balanced InDS

1.1

Rich InDS

1

67,544.6

60,763.1

G3:

0

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

G1:

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

G2:

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

G3:

0

0

0

0

0

0

0

0

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

G1:

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

G2:

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

G3:

0

0

0

0

0

0

0

1

1

1

1

1

0

0

1

1

1

1

1

1

1

1

1

0

InDS, Integrated distribution system; TS, transmission system; UC, unit commitment.

Integrated power transmission and distribution systems Chapter | 5

197

TABLE 5.11 The unit commitment results summary (five scenarios). Case

Ramp level

Optimal value ($)

CPU time (s)

No cuts

High

1,047,090

2523

Medium

1,045,180

2602

With cuts

Low

1,042,900

1245

High

1,047,090

2088

Medium

1,045,180

2371

Low

1,042,900

985

TABLE 5.12 The unit commitment results summary (20 scenarios). Case

Ramp level

Optimal value ($)

CPU time (s)

Continuous variables

Binary variables

Constraints

No cuts

High

1.0602 3 106

6871

320,968

2538

292,510

Low

1.0517 3 106

3943

320,968

2538

292,510

High

1.0602 3 10

6

6026

337,688

2538

309,230

Low

1.0517 3 106

2624

337,688

2538

309,230

With cuts

5.4

Conclusions

In this chapter, we proposed an integrated model for the unit commitment problem in a TS combined with the DER management problem in DSs. Through numerical results, we showed that the InTDS leads to significant cost savings compared to the total costs of the isolated systems solved separately. In addition, we showed that by using the integrated model, we are able to include the characteristics of the DS into the UC analyses of the TS, which performs much better than the case considering the DS as a single load point. In addition, the integrated model allows to include the characteristics of the TS (and other DSs connected to the TS) into the DER management analyses of each DS, which leads to more smooth operation of the power system and solves the issue of excessive power generations in DS by turning them into profit. Finally we demonstrated that it is possible to tighten the proposed MILP formulation by adding valid inequalities for the problem. The computational results verified the efficiency of our proposed integrated model combined with the valid inequalities.

198

PART | I Technologies

References [1] Global renewable generation continues its strong growth, new irena capacity data shows. ,https://www.irena.org/newsroom/pressreleases/2018/Apr/Global-Renewable-GenerationContinues-its-Strong-Growth-New-IRENA-Capacity-Data-Shows.. [2] Ackermann T, Knyazkin V. Interaction between distributed generation and the distribution network: operation aspects. In: IEEE/PES transmission and distribution conference and exhibition, vol. 2. IEEE; 2002. p. 1357
62. [3] Barbose G. US renewables portfolio standards: 2017 annual status report. Berkeley, CA: Lawrence Berkeley National Lab. (LBNL), Technical Report; 2017. [4] Fathabad AM, Cheng J, Pan K, Qiu F. Data-driven planning for renewable distributed generation integration. IEEE Trans Power Syst, Early Access, https://doi.org/10.1109/ TPWRS.2020.3001235, 2020. [5] Bird L, Lew D, Milligan M, Carlini EM, Estanqueiro A, Flynn D, et al. Wind and solar energy curtailment: a review of international experience. Renew Sustain Energy Rev 2016;65:577
86. [6] Root C, Presume H, Proudfoot D, Willis L, Masiello R. Using battery energy storage to reduce renewable resource curtailment. In: 2017 IEEE power and energy society innovative smart grid technologies conference (ISGT). IEEE; 2017. p. 1
5. [7] Jiang R, Wang J, Guan Y. Robust unit commitment with wind power and pumped storage hydro. IEEE Trans Power Syst 2011;27(2):800
10. [8] Cleary B, Duffy A, OConnor A, Conlon M, Fthenakis V. Assessing the economic benefits of compressed air energy storage for mitigating wind curtailment. IEEE Trans Sustain Energy 2015;6(3):1021
8. [9] Khodayar ME, Manshadi SD, Wu H, Lin J. Multiple period ramping processes in dayahead electricity markets. IEEE Trans Sustain Energy 2016;7(4):1634
45. [10] Kasembe A, Maslo K, Hanka L, Hruska Z. Interaction of transmission and distribution systems from voltage control and protection settings point of view. In: 20th International conference and exhibition on electricity distribution (CIRED 2009). IET; 2009, p. 0623
24. [11] Li Z, Guo Q, Sun H, Wang J. Coordinated economic dispatch of coupled transmission and distribution systems using heterogeneous decomposition. IEEE Trans Power Syst 2016;31 (6):4817
30. [12] Li Z, Guo Q, Sun H, Wang J. A new LMP-sensitivity-based heterogeneous decomposition for transmission and distribution coordinated economic dispatch. IEEE Trans Smart Grid 2016;9(2):931
41. [13] Caramanis M, Ntakou E, Hogan WW, Chakrabortty A, Schoene J. Co-optimization of power and reserves in dynamic T&D power markets with nondispatchable renewable generation and distributed energy resources. Proc IEEE 2016;104(no. 4):807
36. [14] Jain H, Rahimi K, Tbaileh A, Broadwater RP, Jain AK, Dilek M. Integrated transmission and distribution system modeling and analysis: need and advantages. In: 2016 IEEE power and energy society general meeting (PESGM). IEEE; 2016. p. 1
5. [15] Venkatraman R, Khaitan SK, Ajjarapu V. A combined transmission-distribution system dynamic model with grid-connected DG inverter. In: 2017 IEEE power and energy society general meeting. IEEE; 2017. p. 1
5. [16] Samaan N, Elizondo MA, Vyakaranam B, Vallem MR, Ke X, Huang R, et al. Combined transmission and distribution test system to study high penetration of distributed solar generation. In: 2018 IEEE/PES transmission and distribution conference and exposition (T&D). IEEE; 2018. p. 1
9.

Integrated power transmission and distribution systems Chapter | 5

199

[17] Bragin M, Dvorkin Y, Darvishi A. Toward coordinated transmission and distribution operations. In: 2018 IEEE power and energy society general meeting (PESGM). IEEE; 2018. p. 1
5. [18] Carrion M, Arroyo JM. A computationally efficient mixed-integer linear formulation for the thermal unit commitment problem. IEEE Trans power Syst 2006;21(3):13711378. [19] Farivar M, Low SH. Branch flow model: relaxations and convexification - Part I. IEEE Trans Power Syst 2013;28(3):2554
64. [20] Ben-Tal A, Nemirovski A. On polyhedral approximations of the second-order cone. Math Oper Res 2001;26(2):193
205. [21] Chen X, Wu W, Zhang B. Robust restoration method for active distribution networks. IEEE Trans Power Syst 2015;31(5):4005
15. [22] Pan K, Guan Y. A polyhedral study of the integrated minimum-up/-down time and ramping polytope. arXiv preprint. arXiv:1604.02184; 2016. [23] Fathabad AM. Test data of modified ieee 33-bus system. ,https://www.dropbox.com/s/ psqv9yr3atg46bk/Configuration%20and%20Parameters%20of%20Modified%20IEEE% 2033-bus%20System.docx?dl 5 0.. [24] Solar power data for integration studies. ,https://www.nrel.gov/grid/solar-power-data.html.. [25] More than the largest source of energy data and water data. ,https://dataport.cloud/.. [26] Hourly aggregated wind output. ,http://mis.ercot.com/misapp/GetReports.do?reportType Id 5 13424&reportTitle 5 Hourly%20Aggregated%20Solar%20Output&showHTMLView 5 &mimicKey..

This page intentionally left blank

Part II

Modelling

This page intentionally left blank

Chapter 6

Integrated inexact optimization for hybrid renewable energy systems Y. Zhou and Z.X. Zhou Key Laboratory for City Cluster Environmental Safety and Green Development of the Ministry of Education, Institute of Environmental and Ecological Engineering, Guangdong University of Technology, Guangzhou, P.R. China

Chapter Outline 6.1 Introduction 203 6.2 Deterministic optimization techniques 204 6.2.1 Classical techniques 205 6.2.2 Metaheuristic algorithm 206 6.2.3 Commercial software 207 6.3 Inexact mathematical programming methods 207 6.3.1 Stochastic mathematical programming 208 6.3.2 Robust optimization 212

6.1

6.3.3 Fuzzy mathematical programming 214 6.3.4 Interval mathematical programming 218 6.3.5 Hybrid inexact mathematical programming 219 6.4 Integrated inexact optimization framework 220 6.5 Conclusions 222 References 223

Introduction

With fast economic development and rapid depletion of fossil fuels, an everincreasing amount of greenhouse gas has been released into the atmosphere over the last 150 years [1]. The Paris Climate Agreement pledges have been made by countries in the world to mitigate greenhouse gas emissions in the form of Nationally Determined Contributions (NDCs), yet according to the UN Emissions Gap Report 2018 [2], current commitments expressed in the NDCs are inadequate to bridge the emissions gap by 2030. The global greenhouse gas emissions reached a record high of 49.2 Gt CO2e in 2017 after 3 years of stagnation [2]. While renewable energy has large mitigation potentials, which

Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00006-6 © 2021 Elsevier Inc. All rights reserved.

203

204

PART | II Modelling

FIGURE 6.1 Hybrid renewable energy system.

are considered as the principal alternative to traditional fossil fuels [2,3], the main problem of it lies in its variability and vulnerability to environmental conditions such as solar irradiance, wind speed, and runoff. Individual renewable energy sources cannot provide a continuous power supply to the load because of the random and intermittent nature of these factors [4
7]. The fluctuant power output of the renewable energy plant has great effects on the voltage stability of power systems [6], but integrating renewable energy sources and energy storage into a hybrid system can overcome such challenges. A hybrid renewable energy system (HRES) is a combination of two or more renewable and nonrenewable energy sources and storage devices, such as wind, solar, hydro, biogas, diesel, and fuel cells [8
10]. Power sources (i.e. solar arrays, wind turbine and diesel engine generator), the battery and power management centre are the basic components that regulate power productions to meet the load demand in both grid-connected and standalone modes (as shown in Fig. 6.1). Traditional energy resources and energy storage play the roles of back-up power sources. If the output of renewable resources is not enough, the rest of the load can be supplied by traditional energy resources. If renewable resources have extra generation, the excess energy can be absorbed by energy storage or be sold to the grid under the high electricity prices [11]. HRES has the advantages of improved energy efficiency and system reliability, as well as reduced energy cost and greenhouse gas emission [4,11]. Renewable energy also demonstrates its superiority in power supply in small islands and rural areas [6]. HRES has aroused wide concerns in recent years, primarily attributed to the technical advances of renewable energy and the subsequent rise in petroleum prices [12
15], and the optimal planning and operation of HRES requires robust optimization (RO) techniques. This chapter overviews deterministic and inexact mathematical modelling in HRES and proposes an integrated inexact optimization framework.

6.2

Deterministic optimization techniques

HRES has become a cost-effective and reliable option for the regional power supply system [16]. As one of the most important issues in HRES, optimal

Integrated inexact optimization for hybrid Chapter | 6

205

planning and operation relies on the performance of the system components, such as the number of solar arrays and batteries, capacity of generators and converters, availability of renewable energy, load profile and nature of storage elements [17,18]. The mismatch between available power and load demand could lead to the unreliability of the system, as well as increasing storage demand or nonrenewable energy consumption. A better performance with minimized system costs or maximized system reliability can be achieved through optimal planning of the energy system. Many optimization techniques and tools are developed to address these problems in HRES [19
23], which can be classified into classical techniques, metaheuristic algorithm and commercial software.

6.2.1

Classical techniques

Classical techniques include linear programming (LP), mixed-integer linear programming (MILP), nonlinear programming (NLP), multiobjective programming (MOP), and analytical methods. The LP method has been widely used in HRES optimization, where the objective function is linear and design variables space is specified using only linear equalities and inequalities. Kusakana et al. [24] developed a linear mathematical model for HRES in rural South Africa, where the capital investment cost of each component of renewable energy was minimized subject to the energy sources, size of components and energy demand. Here, the method is general and can be applied to different locations and energy scenarios. The LP method is competent to perform economic analysis under technical and reliability constraints, though the breakdown of any of the renewable elements may drastically affect the overall system energy delivery capability [19]. The MILP method could address the binary variables problem and deliver the performance in determining the location and size of the HRES. Camargo et al. [25] applied the MILP method to size the hybrid PV-windbattery system with minimized total installation costs. Bartolucci et al. [26] developed an optimization method by coupling MILP and model predictive control strategy, which improved the accuracy of the load forecasting algorithm. Li et al. [27] proposed a simulation-based mixed-integer optimization model for optimizing the installation capacity and operation strategy of the HRES with offshore wind energy for container terminals in northeast China. The NLP method is applicable for the cases in which the objective functions or/and the constraints include nonlinear parts. Vaccari et al. [28] developed a sequential linear programming algorithm to solve a general nonlinear problem in operation optimization of HRES. Different device models- ranging from conventional, renewable, combined heat and power generators, to electrical/thermal load and accumulators - have been considered. The method was applied to real HRES in Tuscany, Italy, for consideration of different

206

PART | II Modelling

energy policies available. Although the NLP method enables solving complex problems with simple operations, it works with more iterations. The MOP method can deal with conflicting objectives (i.e. system cost, system reliability and efficiency, and environmental indicator) and coordinate the interests of different stakeholders. Forough and Roshandel [29] presented the multiobjective receding horizon optimization for optimal scheduling of HRES, with factors of diesel fuel cost and battery wear cost considered to evaluate the economic and technical performances of the system. Li et al [30] proposed a multiobjective optimization model for optimal process selection and sizing of standalone HRES in remote rural areas, while the system profits and carbon abatement capability were maximized. However, the MOP method often encountered difficulties in objectively weighing multiple objectives, especially when their units or orders of magnitude were distinct. The analytical method is used for optimal components sizes of HRES based on mathematical, theoretical analysis, and calculations [31]. It has the advantages of less computational time [32], leading to a better computational efficiency in comparison with the Monte-Carlo simulation method [33]. Jakhrani et al. [34] presented a novel analytical model for the determination of optimal sizing of standalone photovoltaic systems with the least cost and predetermined reliability to satisfy the load. Fantauzzi et al. [35] developed an analytical methodology for the optimal sizing of energy storage devices. The formulation of the problem allows an analytical solution to be obtained that represents a suitable tool for the purpose of designing storage.

6.2.2

Metaheuristic algorithm

In real-world planning and operation of HRES, complexities exist in the uncertainties associated with renewable resources and load demand, system components and location constraints [36]. Classical techniques have difficulties in solving these complex optimization problems. Metaheuristic algorithms have been applied in HRES to address nonlinear or continuous optimization problems and are more accurate, efficient and capable of achieving optimal solutions [37]. Metaheuristic algorithms can be classified into trajectory-based metaheuristics
which develop a single solution at a time - and population-based metaheuristics [38], which process a whole population of solutions at the same time. Simulated annealing belongs to trajectory-based metaheuristics, while genetic algorithm (GA) and particle swarm optimization (PSO), and ant colony optimization are examples of population-based metaheuristics. Sharafi et al. [39] proposed a novel approach based on PSO to minimize the total cost of the system, unmet load and fuel emission simultaneously for optimal design of HRES, which included wind turbines, PV panels, a diesel generator and two storage devices. Fetanat and Khorasaninejad [40] proposed ant colony optimization for continuous domains-based integer programming

Integrated inexact optimization for hybrid Chapter | 6

207

for size optimization in a hybrid PV-wind energy system. Mellouk et al. [41] developed a new parallel hybrid genetic algorithm-particle swarm optimization algorithm (P-GA-PSO) to solve both sizing and energy management problems for microgrids. The algorithm enabled the system to obtain the optimized energy management strategy. However, the metaheuristics algorithm is complicated and the implementation cost is high. Sharafi and Elmekkawy [42] adopted a particle swarm optimization-simulation based approach for sizing HRESs, and three objectives including minimum cost, CO2 emission and maximum reliability were considered. Metaheuristic algorithms have no requirements to characterize optimization problems. However, with the increase in the complexity of the hybrid system, there are shortcomings such as falling into local optimal solutions or slow convergence speed [21].

6.2.3

Commercial software

Hybrid Optimization Model for Multiple Energy Resources (HOMER) software was developed by the US National Renewable Energy Laboratory and has been widely employed for HRES optimization. HOMER aims to minimize the net present value to obtain the optimal capacity scheme of HRESs by using hourly data. Weber et al. [43] applied HOMER Pro software to optimize the output power and cost of the system, and the optimal allocation of different renewable energy allocation combinations was obtained. Murugaperumal and Raj [44] used HOMER for optimal design and economic analysis of HRES for rural electrification in Korkadu, India, while Padro´n et al. [45] adopted HOMER for optimal design and configuration of HRES for autonomous desalination systems on two islands. HYBRID2 is a timeseries computer model developed by the Renewable Energy Research Laboratory of the University of Massachusetts and with the calculation step of the software being 10 minutes to 1 hour, it has a higher accuracy than HOMER [46]. A variety of different control strategies/options could be implemented which incorporate detailed diesel dispatch and interactions between diesel gensets and batteries. The National Renewable Energy Laboratory suggests using HOMER to optimize the system first, and then using HYBRID2 to improve the system [47]. However, HOMER mainly focuses on small-scale HRES, integrated application of large-scale grid-connected hybrid systems should be further extended. Das et al. [14] and Li et al [48] summarized that the algorithms of HOMER software are encapsulated, making it difficult to modify the codes of the software itself for specific situations such as user-defined constraints and model equations.

6.3

Inexact mathematical programming methods

The optimization modelling often represents a series of interactions between system components, processes and factors by several mathematical

208

PART | II Modelling

equations. However, in practical HRES planning and operation, many modelling inputs and impact factors (i.e. technical, economic, policy and social factor) may have an uncertain nature with multiple dimensions and layers, which are difficult to accurately characterize, such as the volatility and availabilities of renewable resources, the randomness of load demand, the fluctuation of diesel prices in markets, the investment cost of renewable energy conversion technologies, the operational and maintenance cost of hybrid system, and the power market regulation emission reduction target [20,49]. The inherent complexities of HRES are further multiplied as a result of the multiperiod, multi-objective and multi-facility features of system components. Uncertainties associated with dynamics features of the system (i.e. the temporal and/or spatial variations of energy resources and their associated economic and environmental implications) may be presented as multiple forms, which makes conventional deterministic optimization techniques incapable of effectively dealing with such complexities [50]. Neglecting these interactions, complexities, and uncertainties, may lead to biased or unsound results of decisions. Generally, there are three types of uncertainties including probabilistic distributions, fuzzy numbers, and intervals. Inexact mathematical methods include stochastic mathematical programming (SMP), robust optimization (RO), fuzzy mathematical programming (FMP), interval mathematical programming (IMP) and their hybrids, which have been gaining increasing popularity in the decision-making surrounding HRES under uncertainties [21,49
52].

6.3.1

Stochastic mathematical programming

Stochastic mathematical programming derived from probability theory can address planning problems with random variables of which known probability distributions have been known. Since 1950, a number of SMP models and solutions have been developed [53]. Chance-constrained programming (CCP) and stochastic programming with recourse (SPR) are the classical modelling paradigms of SMP.

6.3.1.1 Chance-constrained programming Initiated by Charnes et al. [54], chance-constrained programming has been widely applied in reflecting the reliability of satisfying (or risks of violating) system constraints under uncertainty. Constraints could be satisfied with a certain probability level. A general CCP problem can be formulated as follows: Max f 5 C ðtÞX

ð6:1aÞ

AðtÞX # BðtÞ

ð6:1bÞ

subject to

Integrated inexact optimization for hybrid Chapter | 6

xj $ 0; xj AX; j 5 1; 2; . . .; n

209

ð6:1cÞ

where X is a vector of decision variables; AðtÞ, BðtÞ and CðtÞ are parameter vector with random elements defined on probability space T [55]. When uncertainties of some elements in the right-hand sides (i.e. load demand) can be expressed as probabilistic distributions, CCP can be solved by introducing a certain level of probability pi A [0, 1] for each constrained i. It indicates the condition that the constraint is satisfied with at least a probability of 1 2 pi . The set of feasible solutions is thus restricted by the following constraints [55]:    Pr tai ðtÞX # bi ðtÞ $ 1 2 pi ; ai ðtÞAAðtÞ; i 5 1; 2; . . .; m ð6:2Þ where pi is the significance level, representing the probability of violating the constraint; 1 2 pi is the confidence coefficient, indicating the probability of satisfying the constraint. Since the constraints are generally nonlinear, the set of feasible constraint is convex only for some particular cases, one of which is when Ai ðtÞ are deterministic and bi ðtÞ are random (for all pi values). Under this condition, constraint (6.2) becomes linear: ai ðtÞX # bi ðtÞðpi Þ ; ai ðtÞAAðtÞ; i 5 1; 2; . . .; m

ð6:3Þ

where bi ðtÞðpi Þ 5 Fi21 ðpi Þ, given the cumulative distribution function of bi [i.e. Fi ðpi Þ], and the probability of violating constraint i (pi). When the left-hand side parameter Ai ðtÞ or parameters in both sides of constraint (i.e. renewable power output, load demand) are random variables, constraint (6.1b) can be respectively expressed as follows:   
 Pr tai ðωÞX # bi ðtÞ $ 1 2 pi ; ai ðωÞBN ηi ; τ 2i ; bi ðtÞABðtÞ; i 5 1; 2; . . .; m ð6:4Þ

   Pr ½ai ðωÞX # bi ðωÞ $12 pi ; ai ðωÞBN μi ; δ2i ; bi ðωÞBN ν i ; ζ 2i ; i 5 1; 2; . . .; m ð6:5Þ A series of the solution methods were developed to transform left-hand side/double sides, individual/joint chance-constraints into the deterministic ones [56
60]. When Kamjoo et al. [61] applied CCP to design a standalone hybrid wind turbine/PV and battery bank system, the joint distribution of the wind turbine and PV panel output power was considered to follow an unknown distribution and individual cumulative distribution function based on the hourly historical data of wind speed and solar irradiance. Wu et al. [62] proposed a chance-constrained stochastic programming for the stochastic dayahead scheduling, while the loss of wind probability and transmission line overloading probability are modelled as chance constraints to provide high reliability of power supply and to ensure the high utilization of wind

210

PART | II Modelling

generation. Odetayo et al. [63] proposed a chance-constrained and reliability programming optimization model for long-term planning of natural gas and power system; capital and operating costs of natural gas generators, pipelines and compressors were minimized subject to probabilistic constraints such that the desired confidence level of power and natural gas supply was attained. Zeynali et al. [64] proposed a multiobjective optimization methodology for sizing and siting of power systems concerning the probability of constraints violation which was calculated by the probability distribution function (PDF) obtained from the maximum entropy method.

6.3.1.2 Stochastic programming with recourse In the stochastic programming with recourse, the recourse decisions can be made to compensate for any bad consequences that might have been experienced as a result of the first-stage decision prior to the observation of random parameters. The optimal policy from the SPR model is a single firststage policy and a collection of recourse decisions (a decision rule) that defines which second-stage action should be taken in response to each random outcome [65,66]. Thus, SPR is effective for problems where an analysis of policy scenarios is desired. SPR aims to seek several decision alternatives that are feasible for all (or almost all) of the possible parameter realizations and optimize the expectation of some functions of the decisions and the random variables [53]. SPR includes two-stage stochastic programming (TSP), and multistage stochastic programming (MSP). Consider a general TSP model: z 5 minCT X 1 EωAΩ ½QðX; ωÞ

ð6:6aÞ

xAX

ð6:6bÞ

subject to where XDRn1 , CDRn1 and YDRn2 . ω is a random variable from space ðΩ; F; PÞ with ΩDRk , x is the first-stage decision variable, which needs to be made before the realization of random parameters ω; recourse function QðX; ωÞ means the optimal value of the second-stage problem. Qðx; ωÞ 5 minf ðωÞT y

ð6:7aÞ

DðωÞy $ hðωÞ 1 T ðωÞx

ð6:7bÞ

yAY

ð6:7cÞ

subject to

where f :ΩDRn2 , h:ΩDRm2 , D:ΩDRm2 3 n2 and T:Ω-Rm2 3 n1 . y is the second-stage variable, which is taken to minimize the expected consequences of the preciously-taken decisions. The solution of TSP aims to address the

Integrated inexact optimization for hybrid Chapter | 6

211

expected recourse function, and four kinds of approximation algorithms were developed [67]. Scenario method is to approximate the expected recourse function by the sample average of Qðx; ωi Þ for several samples ω1 ; ω2 ; . . .; ωi [68] or discrete value ωh with different probability levels ph [69]. Stochastic gradient techniques could obtain the smoothed estimates of the gradient of the expected recourse function and updates solution through using stochastic subgradients as directions [70]. Based on Benders’ decomposition, primal and dual decomposition could construct feasible dynamic programming policies through outer approximation of a (convex) recourse function computed using Benders cuts [67,71]. Separable approximation methods replace the expected recourse function with linear or piecewise linear approximation function [72]. Zhou et al. [67] proposed a new convergent hybrid learning algorithm to approximate the expected recourse function for TSP, which provides the first theoretical support for the use of the piecewise linear approximation method in TSP. MSP is the extension of TSP to the multistage setting, which means decisions could be updated sequentially in several points of time in response to realized events/scenarios. In MSP problems, there is a sequential decisionmaking process involving a set of decision stages and a stochastic process b1,. . ., bT representing the future evolution of the random input parameters. At each stage t, the decision-maker makes decisions based on the past observed values of b1,. . ., bT to optimize the objective functions. This indicates that in order to realize random parameters bt , decision vector xt and conditions of the system at stage (t 1 1) are affected. A general MSP model can be expressed as follows: ht ðxt21 ; bt Þ 5 min ct xt 1 Ebt11 jbt ht11 ðxt ;bt11 Þ

ð6:8aÞ

At xt 5 Bt xt21 1 bt

ð6:8bÞ

xt $ 0; t 5 2; . . .; T

ð6:8cÞ

xt

subject to

where xt is decision vector for a particular stage t and bt denotes the stochastic parameter (i.e. power demand at stage t). At term means the model’s structural constraint matrix at stage t (i.e. power balance). The solving thought of MSP is similar to TSP. A discretization of their continuous probability distribution is introduced to approximate the continuous random variables b1,. . ., bT. Uncertainties can thus be conceptualized into a scenario tree, with a one-to-one correspondence between the previous random variable and one of the nodes (state of the system). The scenario tree is built to express the evolution of random parameters over time and the dynamics between decisions, and the MSP problem is converted into a deterministic optimization problem [53,73]. Kadri et al. [74] summarized three kinds of methods in scenario tree design. The first method uses binary or ternary

212

PART | II Modelling

scenario trees and only considers a limited number of possible outcomes at each stage (low, high/low, medium, high) [75,76]. Sampling methods (i.e. Latin Hypercube sampling) are used to discretize a continuous probability distribution [77,78]. Since the size of the scenario tree often grows exponentially with the increasing number of stages and branches at each stage, scenario tree reduction techniques are used to generate scenario trees with acceptable size while keeping the approximate representation of the stochastic process as accurate as possible [74,79,80]. Various decomposition methods and heuristic algorithms are widely used to solve complex MSP problems with high computational efficiency [78,81,82]. Wang et al. [83] applied TSP for the optimal design of a HRES for seaport, where stochastic characteristics of wind energy generation and energy demands were performed. Yu et al. [84] developed a TSP model to optimize the capacity of the energy storage systems, with the model transformed into a MILP problem based on multiple equivalent scenarios. Daneshvar et al. [85] formulated a TSP model for optimal scheduling of the wind-thermalhydropower-pumped storage system considering the competitive interactions between the electrical generation units. Considering the uncertainties associated with electricity demand and wind speed, the first stage focused on the day-ahead scheduling of thermal power plants, while balancing market dispatch was considered in the second stage by using the stochastic producers and quick dispatch units. Hafiz et al. [86] adopted a MSP to identify the optimal storage sizing for each house of a shared community, while multistage decisions related to electricity purchases from the grid, and storage device charge/discharge were considered to deal with uncertainties related with electricity demand and solar power generation. Sahin ¸ et al. [87] formulated a MSP model for demand response optimization where different kinds of smart home appliances under uncertain weather conditions and availability of renewable energy were scheduled; scenario groupwise decomposition was applied to compute lower and upper bounds for instances with a large number of scenarios, which was demonstrated to be effective in finding highquality solutions in small computation times.

6.3.2

Robust optimization

SMP assumes that probability distributions are known or can be estimated. In real-world applications, there are difficulties in acquiring and establishing the PDFs of stochastic parameters. Being initiated by Soyster [88] in the 1970s, RO can effectively solve optimization problems with stochastic parameters with unknown PDFs, which assumes that the uncertain data resides in the so-called uncertainty set [89]. RO aims to find conservative solutions where constraint violation cannot be allowed for any realization of the uncertain parameters across the entire uncertainty set. A general RO problem can be formulated as follows:

Integrated inexact optimization for hybrid Chapter | 6

213

min cT x

ð6:9aÞ

AðξÞx 1 Byðξ Þ # d; ’ξAZ

ð6:9bÞ

x;y

subject to

where xARn is the first-stage ‘here and now’ decision variable, which is made before uncertain parameter ξARL is realized; yARk is the second-stage ‘wait and see’ decision variable, which can be adjusted based on the actual data; cARn , ZARm 3 k , dARm is the certain coefficient matrix and A is the uncertain coefficient matrix that restricted in the user prespecified uncertainty set ZCRL . The solution of x and y is called robust feasible if it satisfies the uncertain constraints Aðξ Þx 1 ByðξÞ # d for all realization of ξAZ. Since model (6.9) is a complex problem, the function yðξÞ is restricted [90]. A factor model is to formulate constraint (6.9b) as an affine function y0 1 Qξ of the primitive uncertain parameter ξAZ, where y0 ARk and QARk 3 L . Model (6.9) can be treated into the tractable reformulation: min cT x

ð6:10aÞ

AðξÞx 1 By0 1 BQξ # d; ’ξAZ

ð6:10bÞ

x;y0 ;Q

subject to

The dimension of the general uncertain parameter Qξ is often much higher than that of the primitive uncertain parameter ξcL. Gorissen et al. [89] summarized that there are two ways, including robust reformulation technique and adversarial approach, to handle the infinitely constraint due to them for all quantifiers imposed by the worst-case formulation. The robust reformulation technique is one of the main techniques in RO [91]. Based on this technique, infinitely constraints are converted to finite ones by worst-case reformulation, strong duality, and robust counterpart. Robust policies are of particular interest in real-world applications where changes in solutions are not easy to implement, such as capacity planning of the energy system [92,93]. In particular, the decision-maker may be willing to sacrifice some expected profits in order to avoid continuous changes in the operations whenever a small change in the data is observed. In the TSP framework, first-stage decisions are scenario invariant, while second-stage recommendations differ by scenario. If the differences are large, then the operations of the real-world system must change substantially every time a new scenario is observed, though the RO modelling framework tries to avoid such situations [92]. The inaccurate long-time forecasts of electricity demand and fuel prices would lead to overcapacity in power systems, which is first presented by Akbari-Dibavar [93]. Akbari-Dibavar [93] adopted the RO approach to a European energy planning problem and compared it to the

214

PART | II Modelling

deterministic decision-making approach and found that considering uncertainties in the long-term planning process can reduce the risk of generating overcapacity in national power systems. Ranjbar et al. [94] proposed a trilevel min-max-min optimization problem for coplanning of transmission investment and merchant distributed energy resources (DER) - where the upper, middle and lower levels are investment decision of transmission lines and DER, worst case realization of uncertain parameters, and the best actions in order to minimize the operation costs. Tan et al. [95] applied RO for dispatching optimization problems of a gas-electric virtual power plant, where the random fluctuations and characteristics of wind and solar were addressed by introducing different robust coefficient that reflects the ability of the system to withstand risk.

6.3.3

Fuzzy mathematical programming

FMP is based on the fuzzy set theory pioneered by Zadeh [96], representing the imprecision in a decision-making situation. Ambiguous and vague information can be effectively addressed by FMP in a direct way without a large number of realizations. Based on these two kinds of uncertainties, FMP can be divided into three categories: (1) fuzzy flexible programming (FFP) which can treat vague information in the objective function and constraints; (2) fuzzy possibilistic programming (FPP) which can tackle ambiguous coefficients in both objective functions and constraints; and (3) fuzzy robust programming (FRP), which can deal with both ambiguous and vague information in the model.

6.3.3.1 Fuzzy flexible programming FFP can be applied to represent the uncertainty in stipulation, which means the constraint does not need to be fully satisfied but should be met to a certain satisfactory level identified through membership function [97]. FFP represents fuzzy relationships between the left-hand and the right-hand sides of the constraints. A general FFP can be formulated as follows: Min f 5 CX

ð6:11aÞ

AX # B

ð6:11bÞ

X$0

ð6:11cÞ

subject to ~

where AAfRgm 3 n , BAfRgm 3 1 , CAfRg1 3 n , XAfRgn 3 1 , R means a set of variables and parameters; # ~ means fuzzy equal to or less than. A maximum-minimum operator was applied to solution the FFP. The control variable λ corresponding to the degree of satisfaction (membership grade)

Integrated inexact optimization for hybrid Chapter | 6

215

for fuzzy decision was introduced, which performs its associated characteristics of being the minimum value among the maximized relaxation conditions to all constraints [97,98]. Compromised solutions under acceptable satisfactory levels (always lower than 100%) can be generated. The above model can be converted into as follows: Max λ subject to


 CX # f 1 2 ð1 2 λÞU f 1 2 f 2
 AX # B1 2 ð1 2 λÞU B1 2 B2

ð6:12aÞ

ð6:12bÞ ð6:12cÞ

X$0

ð6:12dÞ

0#λ#1

ð6:12eÞ

where f 1 and f 2 denote the lower and upper bounds of the objective’s aspiration level from the decision-maker. λ is the control decision variable corresponding to the membership grade of satisfaction for the fuzzy decision. Both the flexibility of the target values of the objective function and the elasticity of the constraints in model can be addressed by membership grade λ. Over the past few decades, a number of FFP associated solution methods were proposed, such as max-min operator [99], compromise solutions [100,101], fuzzy ranking [102,103], evolutionary algorithm [104] and interactive fuzzy satisfying method [105,106]. Hocine et al. [107] proposed multisegment fuzzy goal programming for optimizing renewable energy portfolios under uncertainty. Through minimizing deviations between the achievement of goals and their aspiration levels were minimized and the goal fuzziness was handled. The method was demonstrated to improve the effectiveness and accuracy of the decision. Faddel et al. [108] proposed an algorithm based on fuzzy linear programming to maximize the profits of the electric vehicle aggregator in a bidirectional vehicle-to-grid, with the uncertainties of regulation and responsive reserve prices, deployment signals, and the energy used for EV trips were considered. The method could deal with both fuzzy objective and fuzzy inequality with a good level of satisfaction. Jayachitra et al. [109] proposed a fuzzy programming-based approach to solve an extended binary-integer programming model, in which the degree of satisfying the fuzzy objective under the given constraints was maximized. In this approach, the fluctuations in the demand and the availability of manufacturing facilities in each period are regarded as fuzzy. In the current FFP model, the objective function and all of the constraints are relaxed under the same satisfactory level. That, however, may lead to stringent levels for constraints, where the solutions obtained would not be satisfactory enough. Multiple control variables,

216

PART | II Modelling

respectively corresponding to the objective function and each constraint, should to be considered in order to reflect various relaxation levels to the objective and constrains in the HRES management system.

6.3.3.2 Fuzzy possibilistic programming In FPP, parameters uncertainty is presented in as ambiguous coefficients with possibility distributions in the objective function and/or constraints [110
112]. A general FPP can be formulated as follows: Min f ~ 5 C~X

ð6:13aÞ

A~X # B~

ð6:13bÞ

X$0

ð6:13cÞ

subject to

where AAfRgm 3 n , BAfRgm 3 1 , CAfRg1 3 n , XAfRgn 3 1 , R means a set of fuzzy possibilistic variables and parameters which can be treated as fuzzy membership functions; X denotes the vector of nonfuzzy decision variables. The solution of FPP is the defuzzification of ambiguous coefficients to convert the problem into a deterministic one. As the popular possibility distributions, triangular and/or trapezoidal fuzzy membership functions are applied to reflect ambiguous coefficients. Many defuzzification algorithms have been developed for symmetric and nonsymmetric fuzzy numbers in the objection function (weighted average method [113], interactive method resolution [114,115], fractile approach [116,117]) and constraints (possibility and necessity degree [118,119], vertex method [120], distance ranking method [121], necessity measure [122
124], credibility measure [125]). Mohammadi et al. [126] developed a fuzzy-based scheduling model for optimal scheduling of the wind integrated multienergy systems, while the effect of multiple uncertainties (i.e. energy demands, electricity prices and wind power production) was presented by fuzzy set methods. Moradi et al. [127] employed the interval fuzzy approach to offer an efficient energy management system strategy. The coefficient uncertainties in the right-hand side of constraints (i.e. electrical and thermal demands) and the objective function (i.e. gas and electricity prices) were quantified by fuzzy numbers which represented the upper and lower limits in specific α-cuts. This method can reduce the complexity associated with analysis and implementation while providing an optimal solution to the forecasting problem. Rosso-Cero´n et al. [128] presented a novel approach based on fuzzy multicriteria decisionmaking tools for assessing sustainable alternatives of power generation. The fuzzy parameters (i.e. energy demand, the investment cost and fuel costs, energy resources availability and technical parameters) with triangular possibility distributions were defined by a triplet, representing low, medium (most probable) and high scenarios, respectively. Based on a minimum

Integrated inexact optimization for hybrid Chapter | 6

217

acceptable possibility level, fuzzy objective was converted by an expected value operator, and fuzzy parameters in the left- and right-hand sides of constraints were addressed by the fuzzy ranking concept and weighted average method. The FPP presents the average system performance, and has difficulties in ensuring any realization of uncertain parameters and may encounter challenges in obtaining risk-averse solutions under high variability conditions [129,130]. Robust possibilistic programming was developed for seeking robust solutions through controlling both optimality robustness and feasibility robustness, which can avoid imposing a high risk to decision-makers.

6.3.3.3 Fuzzy robust programming FRP can handle uncertainties in left- and right-hand side coefficients (of all constraints) represented by possibilistic distributions [131
133]. A general FRP model can be formulated as follows: Min f 5 CX

ð6:14aÞ

A~X # B~

ð6:14bÞ

X$0

ð6:14cÞ

subject to ~

where # ~ means fuzzy inequality. Based on the fuzzy set theory, FRP can be converted into a deterministic version through transforming the m imprecise constraints into 2 km precise inclusive ones that correspond to k α-cut levels: Min f 5 CX

ð6:15aÞ

subject to n X

s

asij xj # bi ; i 5 1; 2; . . .; m; s 5 1; 2; . . .; k

ð6:15bÞ

j51 n X

a sij xj # b si ; i 5 1; 2; . . .; m; s 5 1; 2; . . .; k

ð6:15cÞ

j51

xj $ 0; j 5 1; 2; . . .; n s bi

ð6:15dÞ

denote the superior limit value in where asij AðA~ij Þαis , bsi AðB~i Þαis asij and sets asij and bsi , respectively, a sij and b si indicate the inferior limit value in sets asij and bsi , respectively. The robustness of the optimization process can be improved through dimensional enlargement of the original fuzzy constraints. Since the weakness lies in the deterministic coefficients for the objective function, FRP is applied with other methods [134,135]. Li et al. [136] developed a two-stage fuzzy robust integer programming for capacity planning of

218

PART | II Modelling

environmental management systems under uncertainty. Integrating FRP and TSP within a MILP framework would explicitly address the uncertainties presented in terms of both possibilistic and probabilistic distributions such that obtained solutions can possess increased stability and thus enhanced robustness. Compared with SMP, FMP has low data requirements based on possibility distributions which are established based on decision-makers’ personal judgments and description. However, fuzzy inputs may simplify relevant parameters for optimization modelling, and it is difficult to reduce the subjectivity of personal judgment.

6.3.4

Interval mathematical programming

When the probability distributions or possibility distributions were unknown and only collected as discrete intervals with lower and upper bounds, the optimization problem can be solved by interval analysis. Interval analysis was initiated by Moor [137] and Alefeld and Herzberger [138] and was extended to IMP by Huang et al. [139]. Huang and Moore [140] pioneered an interval linear programming approach based on a two-step interactive algorithm. The IMP approach improved upon the existing optimization methods by allowing uncertainties to be directly communicated into the optimization and solution processes. The two-step algorithm did not cause more complicated intermediate models and thus had relatively low computational requirements. Since fluctuation intervals were acceptable as uncertain inputs in both objection function and constraints, IMP showed its superiority in practical applications, especially uncertain data acquisition. According to Huang et al. [140
142], a general IMP model can be formulated as follows: Min f 6 5 C 6 X 6

ð6:16aÞ

A6 X 6 # B6

ð6:16bÞ

subject to

X6 $0 ð6:16cÞ        n 3 1 m 3 n m 3 1 1 3 n where A 6 A R 6 , B6 A R6 , C6 A R6 , X6 A R6 , R means a set of interval numbers. According to two-step interactive algorithm, the model can be converted into two deterministic submodels. The submodel corresponding to f 2 is solved first, the sub-model corresponding to f 1 was then solved based on ithe solution hof the first h i submodel, meaning 6 2 1 6 1 5 fopt ; fopt 5 x2 and xopt the final solutions fopt opt ; xopt can be obtained. The

detailed algorithm can be found in references [140
142]. Based on the two-step interactive algorithm, Wang et al. [143] aimed at the optimal operation of the integrated electrical and natural-gas systems by considering uncertainties of the wind turbine and PV through interval mathematics. Bai et al. [144] proposed an interval optimization-based coordinated

Integrated inexact optimization for hybrid Chapter | 6

219

operating strategy to improve the overall system operation for a gaselectricity integrated energy system, with wind power uncertainty was represented as interval numbers instead of probability distributions. An index of interval possibility degree was used to convert the uncertain constraints into deterministic ones. Taghizadeh et al. [145] applied the interval optimization approach for obtaining a risk-constrained operation of PV/fuel cell/battery hybrid energy system under the uncertainty of upstream network price; the objective function was converted into the average cost and the deviation from the determinist cost of the hybrid energy system. One weakness of IMP is that part of optimum solution points obtained through the two-step method may go beyond the decision space in some cases, namely, solution violation, which may mislead decision-makers towards unreasonable policies, plans, or strategies that play important roles in real-world applications. Wang and Huang [146] pointed out such phenomenon and proposed an improved two-step method to avoid resulting violation by introducing extra constraints in the solving process. In comparison, the improved two-step method can achieve the best optimum objective function value, which is the same as that of the two-step method. Meanwhile, the size of the solution space will not be greatly reduced.

6.3.5

Hybrid inexact mathematical programming

In the HRES, load demand and renewable power availabilities would fluctuate and usually present probability distributions that were estimated by past information, while the mean and variance of distribution would change due to a number of reasons, such as references of consumers and weather conditions. It is difficult to reflect the precise value of parameters, especially without abundant information. Approximately parameters based on expert experience are suitable to describe such parameters. Interval programming is often used, due to the nature of uncertain inputs derived from fluctuating power market and imprecise information (such as natural gas price and demand). With more complexities and uncertainties to be considered in the optimization processes, the results of hybrid optimization methods may be more applicable and closer to the actual situation. Hybrid inexact mathematical programming methods were developed to handle complex decision problems with more than one type of uncertainty. When randomness, fuzziness and interval numbers co-occurred in a decision-making framework, fuzzy stochastic programming, interval stochastic programming, interval fuzzy programming and hybrids of SMP, IMP and FMP were proposed to address various uncertainties embedded within optimization problems. Ji et al. [147] developed a hybrid inexact stochastic-fuzzy CCP for electric schedule management of microgrid system, while interval-parameter programming and TSP were combined into the fuzzy credibility constrained programming (FCCP) framework to manage pollutants and CO2 emissions

220

PART | II Modelling

under uncertainties presented as interval values, fuzzy possibilistic and stochastic probabilities. The fuzzy random variable is introduced by Kwakernaak [148,149] and Puri and Ralescu [150], which is viewed as fuzzy perception/observation/report of a classical real-valued random variable [151]. Random fuzzy variable was provided by Liu [152] and is referred to as those fuzzy variables, whose prominent values can be changed randomly according to the tolerance of the available past records. Zahiri et al. [153] developed a novel multistage possibilistic stochastic programming approach which could enhance TSP and MSP by relaxing the subjectivity, setting the deterministic occurrence probabilities and allowing possible perturbation in their values as fuzzy probabilities. The deep uncertainty of the scenariodependent parameter at the early stage of the scenario tree was accounted for random fuzzy variables. When the precise values of probability distribution parameters of random variables could not be obtained and the probability distribution parameters with limited sampled data are modelled as interval parameters, such hybrid uncertainty is called interval-valued random variable [154,155]. Li et al. [156] developed a fuzzy-boundary interval-stochastic programming method for planning water resource management systems, which can deal with uncertainties expressed as probability distributions and fuzzyboundary intervals. Few studies focus on the optimal design and operation of HRES under multiple uncertainties.

6.4

Integrated inexact optimization framework

In recent years, inexact programming methods, environmental analysis tools, and scenario analysis were integrated into the conventional deterministic optimization framework to handle complex decision problems in HRES. Landil et al. [156] proposed an interactive approach for optimal system configuration of hybrid renewable hybrid energy system by maximizing self-sufficiency, self-consumption and energy, minimizing costs and environmental impacts. Here, life-cycle analysis was conducted to estimate the environmental impacts of the components of PV panel and storage system and energy purchased by the grid; multiobjective optimization was adopted to address energetic, environmental and economic objectives, and different Pareto optimal solutions are obtained by different weight for five conflicting objective functions. Li et al. [157] proposed the concept of the negative emission HRES for the first time and developed a stochastic multiobjective decision-support framework to identify the optimal design of the energy mix and discuss the economic and environmental feasibilities of a hybrid solarwind-biomass renewable energy system with biochar production. Abedi et al. [158] developed a comprehensive method for optimal power management and design of hybrid renewable system-based autonomous energy systems, using a differential evolution algorithm accompanied by fuzzy technique is used to handle the mixed-integer nonlinear multiobjective optimization problem.

Integrated inexact optimization for hybrid Chapter | 6

221

Muh and Tabet [159] combined scenario analysis with HOMER to investigate the feasibilities of off-grid renewable hybrid systems for remote electrification in Southern Cameroons using the climate data of Wum. By comparing economic, energy, technical and environmental performance of eight scenarios with eight combinations of system components, the hybrid PV/wind/small hydro/battery system was the optimum scenario with zero-emission and second for low cost of energy from an environmental perspective. As the proportion of renewable energy in the power system as well as the frequency and magnitude of extreme weather events increases, so does the need for integrated optimization techniques to stabilize the HRESs related to a variety of complex factors. Owing to the formulation and computational limitations, optimization applications under multiple uncertainties in HRESs are still rare, and only consider one kind of uncertainty parameters. An integrated inexact optimization framework is proposed to overcome these barriers, which is beneficial for identifying precise design schemes and operation strategies of HRES and improving the system reliability (as shown in Fig. 6.2). Inexact optimization modelling is integrated into a conventional optimization framework to better quantify the various intrinsic uncertainties and interactions from the production and demand side, as well as relevant social, economic, and policy factors. Multiple uncertainties of complex hybrid systems can be expressed by specific variables and transformed into deterministic constraints and objectives. Among them, RO can tackle uncertain parameters with unknown distribution by a prespecified uncertainty set and ensure protection against worst-case realizations within the uncertainty set [160]. RO shows its superiorities in seeking conservative planning and operation schemes of HRES under multi-uncertainties. Hybrid methods have become the mainstream method by incorporating the advantages of different methods. For example, a hybrid stochastic-robust optimization model was developed to quantify the impacts of climate change and extreme climate events on energy systems under climate-induced energy uncertainty, with energy flow presented for both high-probability, low-impact and lowprobability, high-impact scenarios [161]. Reliability, economic, environmental and social indicators need to be incorporated into the multiobjective optimization framework, allowing the performances of different indicators to be evaluated to help decision-makers balance the tradeoffs. Metaheuristic algorithms have become effective tools to solve multiobjective optimization problems due to their better accuracy and convergence. Life-cycle assessment needs to be applied for exploring cradle-to-grave life-cycle environmental and ecological effects of system components and their combinations, as well as operation alternatives. Uncertainties in renewable energy potential and demand can lead to a significant performance gap brought by future climate variations and a drop in power supply reliability due to extreme weather events [161]. Resilience and ecological evaluation indicators can be evolved to better measure system and environmental changes. In addition,

222

PART | II Modelling

FIGURE 6.2 Integrated inexact optimization framework. MOP, NLP and MILP mean multiobjective programming, nonlinear programming and mixed-integer linear programming, respectively. SMP, RO, FMP and IMP denote stochastic mathematical programming, robust optimization, fuzzy mathematical programming and interval mathematical programming, respectively.

scenario analysis associated with negative emission technologies can be conducted to explore the carbon sequestration potential under different component combinations and multiple uncertainties and handle the conflicting economic and environmental trade-off of hybrid systems.

6.5

Conclusions

Renewable energy resources play significant roles in energy system transition and climate change mitigation. The integration of renewable energy resources with traditional fossil-based resources - besides storages - creates HRESs. The power management, sizing, capacity and operation of each component should be optimally determined to deliver a reliable and costeffective HRES. This chapter summarized the conventional and inexact

Integrated inexact optimization for hybrid Chapter | 6

223

optimization techniques and their applications in HRES. Previous researches have paid more attention to small-scale hybrid systems under standalone mode with consideration of stochastic uncertainties. As a result, optimization applications for large-scale real cases under multiple uncertainties are still rare. The intermittent and random nature of renewable resources and load demand poses great challenges for effective resource allocation to match production and supply. A variety of influence factors concerning price fluctuation, technology process, policy target and social development increase system complexities with multiple dimensions and layers. Such inherent uncertainties and their interactions have significant influences on the precise design of schemes and operation strategies of hybrid system. Climateinduced extreme weather events and weather variations bring forward the higher demands for the design and operation of HRESs. This chapter proposes an integrated inexact optimization framework for such challenges. By analysing the advantages and disadvantages of optimization techniques, hybrid methods show superiorities in handling the complex optimization problems of hybrid systems under multiple uncertainties. RO can support conservative decisions in real-world applications where changes in solutions are not always easy to implement. Life-cycle assessment, scenario analysis associated with negative emission technologies and resilience evaluation indicators can be conducted to seek low-carbon and robust planning and operation alternatives of hybrid systems. For future research, more emphasis should be placed on the comprehensive hybrid system modelling under multiple uncertainties as well as efficient optimization algorithms, especially for large-scale hybrid systems and climate change adaptation and resilience.

References [1] IPCC: Climate Change 2013. The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK and New York, NY: Cambridge University Press; 2013. [2] UN Environment. Emissions Gap Report 2018. Nairobi: United Nations Environment Programme; 2018. [3] Lund H. Renewable energy strategies for sustainable development. Energy. 2007;32: 912
19. [4] Krishna KS, Kumar KS. A review on hybrid renewable energy systems. Renew Sustain Energy Rev 2015;52:907
16. [5] Singh VK, Singal SK. Operation of hydro power plants
a review. Renew Sustain Energy Rev 2017;69:610
19. [6] Misak S, Prokop L. Operation characteristics of renewable energy sources. Green Energy and Technology. Ostrava: Springer; 2017. [7] Wang XN, Palazoglu A, Farra NHE. Operational optimization and demand response of hybrid renewable energy systems. Appl Energy 2015;143:324
35. [8] Wang JX, Zhong HW, Ma ZM, Xia Q, Kang CQ. Review and prospect of integrated demand response in the multi-energy system. Appl Energy 2017;202:772
82.

224

PART | II Modelling

[9] Roth A, Boix M, Gerbaud V, Montastruc L. A flexible metamodel architecture for optimal design of hybrid renewable energy systems (HRES)-case study of a stand-alone HRES for a factory in tropical island. J Clean Prod 2019;223:214
25. [10] Jafari M, Malekjamshidi Z. Optimal energy management of a residential
based hybrid renewable energy system using rule-based real-time control and 2D dynamic programming optimization method. Renew Energy 2020;146:254
66. [11] Bahramara S, Moghaddam MP, Haghifam MR. Optimal planning of hybrid renewable energy systems using HOMER: a review. Renew Sustain Energy Rev 2016;62:609
20. [12] Trivin˜o PG. Power control based on particle swarm optimization of grid-connected inverter for hybrid renewable energy system. Energy Convers Manag 2015;91:83
92. [13] Wang XN, Palazogluy A, Farra NHE. Operation of residential hybrid renewable energy systems: integrating forecasting, optimization and demand response. American Control Conference, Portland. 2014. [14] Das M, Singh MAK, Biswas A. Techno-economic optimization of an off-grid hybrid renewable energy system using metaheuristic optimization approaches
case of a radio transmitter station in India. Energy Convers Manag 2019;185:339
52. [15] Kim MH, Kim DK, Heo J, Lee DW. Techno
economic analysis of hybrid renewable energy system with solar district heating for net zero energy community. Energy. 2019; 187:115916. [16] Krishan O, Suhag S. Techno-economic analysis of a hybrid renewable energy system for an energy poor rural community. J Energy Storage 2019;23:305
19. [17] Chang KH. A quantile-based simulation optimization model for sizing hybrid renewable energy systems. Simul Model Pract Theory 2016;66:94
103. [18] Lujano-Rojas JM, Dufo-Lo´pez R, Bernal-Agust´ın JL. Probabilistic modelling and analysis of stand-alone hybrid power systems. Energy. 2013;63:19
27. [19] Siddaiah R, Saini RP. A review on planning, configurations, modeling and optimization techniques of hybrid renewable energy systems for off grid applications. Renew Sustain. Energy Rev 2016;58:376
96. [20] Liu YF, Yu SS, Zhu Y, Wang DJ, Liu JP. Modeling, planning, application and management of energy systems for isolated areas: a review. Renew Sustain Energy Rev 2018;82: 460
70. [21] Lian JJ, Zhang YS, Ma C, Yang Y, Chaima E. A review on recent sizing methodologies of hybrid renewable energy systems. Energy Convers Manag 2019;199:112027. [22] Ahmed A, Khalid M. A review on the selected applications of forecasting models in renewable power systems. Renew Sustain Energy Rev 2019;100:9
21. [23] Forough AB, Roshandel R. Lifetime optimization framework for a hybrid renewable energy system based on receding horizon optimization. Energy. 2018;150:617
30. [24] Kusakana K, Vermaak HJ, Yuma GP. Optimization of hybrid standalone renewable energy systems by linear programming. J Comput Theor Nanosci 2013;19:2501
4. [25] Camargo LR, Valdes J, Macia YM, Dorner W. Assessment of on
site steady electricity generation from hybrid renewable energy systems in Chile. Appl Energy 2019;250: 1548
58. [26] Bartolucci L, Cordiner S, Mulone V, Santarelli M. Hybrid renewable energy systems: influence of short term forecasting on model predictive control performance. Energy. 2019;172:997
1004. [27] Li XD, Peng Y, Wang WY. A method for optimizing installation capacity and operation strategy of a hybrid renewable energy system with offshore wind energy for a green container terminal. Ocean Eng 2019;186:106
25.

Integrated inexact optimization for hybrid Chapter | 6

225

[28] Vaccari M, Mancuso GM, Riccardi J, Cantu` M, Pannocchia G. A sequential linear programming algorithm for economic optimization of hybrid renewable energy systems. J Process Control 2019;74:189
201. [29] Forough AB, Roshandel R. Multi objective receding horizon optimization for optimal scheduling of hybrid renewable energy system. Energy Build 2017;150:583
97. [30] Li LY, You SM, Wang XN. Optimal design of standalone hybrid renewable energy systems with biochar production in remote rural areas: a case study. Energy Procedia 2017; 158:688
93. [31] Upadhyay S, Sharma MP. A review on configurations, control and sizing methodologies of hybrid energy systems. Renew Sustain Energy Rev 2014;38:47
63. [32] Luna-Rubio R, Trejo-Perea M, Vargas-V´azquez D, R´ıosMoreno GJ. Optimal sizing of renewable hybrids energy systems: a review of methodologies. Sol Energy 2012;86: 1077
88. [33] Khatod DK, Pant V, Sharma J. Analytical approach for well-being assessment of small autonomous power systems with solar and wind energy sources. IEEE Trans Energy Convers 2010;25:535
45. [34] Jakhrani AQ, Othman AK, Rigit ARH, Samo SR, Kamboh SA. A novel analytical model for optimal sizing of standalone photovoltaic systems. Energy 2012;46:675
82. [35] Fantauzzi M, Lauria D, Mottola F, Scalfati A. Sizing energy storage systems in DC networks: a general methodology based upon power losses minimization. Appl Energy 2017; 187:862
72. [36] Emad D, El Hameed MA, Yousef MT, El Fergany AA. Computational methods for optimal planning of hybrid renewable microgrids: a comprehensive review and challenges. Arch Comput Methods Eng 2019;20:1
23. [37] Mahesh A, Sandhu KS. Hybrid wind/photovoltaic energy system developments: critical review and findings. Renew Sustain Energy Rev 2015;52:1135
47. [38] Tezer T, Yaman R, Yaman G. Evaluation of approaches used for optimization of standalone hybrid renewable energy systems. Renew Sustain Energy Rev 2017;73:840
53. [39] Sharafi M, Elmekkawy TY, Bibeau EL. Optimal design of hybrid renewable energy systems in buildings with low to high renewable energy ratio. Renew Energy 2015;83: 1026
42. [40] Fetanat A, Khorasaninejad E. Size optimization for hybrid photovoltaic
wind energy system using ant colony optimization for continuous domains based integer programming. Appl Soft Comput 2015;31:196
209. [41] Rezvani A. Intelligent hybrid power generation system using new hybrid fuzzy-neural for photovoltaic system and RBFNSM for wind turbine in the grid connected mode. Front Energy 2019;13:131
48. [42] Sharafi M, Elmekkawy TY. Multi-objective optimal design of hybrid renewable energy systems using PSO-simulation based approach. Renew Energy 2014;68:67
79. [43] Weber JA, Gao DW, Gao TL. Affordable mobile hybrid integrated renewable energy system power plant optimized using HOMER Pro. North American Power Symposium (NAPS), Vol.12. IEEE; 2016. p. 344
65. [44] Murugaperumal K, Raj PADV. Feasibility design and techno
economic analysis of hybrid renewable energy system for rural electrification. Sol Energy 2019;188:1068
83. [45] Padro´n I, Avila D, Marichal GN, Rodr´ıguez JA. Assessment of hybrid renewable energy systems to supplied energy to autonomous desalination systems in two islands of the Canary Archipelago. Renew Sustain Energy Rev 2019;101:221
30.

226

PART | II Modelling

[46] Zhou W, Lou CZ, Li ZS, Yang HX. Current status of research on optimum sizing of standalone hybrid solar
wind power generation systems. Appl Energy 2010;87:380
9. [47] Chauhan A, Saini RP. A review on integrated renewable energy system based power generation for stand-alone applications: configurations, storage options, sizing methodologies and control. Renew Sustain Energy Rev 2014;38:99
120. [48] Li Q, Jorge LB, Nam K, Hwangbo S, Rashidi J. Sustainable and reliable design of reverse osmosis desalination with hybrid renewable energy systems through supply chain forecasting using recurrent neural networks. Energy 2019;178:277
92. [49] Nie S, Huang CZ, Huang GH, Li YP, Chen JP, Fan YR. Planning renewable energy in electric power system for sustainable development under uncertainty
a case study of Beijing. Appl Energy 2016;162:772
86. [50] Cai YP, Huang GH, Yang ZF, Tan Q. Identification of optimal strategies for energy management systems planning under multiple uncertainties. Appl Energy 2009;86:480
95. [51] Zeng Y, Cai YP, Huang GH, Dai J. A review on optimization modeling of energy systems planning and GHG emission mitigation under uncertainty. Energies 2011;4:1624
56. [52] Yu L, Li YP, Huang GH. Planning municipal-scale mixed energy system for stimulating renewable energy under multiple uncertainties
the city of Qingdao in Shandong province, China. Energy 2019;166:1120
33. [53] Birge JR, Louveaux F. Introduction to stochastic programming. Springer Series in Operations Research and Financial Engineering. New York, NY: Springer; 1997. [54] Charnes A, Cooper WW, Symonds GH. Cost horizons and certainty equivalents: an approach to stochastic programming of heating oil. Manag Sci 1958;4:235
63. [55] Charnes A, Cooper WW, Kirby P. Chance constrained programming: an extension of statistical method. In: Rustagi JS, editor. Optimizing Methods in Statistics. Columbus: Academic Press; 1972. [56] Abdelaziz FB, Aouni B, Fayedh RE. Multi-objective stochastic programming for portfolio selection. Eur J Oper Res 2007;177:1811
23. [57] Poojari CA, Varghese B. Genetic algorithm based technique for solving chance constrained problems. Eur J Oper Res 2008;185:1128
54. [58] Abdelaziz FB. Solution approaches for the multiobjective stochastic programming. Eur J Oper Res 2012;216:1
16. [59] Sun W, Huang GH, Lv Y, Li G. Inexact joint-probabilistic chance-constrained programming with left-hand-side randomness: an application to solid waste management. Eur J Oper Res 2013;228:217
25. [60] Ji Y, Huang GH, Sun W. Nonpoint-source water quality management under uncertainty through an inexact double-sided chance-constrained model. Water Resour Manag 2015; 29:3079
94. [61] Kamjoo A, Maheri A, Putrus GA. Chance constrained programming using non-Gaussian joint distribution function in design of standalone hybrid renewable energy systems. Energy. 2014;66:677
88. [62] Wu Z, Zeng PL, Zhang XP, Zhou QY. A solution to the chance-constrained two-stage stochastic program for unit commitment with wind energy integration. IEEE Trans Power Syst 2016;31:4185
96. [63] Odetayo B, MacCormack J, Rosehart WD, Zareipour H, Seifi AR. Integrated planning of natural gas and electric power systems. Int J Electr Power Energy Syst 2018;103: 593
602. [64] Zeynali S, Rostami N, Feyzi MR. Multi-objective optimal short-term planning of renewable distributed generations and capacitor banks in power system considering different

Integrated inexact optimization for hybrid Chapter | 6

[65]

[66]

[67] [68] [69] [70] [71] [72] [73] [74]

[75] [76]

[77] [78]

[79] [80] [81]

[82] [83]

227

uncertainties including plug-in electric vehicles. Int J Electr Power Energy Syst 2020;119:105885. Kall P, Meyer J. Stochastic linear programming: models, theory, and computation. International Series in Operations Research & Management Science. New York: Springer; 2005. Shapiro A, Dentcheva D, Ruszczy´nski A. Lectures on stochastic programming: modeling and theory. MOS-SIAM Series on Optimization. Society for Industrial and Applied Mathematics; 2009. Zhou S, Zhang H, Shi N, Xu Z, Wang F. A new convergent hybrid learning algorithm for two-stage stochastic programs. Eur J Oper Res 2020;283:33
46. Kleywegt AJ, Shapiro A, Homem-de-Mello T. The sample average approximation method for stochastic discrete optimization. SIAM J Optim 2001;12:479
502. Huang GH, Loucks DP. An inexact two-stage stochastic programming model for water resources management under uncertainty. Civ Eng Environ Syst 2000;17:95
118. Ermoliev Y. Stochastic quasigradient methods. Numerical Techniques for Stochastic Optimization. New York, NY: Springer; 1988. Pereira MVF, Pinto LMVG. Multi-stage stochastic optimization applied to energy planning. Math Prog 1991;52:359
75. Cheung RK, Powell WB. SHAPE
a stochastic hybrid approximation procedure for twostage stochastic programs. Oper Res 2000;48:73
9. Pflug GC. Scenario tree generation for multiperiod financial optimization by optimal discretization. Math Prog 2001;89:251
71. Kadri AA, Perrouault R, Boujelben MK, Gicquel C. A multi-stage stochastic integer programming approach for locating electric vehicle charging stations. Comput Oper Res 2020;117:104888. Pimentel BB, Mateus GR, Almeida FF. Stochastic capacity planning and dynamic network design. Int J Prod Econ 2013;145:139
49. Li YP, Huang GH, Huang YF, Zhou HD. A multistage fuzzy-stochastic programming model for supporting sustainable water-resources allocation and management. Environ Model Softw 2009;24:786
97. Hoyland K, Wallace SW. Generating scenario trees for multistage decision problems. Manag Sci 2001;47:295
307. Fattahi M, Govindan K, Keyvanshokooh E. A multi-stage stochastic program for supply chain network redesign problem with price-dependent uncertain demands. Comput Oper Res 2018;100:314
32. Dupacov´a J, Gro¨we-Kuska N, Ro¨misch W. Scenario reduction in stochastic programming. Math Prog 2003;95:493
511. Heitsch H, Ro¨misch W. Scenario tree modeling for multistage stochastic programs. Math Prog 2009;118:371
406. Beltran-Royo C, Escudero LF, Monge JF, Rodriguez-Ravines RE. An effective heuristic for multistage linear programming with a stochastic right-hand side. Comput Oper Res 2014;51:237
50. Maggioni F, Pflug GC. Bounds and approximations for multistage stochastic programs. SIAM J Optim 2016;26:831
55. Wang WY, Peng Y, Li XD, Qi Q, Feng P, Zhang Y. A two-stage framework for the optimal design of a hybrid renewable energy system for port application. Ocean Eng 2019; 191:106555.

228

PART | II Modelling

[84] Yu J, Ryu JH, Lee IB. A stochastic optimization approach to the design and operation planning of a hybrid renewable energy system. Appl Energy 2019;247:212
20. [85] Daneshvar M, Mohammadi-Ivatloo B, Zare K, Asadi S. Two-stage stochastic programming model for optimal scheduling of the wind-thermal-hydropower-pumped storage system considering the flexibility assessment. Energy 2020;193:116657. [86] Hafiz F, Queiroz ARD, Fajri P, Husain I. Energy management and optimal storage sizing for a shared community: a multi-stage stochastic programming approach. Appl Energy 2019;236:42
54. ¨ , Yaman H. Multi-stage stochastic programming for demand [87] Sahin ¸ MK, C ¸ avu¸s O response optimization. Comput Oper Res 2020;118:104928. [88] Soyster AL. Convex programming with set-inclusive constraints and applications to inexact linear programming. Oper Res 1973;21:1154
7. [89] Gorissen BramL, Yanıko˘glu ˙I, Hertog DD. A practical guide to robust optimization. Omega. 2015;53:124
37. [90] Ben-Tal A, Ghaoui LE, Nemirovski A. Robust optimization. Princeton Series in Applied Mathematics. Princeton University Press; 2009. [91] Bertsimas D, Brown DB, Caramanis C. Theory and applications of robust optimization. SIAM Rev 2011;53:464
501. [92] Vladimirou H, Zenios SA. Stochastic linear programs with restricted recourse. Eur J Oper Res 1997;101:177
92. [93] Moret S, Babonneau F, Bierlaire M, Mare´chal F. Overcapacity in European power systems: analysis and robust optimization approach. Appl Energy 2020;259:113970. [94] Ranjbar H, Hosseini SH, Zareipour H. A robust optimization method for co-planning of transmission systems and merchant distributed energy resources. Int J Electr Power Energy Syst 2020;118:105845. [95] Tan ZF, Fan W, Li HF, De GJRF, Ma JL, Yang SB, et al. Dispatching optimization model of gas-electricity virtual power plant considering uncertainty based on robust stochastic optimization theory. J Clean Prod 2020;247:119106. [96] Zadeh LA. Fuzzy sets. Inf Control 1965;8:338
53. [97] Zimmermann HJ. Applications of fuzzy sets theory to mathematical programming. Inf Sci 1985;36:29
58. [98] Huang GH, Baetz BW, Patry GG. A gray fuzzy linear-programming approach for municipal solid-waste management planning under uncertainty. Civ Eng Syst 1993;10:123
46. [99] Bellman R, Zadeh LA. Decision making in a fuzzy environment. Manag Sci 1970;17: 141
64. [100] Sobral MM, Hipel KW, Fargugar GJ. A multicriteria model for solid waste management. J Environ Manag 1981;12:97
110. [101] Abd El-Wahed WF, Lee SM. Interactive fuzzy goal programming for multi-objective transportation problems. Omega 2006;34:158
66. [102] Chen SH. Ranking fuzzy numbers with maximizing set and minimizing set. Fuzzy Sets Syst 1985;17:113
29. [103] Arana-Jime´nez M, Blanco V. On a fully fuzzy framework for minimax mixed-integer linear programming. Comput Ind Eng, 218. 2019. p. 170
9. [104] Buckley JJ, Feuring T. Evolutionary algorithm solution to fuzzy problems: fuzzy linear programming. Fuzzy Sets Syst 2000;109:35
53. [105] Sakawa M, Kato K. An interactive fuzzy satisficing method for general multiobjective 0
1 programming problems through genetic algorithms with double strings based on a reference solution. Fuzzy Sets Syst 2002;125:289
300.

Integrated inexact optimization for hybrid Chapter | 6

229

[106] Li SY, Hu CF. An interactive satisfying method based on alternative tolerance for multiple objective optimization with fuzzy parameters. IEEE Trans Fuzzy Syst 2008;16: 1151
60. [107] Hocine A, Kouaissah N, Bettahar S, Benbouziane M. Optimizing renewable energy portfolios under uncertainty: a multi-segment fuzzy goal programming approach. Renew Energy 2018;129:540
52. [108] Faddel S, Aldeek A, Al-Awami ATli, Sortomme E, Al-Hamouz Z. Ancillary services bidding for uncertain bidirectional V2G using fuzzy linear programming. Energy. 2018; 160:986
95. [109] Jayachitra R, Revathy A, Prasad PSS. A Fuzzy Programming approach for formation of virtual cells under dynamic and uncertain conditions. Int J Eng Sci Technol 2010;2: 1708
24. [110] Inuiguchi M, Ramik J. Possibilistic linear programming: a brief review of fuzzy mathematical programming and a comparison stochastic programming in portfolio selection problem. Fuzzy Sets Syst 2000;111:3
28. [111] Inuiguchi M, Tanino T. Portfolio selection under independent possibilistic information. Fuzzy Sets Syst 2000;115:83
92. [112] Tanaka H, Guo PJ, Zimmermann HJ. Possibility distributions of fuzzy decision variables obtained from possibilistic linear programming problems. Fuzzy Sets Syst 2000;113: 323
32. [113] Liou TS, Wang MJ. Fuzzy weighted average: an improved algorithm. Fuzzy Sets Syst 1992;49:307
15. [114] Jime´nez M, Arenas M, Bilbao A, Rodriguez MV. Linear programming with fuzzy parameters: an interactive method resolution. Eur J Oper Res 2007;177:1599
609. [115] Heilpern S. The expected value of a fuzzy number. Fuzzy Sets Syst 1992;47:81
6. [116] Torabi SA, Hassini E. An interactive possibilistic programming approach for multiple objective supply chain master planning. Fuzzy Sets Syst 2008;159:193
214. [117] Zhou Y, Li YP, Huang GH. Integrated modeling approach for sustainable municipal energy system planning and management - a case study of Shenzhen, China. J Clean Prod 2014;75:143
56. [118] Campos L, Verdegay JL. Linear programming problems and ranking of fuzzy numbers. Fuzzy Sets Syst 1989;32:1
11. [119] Liu BD, Iwamura K. Chance constrained programming with fuzzy parameters. Fuzzy Sets Syst 1998;94:227
37. [120] Otto KN, Lewis AD, Antonsson EK. Approximating α-cuts with the vertex method. Fuzzy Sets Syst 1993;55:43
50. [121] Chang NB, Wen CG, Chen YL, Yong YC. A grey fuzzy multiobjective programming approach for the optimal planning of a reservoir watershed. Part A: theoretical development. Water Res 1996;30:2329
34. [122] Inuiguchi M, Ramik J. Possibilistic linear programming: a brief review of fuzzy mathematical programming and a comparison stochastic programming in portfolio selection problem. Fuzzy Sets Syst 2000;111:3
28. [123] Zhang JD, Rong G. Fuzzy possibilistic modeling and sensitivity analysis for optimal fuel gas scheduling in refinery. Eng Appl Artif Intell 2010;23:371
85. [124] Zhou Y, Li YP, Huang GH. Planning sustainable electric-power system with carbon emission abatement through CDM under uncertainty. Appl Energy 2015;140:350
64. [125] Huang XX. Credibility-based chance-constrained integer programming models for capital budgeting with fuzzy parameters. Inf Sci 2006;176:2698
712.

230

PART | II Modelling

[126] Mohammadi M, Noorollahi Y, Mohammadi-ivatloo B. Fuzzy-based scheduling of wind integrated multi-energy systems under multiple uncertainties. Sustain Energy Technol Assess 2020;37:100602. [127] Moradi MH, Hajinazari M, Jamasb S, Paripour M. An energy management system (EMS) strategy for combined heat and power (CHP) systems based on a hybrid optimization method employing fuzzy programming. Energy 2013;49:86
101. [128] Rosso-Cero´n AM, Kafarov V, Latorre-Bayona G, Quijano-Hurtado R. A novel hybrid approach based on fuzzy multi-criteria decision-making tools for assessing sustainable alternatives of power generation in San Andre´s Island. Renew Sustain Energy Rev 2019;110:159
73. [129] Pishvaee MS, Razmi J, Torabi SA. Robust possibilistic programming for socially responsible supply chain network design: a new approach. Fuzzy Sets Syst 2012;206:1
20. [130] Zhou Y, Li YP, Huang GH. A robust possibilistic mixed-integer programming method for planning municipal electric power systems. Int J Electr Power Energy Syst 2015;73:757
72. [131] Dupacova J. Reflections on robust optimization. Lecture Notes in Economics and Mathematical Systems, 458. Springer; 1998. p. 111
27. [132] Ben-Tal A, Nemirovski A. Robust solutions to uncertain linear programs. Oper Res Lett 1999;25:1
13. [133] Ben-Tal A, Nemirovski A. Robust optimization-methodology and applications. Math Progr 2002;92:453
80. [134] Zhang XD, Huang GH, Chan CW, Liu ZF, Lin QG. A fuzzy-robust stochastic multiobjective programming approach for petroleum waste management planning. Appl Math Model 2010;34:2778
88. [135] Amirkhan M, Didehkhani H, Khalili-Damghani K, Hafezalkoto A. Mixed uncertainties in data envelopment analysis: a fuzzy-robust approach. Expert Syst Appl 2018;103:218
37. [136] Li YP, Huang GH, Nie XH, Nie SL. A two-stage fuzzy robust integer programming approach for capacity planning of environmental management systems. Eur J Oper Res 2008;189:399
420. [137] Moore RE. Method and application of interval analysis. Studies in Applied and Numerical Mathematics. Society for Industrial and Applied Mathematics; 1979. [138] Alefeld G, Herzberger J. Introductions to Interval Computations. New York, NY: Academic Press; 1983. [139] Huang GH, Baetz BW, Patry GG. A grey linear programming approach for municipal solid waste management planning under uncertainty. Civ Eng Environ Syst 1992;9: 319
35. [140] Huang GH, Moore RD. Grey linear programming, its solving approach, and its application. Int J Syst Sci 1993;24:159
72. [141] Huang GH, Cao MF. Analysis of solution methods for interval linear programming. J Environ Inform 2011;17:54
64. [142] Fan YR, Huang GH. A robust two-step method for solving interval linear programming problems within an environmental management context. J Environ Inform 2012;19:1
9. [143] Wang SX, Yuan SC. Interval optimization for integrated electrical and natural-gas systems with power to gas considering uncertainties. Int J Electr Power Energy Syst 2020; 119:105906. [144] Bai LQ, Li FX, Cui HT, Jiang T, Sun HB, Zhu JX. Interval optimization based operating strategy for gas-electricity integrated energy systems considering demand response and wind uncertainty. Appl Energy 2016;167:270
9.

Integrated inexact optimization for hybrid Chapter | 6

231

[145] Taghizadeh M, Bahramara S, Adabi F, Nojavan S. Optimal thermal and electrical operation of the hybrid energy system using interval optimization approach. Appl Therm Eng 2020;169:114993. [146] Wang XQ, Huang GH. Violation analysis on two-step method for interval linear programming. Inf Sci 2014;281:85
96. [147] Ji L, Niu DX, Xu M, Huang GH. An optimization model for regional micro-grid system management based on hybrid inexact stochastic-fuzzy chance-constrained programming. Int J Electr Power Energy Syst 2015;64:1025
39. [148] Kwakernaak H. Fuzzy random variables
I. definitions and theorems. Inf Sci 1978;15: 1
29. [149] Kwakernaak H. Fuzzy random variables
II. Algorithms and examples for the discrete case. Inf Sci 1979;17:253
78. [150] Puri ML, Ralescu DA. Fuzzy random variables. J Math Anal Appl 1986;114:409
22. ´ , Lo´pez-D´ıaz M, Ralescu DA. Overview on the development of fuzzy random [151] Gila MA variables. Fuzzy Sets Syst 2006;157:2546
57. [152] Liu BD. Theory and practice of uncertain programming. Studies in Fuzziness and Soft Computing. Berlin: Springer; 2002. [153] Zahiri B, Torabi SA, Tavakkoli-Moghaddam R. A novel multi-stage possibilistic stochastic programming approach (with an application in relief distribution planning). Inf Sci 2017;385
386:225
49. [154] Chen W, Tan SH, Yang DQ. Worst-case VaR and robust portfolio optimization with interval random uncertainty set. Expert Syst Appl 2011;38:64
70. [155] Chen N, Yu DJ, Xia BZ, Ma ZD. Topology optimization of structures with interval random parameters. Comput Methods Appl Mech Eng 2016;307:300
15. [156] Landil D, Castorani V, Germani M. Interactive energetic, environmental and economic analysis of renewable hybrid energy system. Int J Interact Des Manuf 2019;13:885
99. [157] Li LY, Yao ZY, You SM, Wang CH, Chong C, Wang XN. Optimal design of negative emission hybrid renewable energy systems with biochar production. Appl Energy 2019;243:233
49. [158] Abedi S, Alimardani A, Gharehpetian GB, Riahy GH, Hosseinian SH. A comprehensive method for optimal power management and design of hybrid RES-based autonomous energy systems. Renew Sustain Energy Rev 2012;16:1577
87. [159] Muh E, Tabet F. Comparative analysis of hybrid renewable energy systems for off-grid applications in southern Cameroons. Renew Energy 2019;135:41
54. [160] Moret S, Babonneau F, Bierlaire M, Mare´chal F. Decision support for strategic energy planning: a robust optimization framework. Eur J Oper Res 2020;280:539
54. [161] Perera ATD, Nik VM, Chen DL, Scartezzini JL, Hong TZ. Quantifying the impacts of climate change and extreme climate events on energy systems. Nat Energy 2020;5: 150
9.

This page intentionally left blank

Chapter 7

Large-scale integration of variable renewable resources R Go´mez-Calvet1, A.R. Go´mez-Calvet2 and J.M. Mart´ınez-Duart3 1

Business Department, Faculty of Social Sciencies, Universidad Europea de Valencia, Valencia, Spain, 2Finance Department, Facultad de Econom´ıa. Avda, Tarongers, Valencia, Spain, 3 Departamento de F´ısica Aplicada, C-XII, Universidad Auto´noma de Madrid, Campus de Cantoblanco, Madrid, Spain

Chapter Outline 7.1 Introduction 7.2 Climate change and greenhouse gas emissions trends 7.3 Global renewable power deployment 7.4 High penetration of renewable sources in the power sector 7.4.1 Optimal development of nondispatchable resources (solar and wind) 7.4.2 Surplus and backup powers—curtailment 7.4.3 Energy storage

7.1

233 234 238 239

241 243 244

7.5 Main strategies for the 2030 European energy transition 7.5.1 Coal phase-out 7.5.2 Decrease in renewable energy costs 7.5.3 International interconnections 7.5.4 Digitalization and smart grids 7.5.5 Demand response Acknowledgements References

247 247 248 251 253 253 254 254

Introduction

The World Climate Summit recently held in Madrid (November 2019) has been focussed in seven key areas. Among these areas, four of them are directly related to energy, namely: clean energy infrastructure, decarbonization of the energy matrix, E-Mobility and advanced transportation, and smart cities and green buildings. All these topics imply a profound change in the generation, transformation, storage, and use of energy. The intensive use of fossil fuels during the last decades has led to a significant increase of Greenhouse Gas (GHG) concentrations in the atmosphere and global

Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00007-8 © 2021 Elsevier Inc. All rights reserved.

233

234

PART | II Modelling

warming [1]. Climate change is predicted to impact livestock, oceans, freshwater, and terrestrial life from plankton, corals and forests [2]. To secure a sustainable future it is necessary to analyse the drivers on the increase of CO2 and shift current trends. Economic development and population growth are among these drivers, and the intensive use of fossil fuel combustion is the critical issue [3,4]. Currently the energy sector and transport sector heavily relay on fossil fuels as the primary source of energy, and consequently they represent the main source of pollution and GHG emissions. The replacement of fossil fuels and low-carbon renewable technologies, jointly with the improvement of energy efficiency, present the only alternatives to implement long-term sustainable solutions. Transport sector is currently the primary user of petroleum, consuming about 50% of oil used in the world [5]. However, the substitution of this primary source is clearly dependent on the power sector: the development of the electric vehicle (EV), hybrid electric vehicle (HEV), and fuel cell vehicles mainly rely on electricity as the most promising energy vector. Since electricity is a secondary source of energy that needs other primary sources, the future of the electricity production will be highly influenced by the transport sector requirements. According to IRENA [6], deep electrification of the end-use sectors could reduce the emissions from buildings, transport, and industry by 25%, 54%, and 16%, respectively. It can be deduced from these figures that transport presents the greatest abatement opportunity. Traditionally power systems have been designed to deal with variable loads. The implementation of variable renewable sources (VRES), such as wind and solar, is adding supply side variability and further uncertainty on the power system, raising new challenges for utilities and system operators. In this line, this chapter analyses in detail the challenge of large integration of VRES and CO2 emission mitigation in the electricity power sector.

7.2

Climate change and greenhouse gas emissions trends

Analyses of anthropogenic emissions during the last century have raised awareness about their impact in global warming and subsequent changes in climate. Gases that trap heat in the atmosphere are called GHG and can be categorized according to their global warming potential (GWP). This indicator shows how long the gases remain on the atmosphere and how strong their energy absorption is. The most relevant GHGs are: Carbon dioxide (CO2). The main source of this substance is through burning fossil fuels (coal, natural gas, and oil), solid wastes, trees, and other biological materials, and also from the manufacture of materials like cement and other chemical reactions. Part of this gas is removed from the atmosphere by absorption from plants as part of the biological carbon cycle.

Large-scale integration of variable renewable resources Chapter | 7

235

However, this removal has not been enough to contain the increase the concentration of CO2 during the last century. According to the U.S. Environmental Protection Agency [7], 82% of U.S. GHG emissions in 2017 are due to CO2. Methane (CH4). Emissions of this substance come mainly from the production and transport of coal and extraction of natural gas and oil. They also come from livestock and agriculture and from the decay of organic waste. These emissions represent about 10% of total GHG emissions [7]. Nitrous oxide (N2O). Nitrous oxide is mainly emitted from agriculture and industrial activities, jointly with combustion of fossil fuels and solid waste and also from the treatment of waste water. It has been estimated a 6% contribution of N2O in U.S. total GHG emissions. Fluorinated gases. The most relevant fluorinated gases are hydrofluorocarbons, perfluorocarbons, sulphur hexafluoride, and nitrogen trifluoride. All these gases account for high GWP and their share in U.S. total GHG emissions is about 3%. Regarding other pollutants, it is important to bear in mind that a recent study from the International Energy Agency [8] has raised awareness about the negative impact on pollution of biomass burning. In Europe, particulate matter concentration directly released into the air and emissions of volatile organic compounds (VOCs) has been increased because of the growth in biomass burning since 2005 (PM2.5 increased by 11%, PM10 by 7% and VOCs by 4%). When we evaluate the origin of GHGs we found that electricityderived heat generation and transport are responsible for the largest share of emissions. The decarbonization of transport will probably shift emissions from this sector to the power generation one, and this latter sector will play a key role in the future evolution of GHG emissions. In Fig. 7.1 we depict the historic contribution of the different sectors to GHG emissions. This figure shows a trend change to a steady state in overall emissions during the period 2014
17. However, the latest report from the BP Statistical Review of World Energy 2019 [10] indicates that global energy consumption and carbon emissions have grown 2.9% and 2%, respectively, during 2018. This would be the fastest growth for the last 7 years. This publication also places emphasis on the mismatch between hopes and reality. Recent studies from the same organization indicate that the upwards trend during 2018 might be related to weather effects with a combination of unusual large number of hot and cold days across many of the world’s major demand countries. If this causality is confirmed, we would be facing a new adverse effect of climate change in GHG emissions. Focusing attention on GHG emissions at a country level, in the summary depicted in Fig. 7.2, we find China as the largest emitter, followed by USA and India. However when we compare emissions per capita, we find that developed countries such as USA or Canada contribute the largest share of

236

PART | II Modelling

35,000 30,000

Electricity and heat Industry

Other energy industries

Commerc./Pub. Serv.

Agriculture

Transport

25,000 20,000 15,000 10,000 5000

1990

1994

1998

2002

2006

2010

2014

2017

FIGURE 7.1 World GHG emissions from fuel consumption by sector. Units: MtCO2. GHG, Greenhouse gas. Adapted from IEA. Data and statistics ,https://www.iea.org/data-andstatistics.; 2019a [9].

China (main land): 8501 United States 5700 India 1700 Russian Federation 1614 Japan 1214 Germany 832 South Korea 566 Islamic Republic of Iran 564 Canada 556 Saudi Arabia 518 United Kingdom 512 0

2000

4000

6000

8000

FIGURE 7.2 Ranking of the world’s countries by 2018 total CO2 emissions from fossil-fuel burning, cement production, and gas flaring. Units: Million metric tons of carbon. Length of the bar stand for the absolute amount and colouring of the bar is proportional to CO2 emission per capita. Authors using data from The Atlas team. Global carbon atlas ,http://www.globalcarbonatlas.org.; 2020 [accessed 20.1.2020] [11] and emission per capita from Canadell P, Sharifi A. Global carbon project ,https://www.globalcarbonproject.org.; 2020 [accessed 01.04.20] [12].

237

Large-scale integration of variable renewable resources Chapter | 7

China

1000

United States India Southeast Asia World European Union

800 600 400 200

2000

2005

2010

2015

2020

2025

2030

2035

2040

FIGURE 7.3 Carbon intensity of electricity generation in selected regions. Solid line: observed, dashed line: predicted under the sustainable development scenario, 2000
40. Units: gCO2/kWh. IEA. Tracking power, International Energy Agency, Paris ,https://www.iea.org/reports/tracking-power-2019 . ; 2019b [13].

emissions. The desirable trade-off between economic development and low GHG emissions per capita does not apply in these regions. Since the largest sources of GHG emissions is the power sector, it is a common reference to use the carbon intensity of electricity generation1 as a parameter for evaluating the performance of power generation. In Fig. 7.3 we can see that China, India, and Southeast Asia present the highest intensity. However, except Southeast Asia, almost all regions present decreasing slopes during the last decade in the carbon intensity. A recent report from the International Energy Agency [13] presents the expected trend shown in Fig. 7.3 and predicts a carbon intensity below 200 gCO2/kWh for most of the areas, and almost zero intensity for Europe and other main economies for 2040. The EU is a leader in renewable energy deployment and implementation of environmental policies. The European energy system is rapidly shifting to a low-carbon and resource-efficiency path with a forecast of the lowest carbon intensity for 2040 (see Fig. 7.3). EU efforts to double the share of renewable energy in its consumption have paid off, having greatly reduced the amount of fossil fuels used and their associated GHG emissions [8]. In this context, renewable energies development should account for the whole life cycle impact and promote those sources with less impact on the environment. Among these sources, wind and solar present the best alternatives for Europe. The substitution of fossil fuels by renewables has avoided 1. Carbon intensity, or aggregate carbon intensity of electricity, is defined as the ratio of total CO2 emissions from fossil fuels in electricity production to the total electricity produced in the country.

238

PART | II Modelling

GHG emissions, and most of the substitution has taken place in energyintensive industrial sectors under the EU Emission Trading System (EU-ETS), accounting for three quarters of the total savings in EU GHG emissions in 2018 [8]. Currently the sector showing the largest decrease in emissions is the power sector, compared with transport, heating, and industry. At a country level, Germany, Italy, and United Kingdom have shown the largest absolute reduction in domestic fossil fuel use and GHG emissions. However, these reductions have been more effective in terms of relative reductions, in Denmark, Finland, and Sweden, where the share of renewables has increased more rapidly during 2018.

7.3

Global renewable power deployment

Traditionally hydropower has been the dominant renewable source for most of the regions. In Table 7.1 we present the share of renewable energy for the main world regions, and there is no doubt that many countries of Latin America are greatly using hydropower. The potential energy of waterfalls can be used to produce electricity from this natural resource. Also the use of dams creates a reservoir that can control the flow through the generator units and hence electricity generation. Water can also be used to store surplus energy from other generation sources by pumping water to a higher level, that is, the electrical power for pumping water is stored as the gravitational potential energy of water. During peak hours, with high demand of electricity, water is allow to flow

TABLE 7.1 Share of renewable energy in electricity production (including hydro) (%), in 2005, 2010, and 2015. 2005

2010

2015

Africa

16.90%

17.40%

18.90%

Asia

13.90%

16.10%

20.30%

CIS

18%

16.70%

16.10%

Europe

20.10%

25.70%

34.20%

Latin America

59.30%

57.70%

52.40%

Middle East

4.30%

2%

2.20%

North America

24%

25.80%

27.70%

Pacific

17.90%

18.60%

25.00%

Source: Erias A, Karaka C, Grajetzki C, Carton J, Paulos M, Jantunen P, et al. World energy resources 2016. World Energy Council 2016: 6
46 [14].

Large-scale integration of variable renewable resources Chapter | 7

239

back to a lower height through turbines and convert it back its gravitational potential energy of water in electric energy. In addition to hydropower, other renewable sources are used. Among these sources, wind and solar photovoltaics account for the second and third in the generation ranking. Fig. 7.4 presents the top 10 in the rank of countries with largest VRES generation, and Table 7.1 shows the percentage of renewable energy in electricity production for the most important world regions. In this summary we find strong increase of the renewable share in Europe and Asia. On the contrary regions such as Commonwealth of Independents States (CIS), Latin America, and Middle East show a decreasing trend. During the last decades there has been a continuous deployment of renewable energy plants; among them the amount of electricity generated is growing with different intensities. The source that shows the largest increase is solar photovoltaics followed by wind generation. Fig. 7.5 summarizes the world generation of electricity from renewable sources in 2007 and 2017, showing the total worldwide amount jointly with the growth rate.

7.4 High penetration of renewable sources in the power sector A relevant question in the renewable development strategy at a country level is finding the optimal mix of generation sources in order to satisfy the demand while minimizing the need of backup and storage. As shown before,

Share of renewables (%)

60

40

Wind

Solar PV

20

FIGURE 7.4 Share of electricity generation from variable renewable energy. Top 10 countries, 2017. Authors using data from Sawin JL, Sverrisson F, Rutovitz J, Dwyer S, Teske S, Murdock HE, et al. Renewables 2018-Global status report. A comprehensive annual overview of the state of renewable energy. Advancing the global renewable energy transition-highlights of the REN21 renewables 2018 global status Report in perspective; 2018 [15].

240

PART | II Modelling

32% increase Hydropower 2017: 4158.1 Hydropower 2007: 3158.1

575% increase Wind 2017: 1134.4 Wind 2007: 168.1

106% Bioenergy 2017: 495.3 Bioenergy 2007: 240.2

5286% Solar 2017: 437.2 Solar 2007: 8.1 Geothermal 2017: 85.9 Geothermal 2007: 62.6 Marine 2017: 1.0 Marine 2007: 0.5 0

1000

2000

3000

4000

FIGURE 7.5 Electricity generated by renewable sources at worldwide level in 2007 and 2017. Units: TWh. Authors using data from IRENA. Renewable energy statistics 2019, ,https://www. irena.org/Statistics/Download-Data.; 2019b [16].

wind (mainly onshore) and solar PV are currently the most relevant VRES worldwide, but both sources present differences in their generation patterns; while solar PV presents smaller time variance, that is higher predictability, it does not produce electricity during the night. On the contrary, wind has larger variance but usually produces day and night. To get a better idea of these features, Figs. 7.6 and 7.7 depict several real examples of fan charts.2 In these pictures we can see the complementarity between solar PV and wind: solar PV generates more electricity in spring/summer, but wind is more productive in fall/winter. These figures also show interesting features about the variance of generation. As any random variable is scaled up by a given factor (such as an increase in generation capacity), the variance of this variable scales up by a new figure that is the square of the scaling factor. In the particular case of Fig. 7.7, during 2019 there has been an increase in total installed PV capacity and consequently the amount of power has grown but also its variance has experienced a larger increase (see for instance, differences between summer period in 2018 and 2019).

2. A fan chart is a plot that visualize the distribution that can aid in communicating the degree of underlying uncertainty in the evolution of certain variable. Using percentile distribution, shading fan charts focus the attention towards the whole distribution away from a single central measure. Fan charts were first introduced by the Bank of England for their inflation forecasts [18] and have become a standard method to display uncertainty in other fields such as climate science [19] and demography [20].

241

Large-scale integration of variable renewable resources Chapter | 7 2018 January−March

2018 April−June

15,000

2018 July−September

15,000 90% 80% 70%

10,000

6 5 0%

40% 30% 20%

5000

15,000

10,000

10,000

0 0

4

8

12

16

20

15,000

10,000

24

90% 7 0% 60% 5 0% 3 0% 20% 10%

5000

0 0

4

2019 January−March

8

12

16

20

24

5000

20% 10%

0 0

2019 April−June

15,000

90% 80% 70% 60% 50% 40% 30%

90% 80% 70% 60% 5 4 0% 30% 20% 10%

5000

10%

0

2018 October−December

4

8

12

16

20

24

0

2019 July−September

4

8

12

16

20

15,000

15,000

15,000

10,000

10,000

10,000

90% 80%

90%

10,000

80% 70% 60% 50% 40% 30% 20% 10%

5000

0

90% 80% 70% 60% 5 4 0% 3 2 0% 10%

5000

0 0

4

8

12

16

20

24

2019 October−December

24

90% 80% 70% 6 0% 4 0% 3 2 0% 10%

5000

0 0

4

8

12

16

20

24

70% 60% 50% 40% 30% 20% 10%

5000

0 0

4

8

12

16

20

24

0

4

8

12

16

20

24

FIGURE 7.6 Seasonal hourly (24 h) wind generation distribution for 2018 and 2019. Units in vertical axis in MW. Authors based on data from Red Ele´ctrica Espan˜ola. Sistema de informacio´n del operador del sistema ele´ctrico en Espan˜a, ,https://esios.ree.es/en.; 2020 [accessed 01.11.20] [17].

2018 January−March

2018 April−June

2018 July−September

2018 October−December

5000

5000

5000

5000

4000

4000

4000

4000

3000

3000

3000

3000

2000

2000

2000

2000

1000

1000

1000

0

2 7 8 5 6 1 4 3 9

4

8

12

16

20

9 8 7 2 3 4 5 6 1

0

24

4

2019 January−March

8

12

16

20

1000 9 1 2 3 4 5 6 7 8

0

24

4

2019 April−June

8

12

16

20

4

2019 July−September

5000

5000

5000

4000

4000

4000

4000

3000

3000

3000

3000

2000

2000

2000

2000

1000

1000

1000

5 4 6 8 1 7 2 3 9

4

8

12

16

20

24

2 9 3 4 5 6 7 8 1

0 4

8

12

16

20

24

8

12

16

20

24

2019 October−December

5000

0

6 7 5 8 9 1 4 2 3

0

24

1000 6 7 9 8 5 2 3 4 1

0 4

8

12

16

20

24

5 7 9 1 2 3 4 6 8

0 4

8

12

16

20

24

FIGURE 7.7 Seasonal daily (24 h) solar photovoltaic generation distribution for 2018 and 2019. Units in vertical axis in MW. Authors based on data from Red Ele´ctrica Espan˜ola. Sistema de informacio´n del operador del sistema ele´ctrico en Espan˜a, ,https://esios.ree.es/en.; 2020 [accessed 01.11.20] [17].

7.4.1 Optimal development of nondispatchable resources (solar and wind) Since VRES sources are temporal variables, depending on external weather conditions, in the deployment of very large percentages, it makes sense to propose an optimal mix analysis of VRES that may help to reduce the differences between instantaneous energy produced and energy demanded.

242

PART | II Modelling

A possible way to perform this optimization process is to consider the electric network as a unique node and use historical data to build a linear optimization programme. This methodology has been developed in Refs. [21,22]. For the purpose of illustrating this procedure, we carry out an example based on data of the Spanish grid distribution operator. Based on historic generation data from VRES on hourly basis of one year (2017)3, and aiming to optimally scale VRES to accomplish 2030 targets [23], the optimization problem looks for the simultaneous minimization of both, the excesses and deficiencies of generation, taking as a reference the instant demand. That is, we aim to determine the optimal VRES power mix that satisfies the generation need and minimizes the need of back up and also the excess generation that would be needed to be stored or lost in case of lack of proper storage capabilities. According to the above approach, we propose an optimization programme that minimizes both the need of backup sources (shortcuts) and the excess of electricity production (surpluses). Taking hourly VRES data generation of one natural year, that is 8760 observations, we look for the optimal combination of installed capacities of wind, solar PV, and solar CSP that would minimize the total algebraic sum of the slacks (surpluses, si and shortcuts, di). The mathematical linear optimization programme for this particular problem would therefore be: ! 8760 8760 X X  min si 1 di β 5 λ;s;d i51

i51

subject to xi  λ1 1 yi  λ2 1 zi  λ3 2 si 1 di 5 Demandi ði 5 1; 2; 3. . .; 8760Þ λ1 ; λ2 ; λ3 ; si ; di $ 0

ð7:1Þ

where: xi, yi, zi are the observed generation from wind, solar PV, and solar CSP, respectively, in hour ‘i’ (or period ‘i’), and λ1, λ2, λ3 are the optimal multipliers for wind, solar PV, and solar CSP, respectively, that we are searching for in the evaluated period. DemandI corresponds to the expected power demand for the hour i. In this example, the programme has 17,523 variables (8760 3 2 slacks 1 3 multipliers) and 26,283 constraints (8760 hourly constraints 1 nonnegativity constraints of all variables). By solving this optimization problem [Eq. (7.1)], a key result is the magnitude of the multipliers (λi) that provide the best trade-off between all slacks so as to minimize the total sums for the whole set of hours of the year. The rest of variables also 3. Both periods can be modified and/or extended, but it is important to bear in mind that we must cover the four seasons of the year, and also that selecting shorter periods between measures, for example, on a 10 minutes basis, will drastically increase computation requirements.

Large-scale integration of variable renewable resources Chapter | 7

243

TABLE 7.2 First row is scale factors for each renewable (λ1, λ2, and λ3). Following rows are comparative yearly generation and installed capacity of VRES. Data are results from optimization analysis in the model of Eq. (7.1). Σ VRES

Wind

Solar PV

Solar CSP

Multiplicative VRES factor

2.01 (λ1)

1.85 (λ2)

4.05 (λ3)

Produced in 2017 (GWh)

47,427.3

7,972.4

5,343.1

60,742.8

Proposed 2030 generation (GWh)

95,328.8

14,748.9

21,639.5

131,717.2

Differences (GWh)

47,901.5

6,776.5

16,296.4

70,974.4

Installed capacity in 2017 (GW)

22,922

4,439

2,304

29,665

Proposed 2030 install. capacity (GW)

46,073

8,212

9,331

63,616

Proposed increase (%)

100.1%

85%

305%

114,45%

contain key information about backup requirements and excess of generation. In particular, surplus results can be used to assess about the size of the storage system, and shortcuts about backup requirements. For further details the reader can refer to Go´mez-Calvet et al. [22]. The results obtained for this particular exercise for the VRES multipliers are shown in the first row of Table 7.2. At this point we should note that the additional proposed VRES capacity is (λi 2 1) multiplied by the existing capacity for each source from Table 7.2. In order to better visualize the obtained results, we have depicted in Fig. 7.8, for a randomly chosen week4, the demand (black curve) and the model output (red curve). In this figure we can appreciate the crossing between both curves. As shown above, this model minimizes the area between both curves, therefore it minimizes the shortages and surpluses as initially desired.

7.4.2

Surplus and backup powers—curtailment

The amount of energy generated by variable renewable sources (VRES) can be considered as a random variable. At very high shares of VRES, electricity produced exceeds demand and needs to be stored over hours, days, weeks or even months. Since it cannot be fully predicted and storage of energy is expensive, and sometimes not possible, electricity systems require great flexibility. 4. Fourth week of the year, from 22nd January until 28th January.

244

PART | II Modelling

40

GW

30 20 10 0 Monday

Tuesday Wednesday Thursday

Friday

Saturday

Sunday

FIGURE 7.8 Comparative plot between current demand (black line) and new generation based on the substitution of coal by the optimized VRES mix (red line) proposed in our model. The sample corresponds to the week between hour 528 and hour 696 of the year. VRES, variable renewable sources.

Nowadays, for most of the countries power systems, solar and wind power still have limited impact on grid operation. However, as the share of VRES rises, the grid needs not only a more flexibility service, but potentially a different mix that favours the quick response of a backup system (for instance, gas turbine power plants availability) and response capabilities of electricity storage. Lack of an adequate system support may lead to unstable supply quality (voltage or frequency problems and eventually blackouts). On the contrary, lack of proper storage capability forces the operator to curtailment, that is, disconnect some VRES from the grid, wasting the corresponding energy. Following the example based on the Spanish case for 2017 in Fig. 7.9 we depict a bivariate density plot with the analysis of the surplus intervals. Based on these results we can also predict storage requirements for different coverage levels.

7.4.3

Energy storage

For the integration of high shares of VRES it is needed electricity storage to synchronize the supply to the electricity demand. This is especially true in the case of wind which in intervals of the order of one hour its intensity can change by a factor of almost one hundred. Also, in the case of countries with high shares of solar energy it might be convenient to accumulate the energy during the central hours of the day, so that it can be later used when the solar resource weakens. Evidently depending on the length of the period of time in which the energy is stored one could speak of hourly, daily or even monthly storage. Next, we will briefly describe those storage technologies which are most frequently used for the high integration of VRES.

Large-scale integration of variable renewable resources Chapter | 7

245

60 54 GWh (75% storage coverage)

50

GWh

40 32 GWh (50% storage coverage)

30 20 10 0 0

1

2

3

4

5

6

7

8

9

10

11

12

h

FIGURE 7.9 Bivariate kernel density plot of stored energy (GWh) and duration (h) for storage assessment based on computed surplus slacks from the optimization model of Eq. (7.1). Black curves: correspond to 25%, 50%, and 75% densities. Colour grades: 20%, 40%, 60%, and 80% densities respectively.

7.4.3.1 Pumped-storage hydropower At present, the most efficient technique for storing large amounts of energy for uses in the high integration of VRES is the pumped-hydroelectric energy storage (PHES) which is based on gravitational potential energy storage. The physical mechanism on which it is based consists of pumping water to a high level reservoir which is later released through a set of several electrical turbines at the desired moments. This kind of very effective storage is often used in many European mountainous countries with large altitude reliefs (Italy, Spain, Switzerland). It is important to emphasize that there is in the world about 170 GW of hydropumping which represents about 99% of all the world storage capacity leaving batteries and other storage technologies with a much smaller share (Fig. 7.10). 7.4.3.2 Batteries Although there are many kinds of batteries for energy storage, we will focus ourselves on the lithium (Li)-ion types which present a very bright future for both grid and automotive applications. Besides, they can withstand a very large number of cycles within their lifetime and show very low selfdischarge rates. In addition, their round-trip efficiency ranges between 92% and 96% [25]. In Fig. 7.11 it is shown that the evolution of the costs of storing one kWh of electricity using Lithium batteries, since 2010 with a forecast until 2030. It is interesting to observe that according to IRENA [27] in

246

PART | II Modelling

169,557

Pumped-storage 1 ,629

Lithium-ion battery

931

Flywheels Compressedair

407

189

Sodium sulfur battery 75

Lead Acid battery

72

Flow battery 10

1

10 2

10 3

10 4

10 5

10 6

Storage power worldwide installed (MW) FIGURE 7.10 As of 2018, storage installed capacity in MW by technology type worldwide. Horizontal axis with the logarithmic scale. Authors based on data from Mongird K, Viswanathan VV, Balducci PJ, Alam MJE, Fotedar V, Koritarov VS, et al. Energy storage technology and cost characterization report. Technical report, Pacific Northwest National Lab (PNNL), Richland, WA (United States); 2019 [24].

2030 2029 2028 2027 2026 2025 2024 2023 2022 2021 2020 2019 2018 2017 2016 2015 2014 2013 2012 2011 2010

Forecast (2018 and beyond)

0

200

400

600

800

1000

$USD/kWh stored FIGURE 7.11 Average cost of storing a kWh of electricity using Lithium-ion batteries. Figures for 2018 and beyond are projections. Bloomberg New Energy Finance. New energy outlook 2019, ,https://about.bnef.com/new-energy-outlook/.; 2019 [26].

Large-scale integration of variable renewable resources Chapter | 7

247

the 2030 decade there will be several countries, especially in Europe, with shares of renewables in the range 70%
80% and therefore the storage needs of VRES will become crucial to smooth supply fluctuations over relatively long periods of time. For this reason one of the main priorities for the European Commission is the creation of a ‘strong, competitive and sustainable battery industry in Europe’ [28].

7.4.3.3 Hydrogen The potential of hydrogen to develop a low carbon energy system is currently evolving. In an scenario with high surplus of energy from renewable sources, the opportunity to produce hydrogen can play the role of a promising energy vector that may displace fossil fuels and nuclear power while maintaining a stable electricity grid and meeting the energy needs of a modern economy [29]. The European Parliament and the European Commission have recently established a set of ten priorities for the development of the European Energy Transition 2030 [28]. The eighth priority is to establish the foundation for a scalable green hydrogen economy. Within the upcoming gas package, it is proposed to rise gaseous fuels from 2% in 2022 up to 10% in 2030. This will account for some 370 TWh in 2030. A subquota of one third of this amount should be based on green hydrogen. This will ensure that the EU green hydrogen production and electrolyser capacity would grow to at least 120
125 TWh and 30 GW, respectively, by 2030 [28] with the aim to fully decarbonize fuels by 2050. Following the priorities mentioned above, the last one is prioritize energy transition in the new European budget for 2021
27 and within this priority there is a plan to set EU budget to support decarbonization of the economy and provide funds for the European Energy Transition. This budget accounts for at least 1 billion euros in support from the Horizon Europe and the ETS Innovation Fund for research and innovation projects based on 100% hydrogen grids and hydrogen-based industrial decarbonization technologies. This budget will help countries throughout Europe to invest in the clean energy transition, opening up major opportunities for research and development.

7.5 7.5.1

Main strategies for the 2030 European energy transition Coal phase-out

Coal combustion has historically been at the cornerstone of many country electricity mixes, but due to its high pollution emissions, the Emission Trading Systems, and its low efficiency conversion, it is being been displaced by other generation sources. In the particular case of Europe, coal generation has shown its highest fall in 2019 which has been attributed to the EU’s carbon cap-and-trade

248

PART | II Modelling

system, the Emission Trading System (EU-ETS) Scheme implemented at the beginning of the last decade. During the financial crisis the price on the EU carbon market was rather low, but it has risen from around h5 in 2017 to around h25 per tonne of emitted CO2 in 2019 [30], pushing coal power plants to halt generation. As an evidence of this fact, Fig. 7.12 shows the evolution of coal generation for Spain during a 5-year interval (the period between January 2015 to December 2019) in which it can be observed its progressive diminution during 2019. In addition, strict emissions regulations will make it very difficult to return to coal as a main source of power in Europe. Further analysis of power generation shows that this fall in coal generation has already been replaced mainly by additional renewables sources and natural gas. Not all European countries have shown a similar trend, and the fall in coal power generation has been much smaller in Central and Eastern Europe countries. Focusing attention in other continents, it has been noted a flattering of coal growth generation in China and ‘a sharp turnaround in India, where coal power output is on track to fall for the first time in the last three decades’ [31].

7.5.2

Decrease in renewable energy costs

Coal Power Generation (MWe)

There is no doubt that cost is the principal barrier for the widespread of renewable sources. Until recently generating electricity using renewables sources has been more costly than generating it with fossil fuels or nuclear power plants. In addition, the nature of most of undispatchable renewable sources has also led to additional integration costs (need of backup, storage or more transmission infrastructures). However, during the last three decades, there has been clear concerns about the need to account for all the negative externalities that traditional power plants produce. In this context, the implementation of emission cap and trade schemes for the most harmful pollutants and CO2 emissions has greatly helped to recognize the real cost of the 10,000 8000

Year 2016

Year 2017

Year 2018

35,426.71 GWh

42,751.70 GWh

35,435.35 GWh

Year 2019 11,091.76 GWh)

6000 4000 2000 Year 2015 51,358.36 GWh

0 0

10,000

20,000

30,000

40,000

Hours from 1/1/2015 to 31/12/2019

FIGURE 7.12 Coal generation in Spain during the last 5 years. Red line show a simple moving average taking one year (365 previous days) as interval for the analysis. Authors based on data from Red Ele´ctrica Espan˜ola. Sistema de informacio´n del operador del sistema ele´ctrico en Espan˜a, ,https://esios.ree.es/en.; 2020 [accessed 01.11.20] [17].

Large-scale integration of variable renewable resources Chapter | 7

249

different sources. If emissions are capped and an emission trading system is established, there could be a double beneficial effect: on the one hand, the cost of fossil fuel generation increases due to the emission allowances needed which leads to the internalization of externalities reflecting the real cost, and second, those sources with higher emissions will be set aside in favour of less pollutant alternatives, and consequently greater penetration of renewables will be possible. The proper valuation of environmental profit and loss accounts is a substantial sustainability management framework [32]. The cost of any generating technology incurs costs for capital equipment (i.e. capital expenditure) and cost for the operation of the plant (operational expenditure). Depending on the type of technology, each one of these costs will be different. Also, some costs vary with the amount of energy produced (variable cost), and some costs are fixed. For example, a wind turbine cost is a fixed cost while fuel required for biomass generation is a variable cost. Those alternatives with higher fix costs will usually present larger scales of economy. Since not all expenditures take place simultaneously, it is necessary to take into consideration the financial principle of time value of money, that is, that the discount cash flow analysis should be considered. To do so, we discount all financial flows (inflows minus outflows) to a certain moment of time (usually the present) taking in consideration the adequate discount rate. A key concept in renewable cost analysis is the levelized cost of energy (LCOE), defined as the net present value of life time cost divided by present value of energy production. This is calculated dividing the present value of the total cost of initial establishment and operating a power plant over an assumed lifetime by the present value of expected energy generation during the assumed lifetime, that is: n

P Ii 1 M i 1 F i LCOE 5



i51

n P

i51

ð11rÞi

ð7:2Þ

Ei ð11rÞi

where Ii is the Investment expenditure in the year i and Mi is the Operation & Maintenance expenditures in the year i Fi is the Fuel expenditures in the year i; Ei is the Energy generated in the year i and r is the discount rate; n is the economic life in years of the system. In the above equation, the last two parameters have a great influence on the result of LCOE of renewables. On the one hand, the discount rate (r) has a large impact since given the capital-intensive nature of most renewable power generation technologies and the fact that fuel costs are low, or even zero for renewable sources. On the other hand, the economic life of the plant (n years) is not a trivial estimation issue. A standardized assumption for LCOE calculations are: economic life of 20 years for biomass, 25 years for

250

PART | II Modelling

wind power, solar PV, solar CSP and geothermal and 30 years for hydropower. Regarding the discount rate (r), it is common to assume for OECD countries and China 7.5% and for the rest of the world 10% [33]. Nevertheless, all these values of discount rates should be adapted for each particular project according to the individual risk.

7.5.2.1 Evolution of levelized cost of energy on renewable sources Since 2010, the weighted average cost of LCOE of biomass, geothermal, hydropower, onshore wind and marine have moved closer to the range of conventional fossil fuel power plants. In 2014 the LCOE of solar PV reached the cost of conventional sources. Actually the weighted average of recent auctions price for solar PV has reached $USD 0.048/kWh, a value that is expected to be below the marginal operating cost of almost 900 GW of the operational coal-fired power plants [33]. According to the same study, with global cumulative installed coal-fired generation around 2100 GW by 2020, the prices registered in auction and PPA contracts5 for onshore wind and solar PV are about 40% of the existing coal fleet could be outcompeted by new renewable source deployment. Taking into consideration the current PPA contracts and the auction results, costs should continue to fall in the years to come. The existence of harmonized regulatory frameworks will benefit the implementation of renewable energies and will allow it to reach in a wide number of countries, LCOEs below $USD 0.030/kWh using solar PV technologies, even at those with limited experience in the field, due to the following reasons: (1) the development of international projects in which local actors and highly experienced countries join forces; (2) the low risk and the reduced capital costs; (3) the continuous improvements in technology; (4) the low cost of land in multiple countries and (5) reduced maintenance expenses. Regarding offshore wind and concentrated solar power technologies, costs in 2020 will be set around $USD 0.060 and $USD 0.100/kWh, respectively. However, costs in the near future will not compete with those of the above mentioned technologies. A detailed analysis of LCOE trends during the last decade shows that solar PV technology has drastically reduced its cost. In Fig. 7.13 we present the evolution of solar PV and wind LCOEs between 2010 and 2018. There is no doubt that solar PV will be of paramount importance in the development 5. Power Purchase Agreement (PPA) is a contract between generators (seller of energy) and a second part which is looking to purchase electricity (buyer). Usually these contracts may last anywhere between 5 and 20 years, during which time the buyers agree to buy not only energy, but also capacity and/or other ancillary services.

Large-scale integration of variable renewable resources Chapter | 7

251

0.5

2018 USD$/kWh

0.4

0.3

0.29 0.22

0.2

0.18

0.17 0.13

0.1

2010

0.12

0.08

0.08

0.08

0.08

0.07

0.07

0.1 0.06

0.09 0.06

2011

2012

2013

2014

2015

2016

2017

2018

FIGURE 7.13 LCOE evolution (2010
18) for solar PV (blue) and onshore wind (red). Solid line: global weighted average. Shaded area: interval between 5th and 95th LCOE percentiles. Authors based on IRENA. Renewable power generation cost 2018. Abu Dhabi: International Renewable Energy Agency (IRENA); 2019c [33] data.

of renewable sources. As LCOE of solar PV and wind become closer, the opportunity to develop distributed generation rises. The global energy sector is experiencing substantial changes as renewable power gains strength, becoming the backbone of the power sector transformation.

7.5.3

International interconnections

The development of international interconnections among the different European countries will play a very important role in the high integration of VRES across Europe, and also on the target of reaching 32% of renewables by 2030 in its energy system, which would imply about 57% in the electricity mix [34]. The European international interconnections would also allow to export electricity whenever the national VRES systems are in a situation of surplus, thus avoiding the curtailment of wind or solar energy production as often happened in the past in the case of Germany [35]. Another positive aspect of the development of international interconnections is that they can decisively contribute to the security and continuity of the national electricity services in the case of some failure of the system. In addition they might also contribute to increase the efficiency of the national power systems thus ameliorating the competition amongst neighbouring European countries. Finally we should remark that recently the EU has recommended a country level of interconnection of 15% (ratio of power interconnected with

PART | II Modelling

0

252

FIGURE 7.14 Interchange of electricity power in Europa during 2017. For each country there is an outflow (flow separated form the outer scale) and inflow (flow close to the scale). Based on data from ENTSO-E. Power Flow Tool; 2019. ,https://www.entsoe.eu/data/powerflow-tool/. [accessed 15.05.2019] [37].

neighbouring countries to own capacity power of the country) for the year 2030 [36]. Taking into account that this level in Spain is at present only 5%, this means that it will have to strongly increase its interconnections mainly with France since it has to be crossed to reach many other European countries (Germany, Belgium, Holland, etc.). Currently the degree of interconnection among different countries in Europe presents large differences. Analyzing observed data about flows for the year 2017, in Fig. 7.14 we depict the most relevant6 cumulative energy inflows and outflows within Europe. Countries with large VRES generation

6. In this figure we have removed the flows corresponding to the lowest quartile in order to present the most significant transactions.

Large-scale integration of variable renewable resources Chapter | 7

253

like Germany present the largest outflows. Also France whose electric system mainly relies on nuclear generation presents large outflows. On the opposite side, Italy and Finland were net importers of electricity in 2017.

7.5.4

Digitalization and smart grids

Digitalization of the distribution grid is one of the most significant trends in the EU electricity networks since digital techniques will be essential for the real-time matching of power demand and supply, especially in distributed systems. For instance, smart home systems would allow the optimization of solar photovoltaic systems with storage units like home or electric car batteries. In addition, two-directional smart metres would allow home owners to use electrical appliances when the price of the electricity is lowest. In general, one could say that the use of smart metres and grids would open the way to the joint management of renewable minipower generators (usually solar PV), storage units and distribution lines. We should also remark that digitalization will be also the key technology for the electrification, and consequent decarbonization, of the transport and building conditioning (heating, cooling) sectors [28]. Therefore digitalization will greatly contribute to the high-integration of VRES. Finally we would also like to remark that the implementation of a high share of renewables will open the way to ‘distributed generation’ sources with the corresponding savings in the construction of large transmission lines.

7.5.5

Demand response

Probably the best example in Demand Response techniques consists in the large differences in power demand between the day and night periods which is the reason of the corresponding large difference in the corresponding electricity prices. Based on it, the main purpose of Demand Response (DR) is the transfer of some activities that consume large amounts of electricity (for instance home water heaters or washing machines) from the peak to the valley hours. Of course, the same would apply in the case of working days of the week versus weekend days with much less electricity consumption. Evidently the objective of reducing the demand in peak hours can be realized more effectively by introducing digitalization into the grids [38]. Next let us point out some examples of implementation of DR technologies [39]: (1) Direct load control of utilities (e.g. water heaters) with permission of their customers. (2) ‘Demand limiting’ to maintain the overall energy consumption below a budgetary limit fixed by the customer. (3) Similarly distributed generation would enable consumers to produce power by off-grid generation elements like roof-top solar systems, wind mini-turbines, etc.

254

PART | II Modelling

Acknowledgements We would like to give thanks to the members of the Group Specialized in Energy of the Spanish Royal Physics Society (RSEF) for their fruitful comments during the round table held in the last biennial of the RSEF in Zaragoza (July, 2019). Also to acknowledge the research support from Proyectos I 1 D 1 i «Retos de Investigacio´n» del Programa Estatal de I 1 D 1 i orientada a los retos de la sociedad. Ministry of Science, Innovation and Universities. Ref: RTI2018-100983-B-I00.

References [1] IPCC. Global warming of 1.5oC. Technical report, Intergovernmental Panel on Climate % Change (IPCC), United Nations; 2018. [2] IPCC. Climate change and land. Technical report, Intergovernmental Panel on Climate Change (IPCC), United Nations; 2019. [3] Koomey J, Schmidt Z, Hummel H, Weyant J. Inside the black box: Understanding key drivers of global emission scenarios. Environ Model Softw 2019;111:268
81. [4] Bongaarts J, O’Neill BC. Global warming policy: is population left out in the cold? Science 2018;361(6403):650
2. [5] Ehsani M, Gao Y, Longo S, Ebrahimi K. Modern electric, hybrid electric, and fuel cell vehicles. CRC press. Taylor and Francis Group; 2018. [6] IRENA. Electrification with renewables. Driving the transformation of energy services. Abu Dhabi: International Renewable Energy Agency (IRENA); 2019. [7] U.S. Environmental Protection Agency. Inventory of U.S. greenhouse gas emissions and sinks: 1990
2017. Technical report, U.S. Environmental Protection Agency; 2017. [8] EEA. Renewable energy in Europe: key for climate objectives, but air pollution needs attention. Briefing no. 13/2019. European Energy Agency; 2019. [9] IEA. Data and statistics, ,https://www.iea.org/data-and-statistics.; 2019a. [10] Dudley, B. et al. BP statistical review of world energy. BP Statistical Review, London, UK; 2019, p. 6 [accessed Aug 19]. [11] The Atlas team. Global carbon atlas, ,http://www.globalcarbonatlas.org.; 2020 [accessed 20.1.2020]. [12] Canadell P, Sharifi A. Global carbon project, ,https://www.globalcarbonproject.org.; 2020 [accessed 01.04.20]. [13] IEA. Tracking power, International Energy Agency, Paris, ,https://www.iea.org/reports/ tracking-power-2019.; 2019b. [14] Erias A, Karaka C, Grajetzki C, Carton J, Paulos M, Jantunen P, et al. World energy resources 2016. World Energy Council 2016;;6
46. [15] Sawin JL, Sverrisson F, Rutovitz J, Dwyer S, Teske S, Murdock HE, et al. Renewables 2018-Global status report. A comprehensive annual overview of the state of renewable energy. Advancing the global renewable energy transition-highlights of the REN21 renewables 2018 global status Report in perspective; 2018. [16] IRENA. Renewable energy statistics 2019, ,https://www.irena.org/Statistics/DownloadData.; 2019b.

Large-scale integration of variable renewable resources Chapter | 7

255

[17] Red Ele´ctrica Espan˜ola. Sistema de informacio´n del operador del sistema ele´ctrico en Espan˜a, ,https://esios.ree.es/en.; 2020 [accessed 01.11.20]. [18] Britton E, Fisher P, Whitley J. The inflation report projections: understanding the fan chart. Bank Engl Q Bull 1998;38(1):30. [19] McShane BB, Wyner AJ, et al. A statistical analysis of multiple temperature proxies: Are reconstructions of surface temperatures over the last 1000 years reliable? Ann Appl Stat 2011;5(1):5
44. ˇ c´ıkov´a H, Li N, Gu D, Spoorenberg T, et al. World population [20] Gerland P, Raftery AE, Sevˇ stabilization unlikely this century. Science 2014;346(6206):234
7. [21] Go´mez-Calvet R, Mart´ınez-Duart JM, Serrano-Calle S. Present state and perspec- tives of variable renewable energies in Spain. Eur Phys J Plus 2018;133(3):126. [22] Go´mez-Calvet R, Mart´ınez-Duart JM, Serrano-Calle S. Current state and optimal development of the renewable electricity generation mix in Spain. Renew Energy 2019;135:1108
20. [23] Ministerio para la Transicio´n Ecolo´gica y el Reto Demogr´afico. Plan Nacional Integrado de Energ´ıa y Clima (PNIEC), Spain; 2019. [24] Mongird K, Viswanathan VV, Balducci PJ, Alam MJE, Fotedar V, Koritarov VS, et al. Energy storage technology and cost characterization report. Technical report, Pacific Northwest National Lab (PNNL), Richland, WA (United States); 2019. [25] Crabtree G, Kocs E, Trahey L. The energy-storage frontier: lithium-ion batteries and beyond. MRS Bull 2015;40(12):1067
78. [26] Bloomberg New Energy Finance. New energy outlook 2019, ,https://about.bnef.com/ new-energy-outlook/.; 2019. [27] IRENA. Electricity storage and renewables: costs and markets to 2030. Abu Dhabi: International Renewable Energy Agency (IRENA); 2017. [28] Agora Energiewende and Sandbag. The European energy transition 2050: The big picture; 2019a. [29] Barton J, Gammon R. The production of hydrogen fuel from renewable sources and its role in grid operations. J Power Sources 2010;195(24):8222
35. [30] Sendeco2. Reference market for Southern Europe of CO2 emission allowances, ,https:// www.sendeco2.com/es/.; 2019. [31] Myllyvirta L, Jones D, Buckley T. Global electricity production from coal is on track to fall by around 3% in 2019, the largest drop on record. ,https://www.carbonbrief.org.; 2020 [accessed: 01.12.20]. [32] Danish Environmental Protection Agency, Assessment of potentials and limitations in valuations of externalities. Danish Ministry of the Environment, Environmental Protection Agency; 2014. [33] IRENA. Renewable power generation cost 2018. Abu Dhabi: International Renewable Energy Agency (IRENA); 2019. [34] Agora Energiewende and Sandbag. The European power sector in 2018. Up-to-date analysis on the electricity transition; 2019b. [35] Nicolosi M. Wind power integration and power system flexibility
an empirical analysis of extreme events in Germany under the new negative price regime. Energy Policy 2010;38(11):7257
68 Energy Efficiency Policies and Strategies with regular papers. [36] Directorate-General for Energy (European Commission). Public engagement and acceptance in the planning and implementation of European electricity interconnectors. Third report of the Commission Expert Group on electricity interconnection targets-Study. Publications Office of the European Union; 2019.

256

PART | II Modelling

[37] ENTSO-E Power Flow Tool; 2019. ,https://www.entsoe.eu/data/powerflow-tool/. [accessed 15.05.2019]. [38] Hale ET, Bird LA, Padmanabhan R, Volpi CM. Potential roles for demand re- sponse in high-growth electric systems with increasing shares of renewable generation. Technical report, Golden, CO: National Renewable Energy Lab.(NREL); 2018. [39] Pfeifer A, Dobravec V, Pavlinek L, Krajaˇci´c G, Dui´c N. Integration of renewable energy and demand response technologies in interconnected energy systems. Energy 2018;161: 447
55.

Chapter 8

The climate and economic benefits of developing renewable energy in China Hancheng Dai, MD. Shouquat Hossain and Xiaorui Liu School of Environmental Science and Engineering, Peking University, Bejing, P.R. China

Chapter Outline 8.1 Introduction 257 8.2 Methods and scenarios 259 8.2.1 Integrated model of energy, environment and economy for sustainable development/ computable general equilibrium model 259 8.2.2 Economic assessment of renewable energy 260 8.2.3 Investment in nonfossil power generation 261 8.2.4 Data sources 264 8.2.5 Scenarios 264 8.3 Results 265 8.3.1 Macroeconomic trends towards 2050 265

8.1

8.3.2 Impacts on the energy system 8.3.3 Benefits of developing renewable energy in carbon and air pollutant emissions reduction 8.3.4 Economic impacts of renewable energy development 8.4 Discussion 8.4.1 Policy implications 8.4.2 Comparison with other studies 8.4.3 Sensitivity analysis 8.4.4 Limitations and next step 8.5 Conclusions References

267

271

273 277 277 278 279 279 282 283

Introduction

China is the leading country in the production of clean power from renewable energy (RE) sources, which is twice the second-ranking country, the United States. It has been reported that a total capacity of 728 GW of renewable power will be produced from hydroelectric and wind power at the end of 2018 [1]. However, the RE sector in China is growing faster than fossil fuels and new clean power plants. Energy demand in China is indeed Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00008-X © 2021 Elsevier Inc. All rights reserved.

257

258

PART | II Modelling

increasing rapidly, which can reach an unprecedented level for economic and modern city development. As a result of this outcome, the Chinese government is facing a very critical issue in energy supply security and environmental deterioration, mainly from rising demand for fossil energy. Meanwhile, China has already been recorded as the world’s largest energy consumer and CO2 emitter, whereas the CO2 emissions per unit gross domestic product (GDP) is committed to reduction from 40% to 45% in 2020 and from 60% to 65% in 2030, respectively [2
4]. Therefore low-carbon energy production could play a significant role in reducing emissions. Because of the lack of natural gas and the safety of nuclear energy, replacing fossil fuels with RE could be an option to achieve sustainable development of China’s power sector. In this regard a suitable substructure for the implementation of RE may serve as a long-term sustainable solution [4]. Despite the substantial advancement made by China in the RE sector, particularly solar and wind energies and their quota to the global energy mix production is close to the world average; yet their result in the entire energy mix could almost increase from 8% to 15% in 2020 and 20% in 2030 [5,6]. Currently coal is used in China as the primary source of energy generation, which contributes approximately 70% of the total production [4,7]. Coal will continue to play an important function in economic development in China. However, China has taken a proper charge to create RE in similar steps taken by many developing countries, but there are many drawbacks such as technology and finance that limits the deployment of such an RE [4]. In RE development, climate change serves as a severe issue as it has a connection with both the performance and reliability of the energy system. Previous research on this domain has addressed the weakness of the energy sector from the demand side viewpoint, but various studies have analysed the influence on the supply [8]. On the other hand, other parts of the energy sector may be affected, including the transmission lines. The long lifespan of energy infrastructure has made the energy sector to gain much attention in the literature [9]. The primary sources of RE are directly in connection with climate variables such as solar radiation, temperature, precipitation and wind. Water is a critical variable, which is made available for hydroelectric power plants and needed for other generation plants such as thermal generation or even the carbon capture and storage (CCS) sectors [8]. Based on China’s medium- and long-term plan for RE developed in 2007, there is a set target in the development of solar energy, hydropower, biomass and wind energy as well as other nonfossil energy. However, the aforementioned target was set to serve as climate change and energy-saving policies. Thus the influence of those targets on China’s economy is not known. A computable general equilibrium (CGE) model is proposed for the analysis of the long-term economic impact and to evaluate the policy’s impacts. The model is obtained from the Walrus general equilibrium theory, which stated that there is a balance in demand and supply across all the

The climate and economic benefits Chapter | 8

259

economic markets [10,11]. The model depends on the simulation analysis that joins the intellectual general equilibrium structure formalized by Arrow and Debreu with realistic economic data to solve numerically for the supply level, demand and price that support equilibrium over a specified set of markets. This kind of model is globally used for the analysis of policies impact such as transfer instruments or quotas, taxes and subsidies. They are useful in the areas of development planning, fiscal reform, energy, climate policy and international trade. The model is also useful for the analysis of climate policies and China’s energy. However, few studies have concentrated on the analysis of the impacts of China’s RE policies with hybrid CGE approaches that combine the electrical sector and technology details. In China, the power generation is mostly made from the nonfossil because the electricity absorbs most of the coal. Therefore the electricity sector is the primary source of CO2 emissions. This study carried a detailed analysis of the electricity sector using a hybrid static CGE model for China’s economy. This model has been used for the analysis of the contribution of China’s nonfossil energy plan up to 2050 to carbon intensity reduction and the effect of China’s climate commitment on its economic advancement, CO2 emissions, the dynamics of electricity and energy consumption creation technology [2,7]. As a result of the energy problem, it will take some time for China to optimize the energy consumption arrangement and develop RE to meet the requirement for sustainable development [4]. The rest of this chapter is structured as follows: Section 8.2 describes the methods and scenarios. Section 8.3 shows the results about the trends of macroeconomics, impacts of the energy system, benefits of development RE and the economic effects of RE development. Section 8.4 summarizes the major results and discusses the policy implications, and the last section is the conclusion.

8.2

Methods and scenarios

8.2.1 Integrated model of energy, environment and economy for sustainable development/computable general equilibrium model The integrated model of energy, environment and economy for sustainable development (IMED/CGE model, developed by the Laboratory of Energy & Environmental Economics and Policy at Peking University, is used for this assessment. The base year data, regional coverage and sectorial classification of IMED/CGE model are flexible depending on the specific research purposes and contents. For the purpose of this research, this version selects 2010 as the base year and is solved at a one-year step towards 2050. A representation of 41 Chinese economy sectors is made on this model, which consists of resource-requiring, land-requiring, basic and energy supply sectors. This is a technology-rich hybrid CGE model, which can give a description of

260

PART | II Modelling

series of important technologies of supply energy which include nonfossil fuel supply, power generation and CCS technology. Thus such technology representation treatment will facilitate this model to assess the cost of carbon reduction [12
17]. The model is made of four different blocks, which include a market block, income block of government and household a production block and expenditure block of final demand agents. Sectoral activity is a representation of a nested constant elasticity of substitution (CES) production function, which classifies the input into material, labour, capital energy commodities and resource input. Further information on IMED/CGE model is found in the following site: http://scholar.pku.edu.cn/hanchengdai/imedcge.

8.2.2

Economic assessment of renewable energy

In conventional CGE models the production of electricity is carried out in a single vector without breaking up into various technologies. Hence the initial step is to breakdown the single power sector in China’s traditional input
output table (IOT) into input
output data for eight technologies [coal-, oil- and gas-fired power; nuclear, hydro, wind, solar photovoltaic (PV) and biomass power]. A similar methodology approach found in [18] and the datasets that have been used by Dai et al. [19] to assess China’s nonfossil energy plan towards 2020. However, the three fossil-firing technologies such as gas, coal and oil, as seen in Eq. (8.2) and five nonfossil technologies such as nuclear, hydro, wind, solar PV and biomass, as found in Eq. (8.3), have been used to model the power sector. The aim of each power plant is to maximize its profit πtech [Eq. (8.1)], subject to the CES production technology [Eq. (8.2)]. The energy input of coal, fuel oil or gas is required by the fossil-fired power generation with the low unit cost of the labour and investment. Note that the energy bundle is not merged with capital, but connected directly to the output activity. This implies that there is an existence of a linear relationship between energy input and electricity output. In contrast nonfossil power generation does not require fossil fuel, but the unit costs of capital and labour, which are somehow higher than the fossil-fired power. There is a competition between the fossil and nonfossil power generation technologies as explained in Eqs (8.3)
(8.5). The main goal of the power sector is to increase the total profit π [Eq. (8.3)] by selecting a least-cost technology mix based on the generation cost [Eq. (8.4)]. The electricity generated by each technology can be determined by solving Eqs (8.3) and (8.4) that is Eq. (8.6). The powergeneration technology relies on many factors. The generation cost (ptech ) for a technology decreases as the share (δtech ) in the base year increases, and more power is obtained as the substitution σ for the technology increases. The most crucial factor is the future generation cost ptech , which is dependent on the relative prices of nonfossil energy, capital and labour inputs.

The climate and economic benefits Chapter | 8

maxπtech 5 ptech UQtech 2

X

pi UINPUTi;tech

261

ð8:1Þ

i

such that,

   Qfos 5 LEO1 Mfos ; CES2va CES3va ðKfos ; Lfos Þ ;   CES2e ðelefos ; CES3fos coalfos ; gasfos ; oilfos Þ   Qnfos 5 LEO1 Mnfos ; CES2va ðKnfos ; Lfos Þ X maxπ 5 pele UELE 2 ptech UQtech

ð8:2Þ ð8:3Þ ð8:4Þ

tech

such that, ELE 5 αU

X

δtech UQ2ρ tech

!2ρ1 ð8:5Þ

tech

By solving Eqs (8.3) and (8.4) we obtained the following numerical solution:

pele σ UELE ð8:6Þ Qtech 5 ασ U δtech U ptech where tech is the power-generation technology listed in Table 8.1, including fossil-fired technology, fos, and nonfossil technology, nfos; πtech is profit of technology, tech; π is P profit of whole power P sector; pele is an average electricity price, equal to tech ðQtech Uptech Þ= tech Qtech Þ; ptech is generation cost of technology tech; ELE is total power generation; Qtech is electricity generated by technology tech; INPUTi,tech is input required for power generation, for example labour, capital, energy and other intermediate inputs; α is efficiency P parameter in CES function; δi is share CES function; 0 # δi # 1; i δi 5 1; and ρ is substitution parameter in CES function, in which the elasticity of substitution among technologies, σ, equals 1 11 ρ.

8.2.3

Investment in nonfossil power generation

Another innovative part of this research is the treatment of investment in nonfossil power generation. As depicted in Fig. 8.1, the two types of investment in the year (t) are differentiated. The first type is the traditional investment in sectors outside the nonfossil power generation that is sectors where construction and machinery commodity are the major investment goods. On the other hand the second type is the investment intended for the production of nonfossil power plants. Total investment in specific nonfossil power generation technology in year t is estimated by multiplying the unit cost (COSTnfos;t in Yuan/GW) with the newly installed capacity NCAPnfos;t

262

PART | II Modelling

TABLE 8.1 Investment cost for power generation. Annual reduction rate (%)

Cost (Thousand Yuan/KW) This study

Coal

1

4.38

Oil

1

NA

Gas

1

2.85

Hydro

1

6

Nuclear

1

10

2010 range

2050 estimation

Wind

0.63

9

Biomass

3

20

15
30

7 20

Solar

1.07

20

15
30

20

Source: Coal- and gas-fired power from China Electricity Council. Annual development report of China’s power sector 2013. Beijing [in Chinese]; nonfossil power from CNRE Centre. China renewable energy technology catalogue; 2014. Beijing; Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

FIGURE 8.1 Illustrative treatment of investment in a production function for the power sector. Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

The climate and economic benefits Chapter | 8

263

in GW, as shown in Eq. (8.7). In addition, the investment in nonfossil power generation differs from traditional investment because it involves specific goods such as wind turbines and silicon plates. The demand share matrix for nonfossil power-generation technologies has been prepared in Table 8.2 established on cost data investment from 2011 China Electricity Statistics [20] and data acquired on typical hydropower, wind power, solar PV power and biomass power generation projects in 2010 [21]. The nonfossil power generation investment goods are comprised of a higher share of machinery, transport service, research and development, electronic products and a lower percentage of construction. The variation denotes that the related industries promoted by investment in the creation of nonfossil energy differ from those promoted by traditional investment. This distinction facilitates the assessment of economic impacts of various long-term investment strategies in China. Whether the investment in conventional and backward industries, then the spurred industries will include construction,

TABLE 8.2 Investment demand for non-fossil power generation and other sectors. Other (%)

Hydro (%)

Nuclear (%)

Wind (%)

Biomass (%)

Solar (%)

5

3

3

5

10

Agriculture

2

Metal products

2

Transport equipment

7

2

Machinery

24

28

55

38

40

40

Electronic products

5

7

15

5

8

12

Other manufacturing

1

Construction

49

35

25

15

25

13

Transport

2

8

12

5

8

Research & Development

0

12

8

4

10

Service

8

8

12

5

14

Total

100

100

100

100

100

5

100

Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

264

PART | II Modelling

iron and steel, cement, industries that carry China along an energyintensive course. Otherwise China invests in RE development, the entire upstream industry chain will be motivated, and China will encounter natural economic growth. Infos;year 5 ðCAPnfos;year1ts 2 CAPnfos;year 3 d ts Þ 3 COSTnfos;2010 3 REDyear22010 ð8:7Þ where Infos;year is the investment in nonfossil power generation in a given year (billion Yuan); CAPnfos;year is expected installed capacity of nonfossil power in a given year (GW); CAPnfos;year1ts is expected installed capacity of nonfossil power in the period following a given year (GW); d is depreciation rate (5% per year); ts is time step of the model (5 years); COSTnfos;2010 is unit investment cost of the nonfossil power plant in 2010 (billion Yuan/GW; Table 8.1); and RED is average reduction rate of the unit investment cost of a power plant (%, Table 8.1).

8.2.4

Data sources

The data used in the model include the IOT [24] and energy balance table [25]. For constructing the environmental accounts, carbon emission factors, energy prices of coal, oil and gas, and RE technology costs are used and all datasets are converted to the base year of 2010.

8.2.5

Scenarios

Reference scenario and the REmax scenario envisions has been created. Whilst the reference scenario presumes that the moderate pace is the bedrock for the development of RE, then the REmax scenario predicts the high perception of RE till the year 2050. However, in the year 2020, another two additional carbon constraint scenarios, CM40 and CM45, are created to envisage the effect of carbon reduction on the economy, which is in line with the Copenhagen Commitment that requires China’s carbon intensity in terms of GDP would drop by 40%
45% compared with 2005 level.

8.2.5.1 Reference scenario With regard to the economy, China plans to realize its long-term strategy of modernization in the year 2050. The population rises to 1.45 billion in 2030 and drops to 1.35 billion in 2050. GDP increases by five times over the periods from 2010 to 2050. Consequently the per capita GDP attains the level of a moderately developed country. The economy, meanwhile, did not only increase in size but also improves in structure: the dominance of heavy industry in China’s economic structure diminishes and the country shifts to a service-driven economy [7].

The climate and economic benefits Chapter | 8

265

Naturally China’s air quality in 2050 is not required to decline beneath the stage documented when China started the initial policy at the beginning of the 1980s. Precisely annual SO2 and NOx emissions are predicted not to go beyond 11.5 and 10 million tons, respectively. CO2 emissions peak in around 2030, in correspondence with the current commitment by China (NDRC, [26]), followed by a decrease to around 11 billion tons in 2050.

8.2.5.2 REmax scenario According to the results of the power sector model [27] in the China Renewable Energy Centre, a further attempt is made to develop RE in the REmax scenario (Table 8.3). As the installed capacity of wind (over 2000 GW), solar PV (2500 GW), biomass (over 600 GW), hydro (500 GW) and nuclear (100 GW) increases, coal and natural gas power changes from baseload to peak-load power sources with installed capacities of 500 and 400 GW, respectively. For example hydro and pumped hydropower technologies are aggregated into hydro; biogas, straw and wood power are aggregated into biomass. Furthermore some power generation technologies in the power sector model are not taken into account in the CGE model because of their low-scale in nature, for example geothermal, wave and chemical storage [7].

8.3 8.3.1

Results Macroeconomic trends towards 2050

Macroeconomic analysis is critical to understand the effect of economic development. We need to plan for a long-term energy roadmap, which evaluates the technological costs and economic costs for different pathways. Some studies mentioned the technical possibility of reaching a high share of RE in the energy system in Ireland, Portugal and Denmark [28
30]. On the other hand, they did not examine the full macroeconomic effects of such high penetration of RE. By applying these economic models, the relationship between RE development and the economic impact could be uncovered. Table 8.3 shows the reference scenario outcomes for the economy, energy and environment. These outcomes have been resolved by China’s top national expert investigators and organizations on future trends of the Chinese economy. As for the economy China targets on accomplishing its long-term policy of modernization by the year 2050. Through those periods from 2010 to 2050 in the reference scenario, China’s GDP would grow by around five times. Furthermore, the per capita GDP would rise to 27,000 USD. However, future growth is progressively determined by advancement and final domestic consumption. Concerning economic reform, the shares of capital formation, exports and trade surplus in GDP would gradually decrease from 43%, 40% and 4.7% in 2010 to 26%, 32% and 0.4% in 2050, respectively. On the other hand, the share of household consumption in GDP

TABLE 8.3 Installed capacity of non-fossil power in the reference and REmax scenarios (GW). Coal Reference

REmax

Oil

Gas

Hydro

Nuclear

Wind

Biomass

Solar

2010

700.0

1.14

30.5

260.3

11.9

48.4

5.8

3.8

2015

822.5

1.17

100.3

305.6

39.9

105.3

31.0

42.0

2020

901.9

1.14

198.9

360.2

64.3

249.4

58.0

104.5

2030

1121.8

1.32

274.6

497.3

112.9

592.5

68.6

309.9

2040

1083.4

1.07

297.3

567.7

201.3

847.4

101.7

777.2

2050

954.3

1.07

393.2

692.1

300.0

934.3

125.0

1027.0

2010

700.0

1.14

30.5

260.3

12.6

48.4

5.8

3.8

2015

822.5

1.17

94.3

305.6

42.8

114.9

29.1

42.0

2020

1083.4

1.4

180.4

364.0

50.5

317.1

60.9

157.0

2030

1052.2

1.01

211.6

522.1

66.0

1103.9

87.6

1048.9

2040

892.3

0.82

325.2

579.4

78.0

2092.2

150.4

2206.3

2050

686.7

0.82

432.8

696.4

100.0

2396.6

214.1

2696.2

Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

The climate and economic benefits Chapter | 8

267

would rise from 39% in 2010 to 68% in 2050, implying that domestic demand would replace export and investment as the primary driving force of China’s economic growth by the middle of this century. Meanwhile the share of the primary industry will drop to 2.7% by 2050, and the secondary industry would drop from 46.7% in 2010 to 35.0% by 2050. On the other hand the tertiary industry would increase from 43.2% to 62.4%, which is close to the structure of developed countries. Other outcomes such as CO2 emissions will increase rapidly from 2010 to 2030 then gradually decrease, but SO2 and NOx emissions will dramatically reduce by 2050. It can be seen that the energy demand will increase and, at the same time, increase the pressure on natural gas supply, but coal and oil consumption would gradually decrease because of environmental policy. Table 8.4 shows some previous results for different scenarios. The carbon intensity would decline by about 30.97% and 38.94% in 2020 from the 2005 level under the reference and REmax scenarios, which means China’s nonfossil energy development target can be achieved with a reduction rate of 7.35%. However, this is still not sufficient to meet the target in the Copenhagen Commitment of reduction by 40%
45%. Only in the CM40 and CM45 scenarios, where the carbon intensity drops to 1.69 and 1.55 tons of CO2 per 10,000 Yuan in 2020, corresponding to 40% and 45% reductions, respectively, could achieve the Copenhagen target. In the reference and REmax scenarios, no carbon price is in place since there is no imposed carbon imperatives in the economic system. However, the imperatives on carbon emission in the CM40 and CM45 will induce the carbon price to be 10.41 and 66.17 Yuan for every ton of CO2. Furthermore more RE the REmax scenario would help to reduce the carbon prices.

8.3.2

Impacts on the energy system

China requires to amend its energy system in a far better approach in anticipation of achieving its ecological focuses and impressive commitments. Currently the Intergovernmental Panel on Climate Change releases an important statement (IPCC, [31]), which gave highlights on the convincing reason to reduce greenhouse gas (GHG) to net-zero by 2050 to control the universal rise in temperature within 1.5 C [32]. The statement suggested that the GHG emission will decline to about 45% in 2030 and 100% in 2050; this indicates that transitioning about 70%
85% of global fossil
fuel power sources to RE power sources, executing a cost on carbon and increasing that diffusion of CCS. Indeed, renewable energy targets (RETs), such as solar PV and wind energy are required to function in transitioning of the energy systems to avoid climate change [33].

268

PART | II Modelling

TABLE 8.4 Boundary of economic, energy, and environmental indicators for the reference scenario (Modified). 2010

2020

2030

2040

2050

1.34

1.40

1.43

1.40

1.35

Socio-economic Population (billion) GDP (trillion 2010 USD)

5.92

12.26

20.83

29.40

35.90

Per capita GDP (2010 USD)

4432

8716

14624

20976

26590

Primary industry share (%)

10.1

6

4

3.1

2.7

Secondary industry share (%)

46.8

43

39

34.9

32.3

Tertiary industry share (%)

43.1

51

57

62

65

Urbanization (%)

50.0

61

69

74

77

Share of household consumption (%)

39.6

47.3

53.5

60.4

59.9

CO2 emissions (bil. Ton)

5.21

12.68

11.17

10.98

10.04

Carbon intensity (ton CO2/10k yuan)

2.82

1.94

1.72

1.69

1.55

Carbon intensity reduction (%)

-

30.97

38.94

40.00

45.00

Share of government expenditure (%)

14.1

13.1

12.6

12.8

13.8

Share of capital formation (%)

43.4

38.2

33.1

26.3

25.9

Share of export (%)

39.6

37.1

34.8

32.6

31.6

Share of import (%)


36.8


35.8


34.1


32.1


31.2

SO2 emissions (million ton)

22.73

18.50

15.00

12.00

10.00

NOx emissions (million ton)

8.29

7.00

6.00

5.00

4.00

CO2 emissions (billion ton)

0.82

1.15

1.30

1.25

1.10

Energy demand (billion tce)

3.25

4.8

5.8

6.3

6.3

Coal (%)

68.0

58.0

49.3

42.0

34.0

Oil (%)

19.0

17.3

16.5

16.6

16.1

Natural gas (%)

4.4

9.7

12.6

13.7

14.8

Non-fossil (%)

8.6

15.0

21.6

27.8

35.1

Eco-environment

Energy

GDP, Gross domestic product.

The climate and economic benefits Chapter | 8

269

8.3.2.1 Primary energy Fig. 8.2 shows the primary energy consumption in reference and REmax scenarios. The reference scenario depicts that the coal used energy consumed about 2201.11 mtce (million tons of coal equivalent) in 2010 and the peak energy will consume about 2711.36 mtce in 2020 and after that, the consumption goes down a level than 2010. The second largest source of oil used energy consumed about 607.42 mtce in 2010, and the peak oil consumption would be about 1002.54 mtce in 2030. In 2030 the consumption would remain constant, but the level would drop between the years 2040 and 2050. The same behaviours can be observed in the gas, nuclear, and hydro sources of energy consumption. Whereas the gas, nuclear and hydro used energy consumption is estimated to be about 73.33, 73.25 and 315.77 mtce in 2020, respectively. However, biomass and solar used energy consumption performance are slightly different from other energy sources. They started their energy consumption rate in 2015. After that, biomass energy consumption reaches a peak of about 146.11 mtce in 2050, but the solar energy consumption peak is nearly 406.34 mtce in 2050. On the other hand wind energy consumption was found in 2010 and the peak energy consumption of about 531.18 mtce can be observed in 2050. It can be evaluated that RE consumption will be higher than the fossil energy sources. Whereas the REmax scenario shows quite a different trend. It is found that the total primary energy consumption is followed by a reference scenario. This energy blends progressions strikingly with the substitution of fossil energy towards RE in the power sector. Fig. 8.2 shows that coal-fired power will increase in 2020 and then decrease to the levels of 2010. It is observed that oil, gas, nuclear, hydro, biomass, wind and solar energy consumption will be about 925.18, 273.30, 2657.53, 78.33, 72.99, 139.86 and 69.42 mtce in 2020, respectively. However, the REmax scenario shows that all primary energy consumption will decrease in 2050. On the other hand the

FIGURE 8.2 Primary energy consumption by 2050 (Modified). Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

270

PART | II Modelling

use of RE is gradually dropping, which is higher than fossil sources such as coal, oil and gas. Because of future power generation will depend on RE sources. It can be observed that nuclear, hydro, biomass, wind and solar energy consumption will be about 122.96, 488.12, 274.90, 1337.5 and 1077.4 mtce in 2050, respectively.

8.3.2.2 Power structure Fig. 8.3 shows the power structure in reference and REmax scenarios. Currently China depends on coal as a dominant primary energy source, which can also be seen in the reference scenario. In 2010 coal-fired power production was 3.25 PWh, and that production would double its capacity in 2050. But in REmax scenario, the coal power production will be about 3.25 PWh and that production is lower in 2050, whereas the oil power production will be phased out in both scenarios. Gas-fired power generation remains the same for both scenarios. The peak power generation for both scenarios will be about 0.95 PWh in 2050. Another power production such as nuclear power has a less contribution in reference scenario and the power generation had a little increase of about 0.13 PWh in REmax scenarios. The hydropower generation contradiction can be observed in both scenarios. Where the peak power production in both scenarios is 2.15 and 2.82 PWh in 2050, respectively. The development of RE power production impotent and performance, especially the solar and wind energy contribution can be seen in REmax scenario. The maximum shares of renewable and nonfossil energy in power production could attain 74%
78% in 2050, which will be at the range of 56%
60% of total primary energy, respectively. It is projected that the primary energy mix by 2050 will depend on wind and solar energy. The wind and solar-based power generation would increase by 2.24 and 2.92 PWh in the

FIGURE 8.3 Power generation mix by 2050 (PWh) (Modified). Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

The climate and economic benefits Chapter | 8

271

2050 REmax scenario compared with the reference scenario, respectively. Perhaps hydro and biomass-based power production would increase to 2.82 and 0.94 PWh in the 2050 REmax scenario, respectively. The power consumption would rise to 15.7 PWh in 2050, which is 3.8 times more than in 2010 and around 11,700 kWh power consumption per capita, which is close to the level of per capita consumption in developed countries [7].

8.3.3 Benefits of developing renewable energy in carbon and air pollutant emissions reduction The CO2 emissions of the power sector emerge mostly from coal- and gasfired power. Considerable benefits in terms of air pollutants and CO2 emission reduction will be achieved through the development of RE in the power sector by a drastic decline in the consumption of coal. Fig. 8.4 shows the reference and REmax scenarios and the effect of CO2 emission and air pollution. It can be observed that the CO2 emission and air pollutions for both scenarios are in the same trend between 2010 and 2030. Then the trend will fluctuate based on the consumption of fossil sources. The CO2 trend in the reference scenario shows that the emission will increase to a peak of about

FIGURE 8.4 Impacts on air pollutant and CO2, N2O, NOx and SO2 emissions (Modified). Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

272

PART | II Modelling

9200 million tons in 2030. Whereas in the REmax scenario, the emissions will decline to 7319 million tons in 2030. Furthermore the emission reduction rate could be 29.78% in 2050 (from 8665 to 5686 million tons). The N2O trend represents some common trend for the air pollution effect in both scenarios, in which emissions are almost the same until 2020 between reference and REmax scenarios. From 2020 onwards, the patterns of all air pollutants will divide and decrease in similar ways. However, the N2O air pollution rate is significant. Based on the future projection, the N2O air pollution reduction rate could be 23.5% in 2050. Another air pollution is NOx, where the reference and REmax scenarios would follow the same pattern from 2010 to 2020. After that, the trend will divide and decrease in the same ways. Based on the future simulation, the NOx air pollution reduction rate could be 30.7% in 2050. Furthermore the SO2 air pollution reduction rate could be 42.5% in the REmax scenario compared with the reference scenario. The sectoral level emissions and reduction contribution are shown in Fig. 8.5. In general, the majority reduction originates from the manufacturing industry because of the different energy supply. It can be observed that for N2O, whereas from the beginning there is no change between reference and REmax values, between 2030 and 2050 the total air pollution reduction is 11.9
16.5 kilotons in the most remarkable sectors such as household, service, metal, electricity, other energy and other manufacturing. That

FIGURE 8.5 Sectoral emissions of air pollutant (Modified). Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

The climate and economic benefits Chapter | 8

273

comparison for NOx gas between 2030 and 2050 is 3.0
4.4 million tons in household, service, metal, electricity and other manufacture sectors. Other sectors contribute less to this pollution reduction. For SO2 gas, the reduction would be 3.8
4.7 million tons from 2030 to 2050, and the main contribution would be from household, service, metal, electricity sectors, other energy, other manufacturing and transport road sectors. For CO2 emissions the reduction would be 1.8
3.2 billion tons, and the most remarkable contributing sectors are electricity, other manufacturing, metal smelting and so on, whereas the service sector makes less contribution. For NMVOC gas, air pollution reduction values are from 0.1 to 0.2 million tons. Whereas with CH4, the reduction is 0.1 million tons for only 2030. In 2050, the value is unchanged between the two scenarios. For CO and NH3 gases, there is not much reduction, but the household sector is the visible reason for air pollution. Unit: N2O, 10 kilotons; CO2,billion tons; CO, 100 kilotons; NH3, 10 kilotons; Other gases (NOX, SO2, NMVOC, CH4), million tons.

8.3.4

Economic impacts of renewable energy development

It has been investigated that RE brings about significant impacts on the economic system in China. Abstractly there needs to be aid at three levels when it comes to the regulation and backhanded economic impacts. (1) significant expansion has been found in the nonfossil power sector; (2) immense investment in RE growth that might make colossal requests for investing goods and profit on the related upstream streamlined chain concerning design and (3) No less importantly, different conventional sectors could lose, particularly the individuals involved with routine energy supply. The net impact on the general economy is unknown without bringing into account of aggregated positive and negative effects comprehensively. Note that the CGE model offers a distinctive capacity with the consideration on both the positive and negative impacts and will evaluate the general effects on the different viewpoints of the macroeconomic such as the output value of every sector, GDP, household income, job around sectors, commodity price, and import and export.

8.3.4.1 Investment To achieve the RET, it is vital to know the total investment demand for renewable and non-RE sources. In this regard the investment cost has been calculated by using Eqs. (8.2)
(8.7) and the results have been categorized by reference and REmax scenarios as shown in Fig. 8.6. It can be seen that the total investments in developing RE will be increasing at a reasonable amount of 303.58
224.86 billion USD from 2040 to 2050 in the reference scenario. Meanwhile the share of investment would keep increasing in the

274

PART | II Modelling

FIGURE 8.6 Investment in nonfossil power generation and the share of renewable energy investment out of total investment (Modified). Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

total investment of about 3.6% in 2040 for RE development. However, most of the investment demand goes for the development of solar and wind energy, and some portion goes to the hydro and nuclear power sector. On the other hand, the REmax scenario shows that the maximum investment area is RE production, especially solar and wind energy. It can be seen that the peak investment would be 787.63 billion USD in 2040, which accounts for over 9.5% of the total investment.

8.3.4.2 Impacts on industrial output, value-added and employment Energy-intensive industries currently dominate China’s economy. In the future the share from the primary industry will fall continuously to 2.7% by 2050 and the secondary industry will fall starting from 46.7% in 2010 to 35.0% in 2050. On the other hand, the tertiary industry ascents from 43.2% to 62.4% over the same period, which is close to the current industry structure in the developed countries. Those expansions introduced limits for nonfossil power prompts an intense increase in the monetary output of the nonfossil power sectors. Likewise, Fig. 8.7 shows that the monetary output and value included in the nonfossil power sector in 2010 were truly small. In the reference scenario, the output of nonfossil power sectors would expand moderately and the total value included in nonfossil power sectors involves a minor share (1.9%) of China’s GDP in 2050. On the other hand in the REmax situation, wind and solar energy draw in substantially additional investment, whereas the total output of nonfossil power sectors will expand to about 1.98 trillion USD in 2050. It can be seen that the RE reveal output investment was 224.6, 63.5, 165.1, 648.5 and 881.6 billion USD, with new value-added to about 132.7,

The climate and economic benefits Chapter | 8

275

FIGURE 8.7 Output value of nonfossil power sectors in 2050 (2010 prices) (Modified). Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

37.5, 98.1, 386.9 and 525.2 billion USD, respectively. Altogether, the value of nonfossil power sectors will achieve 1.18 trillion USD in 2050, representing 3.4% of the GDP. As a result the nonfossil power sector would turn into a backbone industry by 2050. Fig. 8.8 shows the results from the reference scenario, where solar and wind power-based energy would obtain a higher motivation compared to hydro and nuclear power in 2050. However, the total indirect output value of 1332 and 1179 billion USD from other sectors will be motivated by renewable and nonfossil energy, which corresponds with the REmax scenario. As a result, those energy production values of 342 and 302 billion USD would have been added, and the GDP would increase by about 0.99% and 0.88%, which is an increase of 4.87 and 4.12 million jobs, respectively. Moreover the transport, machinery, electronic and service sectors get the most benefit from RE development. The future net effects of RE development on the overall economy in 2050 are shown in Fig. 8.9, which takes into account both the positive

276

PART | II Modelling

FIGURE 8.8 Stimulating effect of renewable energy development on employment, outputs and value-added of other sectors in 2050 (Modified). Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

FIGURE 8.9 Impacts on employment, output and commodity prices in 2050. Source: Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49 with permission from Elsevier.

The climate and economic benefits Chapter | 8

277

impacts on the stimulated upstream sectors related to RE supply chain and the adverse effects on the sectors linked to the conventional fossil energy supply. The results indicate that the most affected industry is coal, whose output falls by 52%, employment falls by 54% (over one million), price falls by 24% (because of demand decrease) and other sectors such as coke and urban gas prices dropping by around 1%. The negative impacts are also rising in the cement, construction, steel and iron, machinery and mineral mining sectors since the RE power plant got higher investment rather than fossil-fired power or other plants. On the contrary, the RE investment benefit can be befitted in electronic machinery, research and development sectors. In addition, the electricity, coal, natural gas and petrol oil decline their prices, which could push down the energy cost of other nonenergy sectors, which is leading to lower market prices with higher demand for nonenergy supply.

8.4 8.4.1

Discussion Policy implications

This study proposed a future roadmap in terms of three considered stages in reaching the target of RE development. This roadmap has a form-based on analytical results, which has been discussed with China’s energy experts to enlighten their opinion. The development of the RE technologies market has matured from 2010 to 2020. At the same time, the cost of wind power will be able to compete with the traditional fossil-fired power, and that power will be enough for large-scale commercial use. However, system improvements should be made to facilitate the penetration of RE. With these improvements and institutional arrangement, RE progressively necessary and eventually supplant fossil energy as the primary source in the resource-rich areas. Those on medium-haul, starting from 2020 to 2030, will have a chance of a decade of fast infiltration for RE. The cosset from claiming solar PV will fall altogether from the current level. Likewise an aftereffect about progressed expense competitiveness, the share for RE in the energy mix will increase more quickly than it did the previous decade. RE will get a standout amongst the primary energy sources, representing more than 25% of total energy in 2030. China’s fossil energy can be turned into a modern energy system, but compared with more extended periods between 2030 and 2050, which will be dominated by RE. However, the solar PV future cost will decrease to about 0.80 USD/W and wind power plant installed capacity will increase to 2300 GW. Considering the resource distribution, inland wind farms will be principally placed in the Northeast, North and Northwest China. On the other hand, the installed wind power capacity in Inner Mongolia, Hebei, Xinjiang, Liaoning and Shandong provinces could be about 800, 200, 200, 100 and 100 GW, respectively. As a result, in 2050, the solar PV’s installed capacity

278

PART | II Modelling

will increase to nearly 2700 GW. Those incorporated energy plants will be found in resource-rich regions, for example Inner Mongolia, Qinghai and Xinjiang. Moreover, China needs to be more focused on solar PV power plant installation and distribution, which will provide sufficient development across the nation. It is estimated that in 2050 the total RE capacity will be around 5000 GW. In this condition, the RE in power and primary energy would increase by 70% and 50%, respectively.

8.4.2

Comparison with other studies

The primary energy share from RE mainly depends on the total energy utilization, which relies on industrial restructuring. Table 8.5 shows a comparison between energy utilization in the reference scenario and projection by the Chinese Academy of Engineering. According to the chinese academy of engineering (CAE) projection, primary energy and electricity demand will be higher than ours in 2040. However, the current investigation shows that while primary energy production will be high in 2040, then the

TABLE 8.5 Primary energy and electricity demand in the reference scenario (Modified). Year

Population (bil.)

GDP growth rate (%)

Primary energy (bil. tce)

Per capita (tce/ cap)

Electricity demand (TWh)

Per capita (kWh/ cap)

3.25

2.43

4200

3135

CAE study (CAE, 2011, 2011) 2010

1.34

2020

1.42

7

4.58

3.23

7242

5100

2030

1.45

6

5.81

4.01

10,315

7114

2040

1.45

4

6.05

4.17

11,622

8015

2050

1.40

3

5.53

3.95

10,943

7817

Reference scenario of this study 2010

1.34

11.2

3.12

2.32

4177

3117

2020

1.40

7.4

4.53

3.23

7623

5445

2030

1.43

5.3

5.10

3.57

10,813

7562

2040

1.40

3.4

5.16

3.68

13,754

9824

2050

1.30

2.0

4.92

3.78

15,743

12,110

GDP, Gross domestic product.

The climate and economic benefits Chapter | 8

279

utilization will be lower than the CAE report. The electricity demand is higher than the CAE report, and the results show that the per capita electricity utilization is very close to the modern developing countries. On the other hand, China’s economic progress will increase as an increase in primary energy production and electricity demand. As long as we save coal, natural gas and crude oil demand, it is essential to restructure the industrial sector. As a result China will increase its energy production, which will give a surprise to the international energy market and China will face internal energy security challenges. China has big hopes regarding the massive development of RE and this will provide a high share RE market in the future 2050.

8.4.3

Sensitivity analysis

It is important to understand the novel CGE model analysis principle. To do so this model will undergo a sensitivity analysis and the model may work depending on parameter assumptions such as using sensitivity analysis to get the international price of commodities, substitution elasticities of fossil fuels and exchange between capital and energy, and others. The result shows that even a 10% rise or drop from the present value has an impact on GDP in the REmax scenario. However, this impact also has an effect on carbon prices in CM40 and CM45 and primary energy consumption in all scenarios shown in Tables 8.6 and 8.7. International price have a remarkable influence on GDP and GDP change. The higher international price get more commodities will be produced and exported. Therefore as international price increase, the GDP in the reference scenario increases as well. Be that as they have little impact with respect to carbon price and primary energy consumption in CM40 and CM45. The elasticity of substitution amongst fossil fuels determines the degree of the fuel switch. Higher elasticity of substitution permits higher adaptability to energy consumers towards using a greater amount of low-carbon fossil fuels such as natural gas under carbon constraint scenarios. GDP and primary energy utilization are not influenced considerably by the elasticity of substitution amongst fossil fuels. The elasticity of substitution between capital and energy has a noteworthy impact on the results, particularly concerning carbon value. A drop of 10% from its base value leads to carbon cost in CM40 scenario to be 0, which demonstrates that a 40% carbon intensity decrease could be attained in the reference scenario.

8.4.4

Limitations and next step

In this study novel methods have been established. Capital information has been investigated for all nonfossil power generation technology and

280

PART | II Modelling

TABLE 8.6 Sensitivity analysis: GDP in reference scenario and GDP loss in other scenarios (Modified). Parameter

International price

Elasticity of substitution among fossil fuels

Elasticity of substitution between capital and energy

Percentage change (%)

GDP (10 bil. Yuan)

GDP change compared to REmax (%)

REmax

CM40

CM45


10

6355.68

2.21

1.99

0

6498.21


0.03


0.24

1 10

6549.55


0.82


1.03


10

6496.19


0.03


0.26

0

6498.21


0.03


0.24

1 10

6500.10


0.03


0.23


10

6481.63

0.00


0.10

0

6498.21


0.03


0.24

1 10

6513.18


0.15


0.39

GDP, Gross domestic product.

economic effects on RE development. This assessment is rather important, provided that the capital utilized for RE development varies evidently from the conventional capital for other sectors. This investigation is innovative because of our disaggregation of the traditional IOT in two ways: (1) dividing the single power sector into several generation technologies, and (2) disaggregating the capital foundation section, which will separate the capital information on RE and different sectors and thereby unequivocally representing those relationship between RE improvement and those upstream industry chain. The study has gone through several limitations, which will be discussed in this section. First this study only evaluates how renewable power generation would substitute the primary coal power, whereas petrol oil utilization may be just insignificantly influenced. However, several studies have identified that RE can be a potential for the transport sector possibly in the type for low-carbon power or biofuel [34
38]. Second the national CGE model in this study assumes small countries’ assumptions, in which the global market is assumed to be not affected by China’s domestic situation. Therefore the model might have overrated the effects of domestic stimulation. In reality a significant number of the machinery of RE is made supplied from outside nations. In this idea China can make more investment in RE, which will be

TABLE 8.7 Sensitivity analysis: carbon price and primary energy consumption (Modified). Parameter

International price

Elasticity of substitution among fossil fuels

Elasticity of substitution between capital and energy

Relative change (%)

Carbon price (yuan/ton CO2)

Primary fossil energy consumption (mtce)

CM40

CM45

Reference

REmax

CM40

CM45


10

10.40

66.18

4674.92

4093.75

4299.34

4014.09

0%

10.40

66.17

4970.80

4358.38

4299.36

4014.10

1 10

10.40

66.18

5204.64

4587.68

4299.34

4014.09


10

10.74

68.80

4990.67

4371.71

4311.82

4021.15

0

10.40

66.18

4970.80

4358.38

4299.34

4014.09

1 10

10.66

64.57

4951.71

4345.53

4284.02

4003.28


10

0.00

33.90

4623.78

4124.37

4124.37

3990.64

282

PART | II Modelling

an immense benefit of the business prospects for different countries. On the other hand the model did not categorize the export of RE-related materials. A tremendous investment in RE might fast-track the taking in of the procedure and decrease in the cost of investment. These advancements might expand the intensity of China’s RE-related industry in the worldwide market. This trend has already begun, as represented by the fierce competition from Chinese RE producers in the European market in recent years. Additionally this model is not capable of detecting regional economic impact within China. Concerning the third limitation, the study shows that a significant number of contrasts that exist between the traditional power and power generation by variable renewable energy (VRE) sources, such as solar and wind energy. The limitation of VRE power generation strongly depends on the weather and the time of day [39]. It also correlates weakly with hourly load profiles [40]. At those shares of the wind- and sun-based climbs with large amounts in 2050, those joining will end up being additionally challenging and expensive [41]. Supplementary challenges in system coordination might significantly limit the organization for REs [42]. The features for VRE are not unequivocally reflected in this model and are required to be accounted for in future worth of effort.

8.5

Conclusions

The study uses a unique equilibrium model to evaluate the economic and environmental benefits in RE development in China. The key findings are enumerated as follows: 1. RE development could have a positive effect on industry restructuring in a low-carbon economy. It will help China achieve productive energy utilization and the establishment of a high capacity power generation from wind and solar PV. It has been estimated that the RE power capacity could be more than 2000 GW. This change could affect the power grid systems, which could be facilitated by the high penetration of RE. It concludes an annual peak investment of 800 billion USD for RE development, especially wind and solar PV power. The peak investment would account for 9% of all Chinese society and close to that of high-speed railway development. 2. China has reformed its economy and energy structure. In the future, the maximum share of RE in power generation could reach about 74%, which corresponds to 56% of the total primary energy. 3. The development of RE also affects the macroeconomic system and other sectors. Massive-scale RE development would not have substantial adverse effects on the macroeconomy, which has only adverse effects of 0.3% and 1.6% on GDP and welfare, respectively. On the other hand, it will bring substantial stimulating impacts on upstream industries such as electronic device manufacturing, machinery and the R&D sector. The

The climate and economic benefits Chapter | 8

283

substantial scale RE development will invigorate the output value of about $1.18 trillion related to upstream commercial enterprises, which will create 4.12 million occupations in 2050. As a result, renewable power sectors and the upstream industries would turn into backbone industries, equivalent to 3.4% of the GDP, which can be tantamount to other sectors such as iron and steel, construction and agriculture with 3.3%, 2.1% and 2.5%, respectively. 4. The development of RE will eventually bring momentous environmental cobenefits. They would bring down the emissions of carbon dioxide, sulphur microbes dioxide and also nitrogen oxide from the power generation.

References [1] Zhihua L. Renewable energy powers up in 2018 with fast growth, ,http://www.chinadaily.com.cn/a/201906/27/WS5d1470a6a3103dbf1432aa07.html.; 2019. [2] Dai H, Masui T, Matsuoka Y, Fujimori S. Assessment of China’s climate commitment and non-fossil energy plan towards 2020 using hybrid AIM/CGE model. Energy Policy 2011;39(5):2875
87. [3] Liang Y, Yu B, Wang L. Costs and benefits of renewable energy development in China’s power industry. Renew Energy 2019;131:700
12. [4] Zhang D, Wang J, Lin Y, Si Y, Huang C, Yang J, et al. Present situation and future prospect of renewable energy in China. Renew Sustain Energy Rev 2017;76:865
71. [5] NDRC. Medium and long-term development plan for renewable energy; 2007. [6] Schroeder M. Utilizing the clean development mechanism for the deployment of renewable energies in China. Appl Energy 2009;86(2):237
42. [7] Dai H, Xie X, Xie Y, Liu J, Masui T. Green growth: the economic impacts of large-scale renewable energy development in China. Appl Energy 2016;162:435
49. [8] Solaun K, Cerd´a E. Climate change impacts on renewable energy generation. A review of quantitative projections. Renew Sustain Energy Rev 2019;116:109415. [9] Contreras-Lisperguer R, de Cuba K. The potential impact of climate change on the energy sector in the caribbean region. . OAS Pap 2008;. [10] Sue Wing I. The synthesis of bottom-up and top-down approaches to climate policy modeling: electric power technologies and the cost of limiting US CO2 emissions. Energy Policy 2006;34(18):3847
69. [11] Sue Wing I. The synthesis of bottom
up and top
down approaches to climate policy modeling: electric power technology detail in a social accounting framework. Energy Econ 2008;30(2):547
73. [12] Dai H, Mischke P, Xie X, Xie Y, Masui T. Closing the gap? Top
down versus bottom
up projections of China’s regional energy use and CO2 emissions. Appl Energy 2016;162:1355
73. [13] Dai H, Xie Y, Liu J, Masui T. Aligning renewable energy targets with carbon emissions trading to achieve China’s INDCs: a general equilibrium assessment. Renew Sustain Energy Rev 2017;82:4121
31. [14] Dong H, Dai H, Dong L, Fujita T, Geng Y, Klimont Z, et al. Pursuing air pollutant cobenefits of CO2 mitigation in China: a provincial leveled analysis. Appl Energy 2015;144:165
74.

284

PART | II Modelling

[15] Dong H, Dai H, Geng Y, Fujita T, Liu Z, Xie Y, et al. Exploring impact of carbon tax on China’s CO2 reductions and provincial disparities. Renew Sustain Energy Rev 2017;77:596
603. [16] Xie Y, Dai H, Dong H, Hanaoka T, Masui T. Economic impacts from PM2. 5 pollutionrelated health effects in China: a provincial-level analysis. Environ Sci Technol 2016;50 (4836
4843). [17] Xie Y, Dai H, Xu X, Fujimori S, Hasegawa T, Yi K, et al. Co-benefits of climate mitigation on air quality and human health in Asian countries. Environ Int 2018;119:309
18. [18] Sue Wing I. The synthesis of bottom
up and top
down approaches to climate policy modeling: electric power technology detail in a social accounting framework. Energy Econ 2008;30(2):547
73. [19] Dai H, Masui T, Matsuoka Y, Fujimori S. Assessment of China’s climate commitment and non-fossil energy plan towards 2020 using hybrid AIM/CGE model. Energy Policy 2011;39(5):2875
87. [20] China Electricity Council, China Electricity Statistics. Beijing; 2011. [21] China Hydro Power Planning Institute. Investment cost data of hydro, wind, solar PV and biomass power (unpublished internal material). Beijing; 2011. [22] China Electricity Council.. Annual development report of China’s power sector 2013. Beijing: in Chinese]; 2013. [23] CNRE Centre. China renewable energy technology catalogue. Beijing; 2014. [24] NBOSOC (NBS) China energy statistical year book 2008. Beijing, China: China Statistics Press; 2011 [in Chinese]. [25] NBOSOC (NBS). China energy statistical year book 2008. Beijing, China: China Statistics Press; 2008 [in Chinese]. [26] National Development and Reform Commission (NDRC). Enhanced actions on climate change: China’s intended nationally determined contributions, ,http://www4.unfccc.int/ submissions/INDC/Published%20Documents/China/1/China’s%20INDC%20-%20on% 2030%20June%202015.pdf.; 2015 [accessed 3.11.2015]. [27] Energy Foundation China. China 2050 high renewable energy penetration scenario and roadmap study: executive summary,,http://www.efchina.org/Attachments/Report/report20150420/China-2050-High-Renewable-Energy-Penetration-Scenario-and-RoadmapStudy-Executive-Summary.pdf.; 2015 [accessed 2.9.2015]. [28] Connolly D, Lund H, Mathiesen BV, Leahy M. The first step towards a 100% renewable energy-system for Ireland. Appl Energy 2011;88(2):502
7. [29] Krajaˇci´c G, Dui´c N, Carvalho MDG. How to achieve a 100% RES electricity supply for Portugal? Appl Energy 2011;88(2):508
17. [30] Mathiesen BV, Lund H, Karlsson K. 100% Renewable energy systems, climate mitigation and economic growth. Appl Energy 2011;88(2):488
501. [31] IPC Change.. Climate change 2014: mitigation of climate change. Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press; 2015. [32] IPCC. Global Warming of 1.5 C, ,https://unfccc.int/topics/science/workstreams/cooperation-with-the-ipcc/ipcc-special-report-on-global-warming-of-15-degc.; 2018. [33] Emodi NV, Chaiechi T, Alam Beg ABMR. A techno-economic and environmental assessment of long-term energy policies and climate variability impact on the energy system. Energy Policy 2019;128:329
46. [34] Andersen PH, Mathews JA, Rask M. Integrating private transport into renewable energy policy: the strategy of creating intelligent recharging grids for electric vehicles. Energy Policy 2009;37(7):2481
6.

The climate and economic benefits Chapter | 8

285

[35] Connolly D, Mathiesen BV, Ridjan I. A comparison between renewable transport fuels that can supplement or replace biofuels in a 100% renewable energy system. Energy 2014;73:110
25. [36] Lund H, Kempton W. Integration of renewable energy into the transport and electricity sectors through V2G. Energy Policy 2008;36(9):3578
87. [37] Olsson L, Hjalmarsson L, Wikstro¨m M, Larsson M. Bridging the implementation gap: combining backcasting and policy analysis to study renewable energy in urban road transport. Transp Policy 2015;37:72
82. [38] Vorobiev P, Vorobiev Y. About the possibilities of using the renewable energy power sources on railway transport. J Adv Transport 2013;47(8):681
91. [39] Olson A, Jones RA, Hart E, Hargreaves J. Renewable curtailment as a power system flexibility resource. Electricity J 2014;27(9):49
61. [40] Schill W-P. Residual load, renewable surplus generation and storage requirements in Germany. Energy Policy 2014;73:65
79. [41] Ueckerdt F, Brecha R, Luderer G, Sullivan P, Schmid E, Bauer N, et al. Variable renewable energy in modeling climate change mitigation scenarios. In: Proceedings of the 2011 international energy workshop in Standford. US. 2011. [42] Luderer G, Krey V, Calvin K, Merrick J, Mima S, Pietzcker R, et al. The role of renewable energy in climate stabilization: results from the EMF27 scenarios. Climatic Change 2014;123(3):427
41.

This page intentionally left blank

Part III

Applications

This page intentionally left blank

Chapter 9

The utilization of renewable energy for low-carbon buildings Yuan Chang and Yayin Wei School of Management Science and Engineering, Central University of Finance and Economics, Beijing, P.R. China

Chapter Outline 9.1 Building and energy and environmental challenges 289 9.2 Net-zero energy building and lowcarbon building 290 9.3 Building life-cycle systems and greenhouse gas emissions 292 9.4 Renewable energy technologies for low-carbon buildings 292

9.1

9.4.1 Building material extraction and transportation 292 9.4.2 Building construction 295 9.4.3 Building operation 296 9.5 Path forward for advancing lowcarbon buildings 305 References 307

Building and energy and environmental challenges

Buildings are basic necessities for the smooth and prosperous operation of society. As the world continues to urbanize, approximately 68% of the world population (which was estimated to be 7 billion people) will live in cities by 2050 [1], yielding a substantial demand for buildings. Although developed countries are experiencing slow growth of annually new-built buildings, large-scale building constructions are undergoing in the developing and emerging economies such as China and India. For example, more than 4 billion square meter (m2) of buildings have been annually built in China since 2014, about 9 times of the total value of the EU-28 [2]. Undoubtedly, the expanding built environment will pose severe energy and environmental challenges to our planet, which determines the extent to which sustainable development could be achieved. Compared with other consumer goods such as appliances and vehicles, buildings have complex and long supply chains, involving many manufacturing sectors in the economy. The production of building materials such as steel and concrete is energy- and emission-intensive, resulting in high energy Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00009-1 © 2021 Elsevier Inc. All rights reserved.

289

290

PART | III Applications

and environmental impacts. In addition, buildings have long life time, typically ranging from 50 to 100 years. During such long-period existence, the provision of building services including heating, cooling, lighting, cooking and hot water supply consumes a large amount of energy (i.e. building operational energy), resulting in a dominant share of operational energy in building life-cycle energy use, about 80% [3]. As a result, the building sector is responsible for approximately 40% of global energy use [4] and 40% of the world’s energy-related annual greenhouse gas (GHG) emissions [5]. Thus, the sustainability state of the building sector is critical for today’s transition towards a low-carbon human-nature system. Renewable energy resources provide a pathway to the creation of sustainable built environment, attributable to their inexhaustible reserves and low carbon content. Renewable energy consists of biomass, solar, wind, ocean, hydro and geothermal, and currently scale commercial utilizations primarily focus on liquid fuel production (such as ethanol, methanol and biodiesel), electricity generation and heat production [6]. The cleaner and decarbonized grid enabled by hydropower plants and solar and wind farms contributes to mitigating the carbon footprint of building materials, which helps to reduce building embodied GHG emissions. Besides, since the activity of building occupants such as heating, cooling and lighting heavily relies on electricity and heat energy consumption, renewable energy utilizations, either in an onsite or off-site manner, contribute to expanding the energy supply and mitigating the environmental burdens of building operation. Therefore, unlocking the green opportunity of renewable energy utilizations across the life-span of buildings has become a hot pursuit for both academia and industry.

9.2

Net-zero energy building and low-carbon building

Given the potential energy and environmental benefits associated with renewable energy utilizations in buildings, two initiatives about building sustainability globally rise, the net-zero energy building (NZEB) and the low-carbon building (LCB), respectively. The concept of NZEB is overwhelmingly interpreted by utilizing renewable energy to satisfy building energy need. But, until now, global community has not reached consensus on the definition of NZEB [7]. The National Renewable Energy Laboratory (NREL) in the United States has classified NZEB as a range of buildings, and the top of the classification is ‘a building that offsets all of its energy use from renewable energy resources available within the footprint’, whereas the categorization’s lowest end refers to ‘a building that achieves an NZEB definition through a combination of on-site renewables and off-site purchases of renewable energy credits’ [8], see Fig. 9.1. In addition to the sole reliance on building on-site renewable energy sources, the bottom concept of NZEB includes potential utilizations of off-site renewable energy. For example, consuming the electricity generated by wind or solar farms that might be

The utilization of renewable energy for low-carbon buildings Chapter | 9

291

FIGURE 9.1 The concept of NZEB. NZEB, Net-zero energy building.

thousands of miles away from the building site and transmitted by highvoltage power lines. Undoubtedly, the off-site renewable energy consumption helps architects and building occupants to overcome the constraint of insufficient renewable energy resources and limited building site area, greatly reducing the hurdles associated with NZEB delivery. However, interactions with commercial energy suppliers and heavy reliance on national high-voltage transmission network also bring economically and technologically complex environment to NZEBs in society. Therefore, stakeholders should leverage a variety of technological, managerial and policy measures to jointly advance the seamless delivery of NZEBs. Rooted from the United Nations Framework Convention on Climate Change (UNFCCC) adopted in Rio in 1992, the concept of low carbon development has become the direction of human development. NZEBs are inherently low-carbon and cleaner attributable to the environmentally friendly attribute of renewable energy. The concept of LCBs broadly refers to buildings that emit significantly less GHG emissions than regular buildings. For the qualification of LCBs, there is no GHG mitigations threshold. But considering that low-carbon development strategies aim to change our practice modes to reduce reliance on fossil fuels and contribute to global efforts on climate change mitigation, LCBs should reduce their GHG emissions by 80% or more compared to traditional buildings so as to achieve ‘climate-change-neutral’ [9]. Given the wide range of NZEB definitions, the carbon mitigation of LCBs estimated by different accounting boundaries might significantly varies, calling for globally standardized assessment procedures, see Fig. 9.2. In this chapter, we will not explore the threshold GHG reduction for a building to be qualified as LCB, but comprehensively illustrate how renewable energy technologies can be leveraged by buildings throughout their entire life span to create low-carbon built environment.

292

PART | III Applications

FIGURE 9.2 Conceptual tier of NZCB. NZCB, Net-zero-carbon building.

9.3 Building life-cycle systems and greenhouse gas emissions Building life-cycle (cradle-to-grave process) consists of five phases; they are raw material extraction, building material production, construction, operation and maintenance and demolition, see Fig. 9.3. In the life-cycle assessment (LCA) community, the first three phases are defined as ‘cradle-to-gate’ (or embodied) process, whose energy and environmental impacts highly rely on the green performance of industrial, transportation and construction sectors due to the production and transportation of building materials and the on-site construction activity of buildings. The energy and GHG emissions of building operation are primarily determined by building physical features, occupants’ behaviour and the energy mix of building external community. Renewable energy technologies hold the promise to mitigate the energy and environmental impacts of buildings throughout their entire life-cycles, enabling the delivery of LCBs. This is especially the case for building operation, whose GHG emissions dominate buildings’ life-cycle carbon footprint. Renewable energy technologies can satisfy the heat and electricity demand associated with the provision of various energy services, see Fig. 9.4.

9.4 Renewable energy technologies for low-carbon buildings 9.4.1

Building material extraction and transportation

The complex and interlinked sustainability systems result in the energy and environmental opportunities of a product go beyond its operation and use and upstreamly extend to cradle processes. The cognition of LCBs has also penetrated to the extraction and production of building materials and their

The utilization of renewable energy for low-carbon buildings Chapter | 9

293

FIGURE 9.3 Building life-cycle systems.

FIGURE 9.4 Renewable energy utilizations for energy services in buildings.

associated transportations. For example, as one of the worldwide green building rating systems, the Leadership in Energy and Environmental Design developed by the U.S. Green Building Council keeps evolving to cover the complete delivery system of building projects, and its credit category
‘Materials and Resources’
gives credit for using the materials that have published LCAs and environmental product declarations [10]. Globally, advances in automation, information and digitalization have substantially upgrade many products in our daily life, such as electrical appliances, communication products and automobiles, over the last decades, while buildings have been inactive to evolve. Cement, steel, gravel and sand are still the materials for buildings. Undoubtedly, the increased construction

294

PART | III Applications

activities driven by the urbanization of global population will pose huge demands for the products of mining industry. Mining industry is energyintensive, accounting for 1.25%
11% of global energy demand [11]. The sector’s on-site production highly relies on conventional fossil-based fuels, such as oil, gas and coal, taking up about two-third of final energy consumption, while the share of electricity is about one-third, which in most cases is also generated by fossil fuel and transmitted by grid. In 2014, only 0.001% of the final energy consumption of the mining and quarrying sector was generated by renewable energy technologies (such as solar and wind) installed on site [12], urging mining companies to expand renewable energy use. Notably, mining operation requires stable power supply without interruptions, which however might become difficult because mining production always expand into new and often remote locations [13]. On-site deployments of the solar and wind power technologies contribute to the operation of off-grid mining projects. However, to address the limitation of intermittency and reliance on site-specific energy endowments, battery storage and back-up system (usually diesel power generator) are required to build microgrid for uninterruptible power supply. Fortunately, the continuously decreasing battery prices and solar and wind power generation costs help to advance renewable energy penetrations in the mining industry. For example, the Degrussa copper/gold mine in western Australia integrated batteries in its solar-diesel hybrid power system, enabling an annual diesel saving of 5 million litres and a reduction of CO2 emission by 12,000 tonnes per year [14]. The delivery of building materials between mining site, manufacturing plant and construction site involves road and railway transportations, and renewable energy resources help to mitigate the carbon footprint associated with cargo transports, contributing to reduce building embodied GHG emissions. Biofuels, such as ethanol, enable 19%
48% ‘cradle-to-wheel’ GHG reductions compared to conventional gasoline-driven vehicles [15], and this reduction potential could be as high as 101%
115% if advanced biofuels are considered [16]. Comparatively, railway transportation utilizes renewable energy indirectly, consuming the low-carbon electricity generated by various renewable energy technologies, such as wind turbines and solar photovoltaics (PV). However, to expand the market penetration of biofuels in the real world, policy makers cannot be blind to the challenges inherited in biomass utilizations, such as the land use demand associated with energy crops (like sugarcane and maize) cultivation. This is especially the case for countries with high population, and the potential land competition between energy and food crops seems unavoidable. Furthermore, the growth of energy crops requires irrigation, which might exacerbate the water stress in many regions in Asia. As a result, the scale use of biofuels in one country’s transportation sector must be completely evaluated in the food
energy
water nexus context to ensure the secure delivery of the lifeline resources [17].

The utilization of renewable energy for low-carbon buildings Chapter | 9

9.4.2

295

Building construction

Building construction removes buildings from drawings to reality. The GHG emissions during building construction mainly derived from three sources: first, construction activities involve the use of various machinery, such as excavator, bulldozer and grader, and the diesel consumption of these machinery emits GHG to atmosphere; Second, for construction sites that do not have access to municipal grid, diesel generators are used for construction onsite electricity supply to power equipment (such as concrete mixer and construction hoist) and construction trailers and dormitory. Obviously, diesel generator has GHG emissions; Third, coal is usually consumed to generate heat for concrete curing and dormitory heating (in cold regions), and the combustion of coal is a significant GHG emitter. Up to present, utilizations of renewable energy sources during building construction are relatively rare. This might be because construction site usually confronts limited area, which barricades the scale deployment of renewable energy technologies. In addition, unlike mining productions, the period of building construction is short (1
2 years in most cases), making the use of renewable energy on construction site not economically feasible. As a result, contractors mainly used solar energy in auxiliary activities during building construction such as security lighting and hot water supply in construction workers’ dormitory. However, today’s ever-changing economic and ecological environment requires evolution of construction industry towards an efficient and sustainable direction. Building prefabrication, defined as the building delivery approach that manufactures building components (such as stair, floor slab and balcony) in factories, transports them to building site and assembles onsite. Due to the strength in boosted productivity, quality production of building parts and construction waste reductions, building prefabrication holds promise for the sustainable transformation of construction industry [18]. For example, the Chinese government mandates that 15% of the nation’s annual new construction (by building floor area) must be built using the prefabricated approach by 2020, and this number will rise to 30% by 2025 [19]. Compared to onsite construction, the factory production of building components helps to quality control attributable to the more controllable temperature and moisture production conditions, which are less affected by adverse weathers. Moreover, the bulk production of building components is easier to achieve economies of scale, which not only contributes to cost reductions but also unlocks the potential of renewable energy utilizations, for example applying solar PV in the concrete curing of precast manufacturing. As a pioneer prefabricated design and manufacturing company aiming to achieve sustainable construction, materials, processes and operations, the Plant Prefab commits to being net carbon neutral by 2028 by utilizing renewable energy and carbon offsets [20].

296

PART | III Applications

FIGURE 9.5 The operational energy end uses of residential building.

9.4.3

Building operation

The operational energy of a building refers to the energy that is used for heating, cooling, lighting, cooking, appliances and hot water provision, see Fig. 9.5. Musall et al. [21] has identified almost 300 net zero or almost netzero energy buildings constructed worldwide, including both residential and commercial buildings, and found that about one-third of these buildings reduced energy use by 60% compared to local conventional building. Generally, buildings with a spacious roof (e.g. to install PV arrays and wind turbines) and site area (e.g. to drill geothermal wells and to construct biogas digester) in relation to the building’s energy demand are more likely to achieve net-zero energy use. As such, NZEBs are more difficult to build in urban areas because of high occupant density and limited urban land area. (1) Solar energy Buildings utilize solar energy mainly through three types of technology: the solar PV, solar thermal and photovoltaic
thermal (PV-T) technology, respectively. Applications of solar energy technology in buildings are affected by various factors, including the characteristic of solar irradiance, the area of building roof, occupant density and building ownership. These factors need to be comprehensively examined by decision makers before utilizing solar energy in buildings. It should also be noted that since buildings have long lifespan, the availability of solar resource in future decades during building operation should also be evaluated to completely understand the technology’s energy, environmental and economic benefits. The climate change caused by increasing GHGs in atmosphere affects atmospheric water vapour content, cloud cover, rainfall and turbidity, which may impact the resource potential of solar energy [6]. In addition, the deteriorating air quality could also significantly decrease solar

The utilization of renewable energy for low-carbon buildings Chapter | 9

297

FIGURE 9.6 Solar PV technology systems. PV, Photovoltaics.

irradiance. For example, anthropogenic aerosol emissions in China such as PM2.5 block about 20% of sunlight from reaching solar panel arrays [22]; Historically air pollutions reduced China’s PV potential by 11%
15% on average between 1960 and 2015 [23].

9.4.3.1 Solar photovoltaics The solar PV technology primarily uses solar panels to absorb and convert sunlight into electricity. In addition, a balance of system is also needed to enable the utilization of solar electricity, see Fig. 9.6. To be specific, a solar inverter is installed to convert the output from direct current to alternating current electricity. Other electrical accessories, usually including wiring and switches, battery bank and mounting system, are also indispensable to set up a working system. Solar cell is the core of PV technology. With the development of photoelectric materials, solar cells could be divided into three generations. The first-generation solar cell refers to crystalline silicon PV, including monocrystalline silicon (mono-Si) and polycrystalline silicon (poly-Si). The second-generation solar cell refers to the thin-film materials, mainly including amorphous silicon (a-Si), copper indium gallium selenium (CIGS), cadmium telluride (CdTe), dye-sensitized (DSSC) and organic solar cell. The CIGS and CdTe are mainstream products in today’s market attributable to their low economic cost and high conversion efficiency, which could be as high as 23.3% [24] and 22.1% [25], respectively. The thirdgeneration PV cells are to date the conceptual products, such as perovskite solar cell, and have not been commercially applied. For buildings, PV technologies are primarily installed on building rooftops. The majority of these systems consist of solar cells that are mounted off the surfaces of roofs by racking hardware [26]. However, a new solar market segment
building-integrated photovoltaics (BIPV)
rises, calling for deploying the PV materials in different parts of building envelop such as roof, skylights or facades, for electricity generation. The performance of applying PV technologies to buildings is usually evaluated by technologies’ energy payback time and ‘cradle-to-wire’ GHG emissions, see Table 9.1. However, since different regions and countries vary in energy mix, PV

298

PART | III Applications

TABLE 9.1 The energy and GHG implications of using different PV technologies in buildings. Location

Installation

PV types

Energy payback time (year)

GHG emissions (g CO2e/ kWh)

A total of 111 countries in northwest and southwest Europe, Sahara and Tropics

Building roof-top mounted, mono-Si

monoSi

2.9

58.0

polySi

1.7

33.3

CdTe

0.7

13.4

CIGS

1.3

22.5

a-Si

1.7

31.1

Note: The EPBT and GHG emissions are the mean value of 111 countries in the study area, and data source is Louwen et al. [27]. mono-Si, Monocrystalline silicon; poly-Si, Polycrystalline silicon; CdTe, Cadmium telluride; a-Si, Amorphous silicon; CIGS, Copper indium gallium selenium; GHG, Greenhouse gas; PV, Photovoltaics

production technology and transportation pattern and the energy and GHG emissions associated with the same PV system production might be significantly different. For example, for the same poly-Si PV module (manufactured by identical process and technology in China, Australia and United States), the ‘cradle-to-wire’ GHG emissions of China-made PV were 86% and 590% higher than those of the Australian-made and US-made PV, respectively [28]. Thus region-specific GHG emissions accounting is a prerequisite for the robust decision making of building PV deployment.

9.4.3.2 Solar thermal Solar thermal system harvests energy from the sun and uses them for various heating services in buildings, mostly the domestic hot water heating. Hot water provision consumes a large amount of energy, and the growing demand for hot water in buildings is an important driver of a country’s energy consumption, especially for the emerging economies that are experiencing and will continue to experience rapid urbanization and continuously rising living standards. For example, the energy use of urban residential hot water in China has increased from 1.6 million metric tons of coal equivalent (MTCE) in 1996 to 14.5 million MTCE in 2011, accounting for 10% of the country’s total building operational energy [17]; In the United States, the energy consumption associated with water heating is higher compared to that of water supply and treatment [29]. The minimum temperature

The utilization of renewable energy for low-carbon buildings Chapter | 9

299

FIGURE 9.7 Building solar thermal system.

of the domestic hot water is typically between 40 C and 65 C, and buildings could use solar thermal technology to preheat the fresh water to reduce the demand for nonrenewable energy sources [30]. The building solar thermal system consists of at least a solar collector, pipes and a storage tank, but additional components including controller, pumps and valves are usually included, see Fig. 9.7. For lower-temperature (up to 100 C) heat provision, three types of collector are commonly used, including the evacuated tube collectors, the flat plate collectors and the unglazed absorbers [31].

9.4.3.3 Photovoltaic
thermal The PV-T technology simultaneously utilizes solar energy for electricity generation and domestic hot water preheating. Compared to conventional PV solar technology, the PV-T system also includes a solar thermal collector that is installed on the back of the PV panel. Since PV-T technology dual utilizes the solar radiation received for thermal and electrical energy provision, buildings with limited roof area should be prioritized in the technology deployment for higher energy output of per unit building roof space compared to a separate PV and solar thermal system. However, the thermal efficiency of PV-T system is calculated to be less than half of that of the solar thermal systems, the experience from USA showed that to enable economically cost-effective application of PV-T system in buildings, critical factors including accurate information about the daily hot water usage, load profile of the target building, careful selection of heat exchangers and complete calculations of local utility fees and governmental incentives should be considered [32]. (2) Wind energy To advance the transition of our built environment towards a more sustainable direction, wind turbines are considered by building owners and

300

PART | III Applications

architects to be mounted on, or integrated into buildings to harness wind for electricity generation. Given the limits of space and area around building site, microwind turbines are usually installed. Components of the wind power technology consist of nacelle and rotor. The nacelle houses all of the internal parts of the turbine, mainly including gear box, generator, controller and yaw drive, while rotor includes the elements of the turbine that rotate in the wind, including hub and blades, and for pitch-regulated wind turbines, the blade pitch mechanisms and bearings [33], see Fig. 9.8. It should be noted that not all wind turbines include gearboxes, some may use a different ‘direct drive’ generator design. It is through the rotor that the wind energy is transformed into mechanical energy that turns the main shaft of the wind turbine. The blade root is bolted to the hub. Generally, the wind turbines deployed by buildings can be divided into two categories: the horizontal axis wind turbine (HAWT) and the vertical axis wind turbine (VAWT), see Fig. 9.9. Turbine blades of the HAWT rotate around a horizontal axis, while the main rotor shaft of the VAWT is set transverse to the wind. Comparatively, the VAWT is a new rising technology with fast development in the last decade. However, debates still haunt as which technology is more suitable for buildings in urban environment. The HAWT technology yields higher efficiency in conversing wind energy to electricity, while the VAWT technology has fewer components, and the

FIGURE 9.8 Components of wind turbine.

The utilization of renewable energy for low-carbon buildings Chapter | 9

301

blades of a VAWT harvest wind in any direction without directional orientation. Furthermore, the VAWT technology generates fewer vibrations and makes less noise [34]. Buildings can utilize wind energy in both on-site and off-site manners. To decouple economic growth and GHG emissions, countries began to construct wind farms to enable lower-carbon and cleaner power transition, which undoubtedly is beneficial to the electricity consumption throughout building life-cycle systems. For 1-kWh electricity provision (from cradle to wire), the wind electricity only emits less than 1% of the GHG emissions of coal-fired electricity [35]. Notably, large-scale explorations of offshore wind power also gradually emerge. The quantification of China’s first offshore wind farm (the Donghai Bridge Offshore Wind Farm, which is also the first offshore wind farm outside Europe) showed a cradle-to-wire GHG footprint of 25.5 g CO2e/kWh, providing a low-carbon solution to the high-density built environment in eastern coastal area of China [36]. Integrations of wind power technology in buildings are also not rare. For example, the Guangdong Tobacco Tower in China and the Bahrain World Trade Center. Compared to open spaces, the terrain in urban environments is rougher, resulting in a reduced and more turbulent wind flow [37], but mounting turbines on buildings helps to mitigate GHG emissions in cities. Thus, wind power technology is more applicable in rural area, where building sites are

FIGURE 9.9 Horizontal axis wind turbine and vertical axis wind turbine. HAWT, Horizontal axis wind turbine; VAWT, vertical axis wind turbine.

302

PART | III Applications

relative spacious for wind turbine installation, and the electricity consumption of occupants is much lower. However, some issues associated with integrating wind power technology in buildings must be fully considered by architectural designers and building owners. The aeroacoustic noise and mechanical sound generated by wind turbines might disturb building occupants or people around the building, especially for those with low tolerance on noise, such as residents, students, and patients. Furthermore, installing wind turbine blades on a conventional building might be visually and aesthetically unacceptable by occupants, who might even have safety worries about the facility. (3) Biomass energy Globally, biomass accounts for the largest share in building energy sources, and it has been widely used for space heating, hot water production and cooking in many developing countries [38], especially for residential buildings in rural area [3]. Compared to open fires, advanced biomass stoves enable 30%
60% fuel savings and 80%
90% indoor air pollution reductions [39], see Fig. 9.10. Moreover, biomass also provides residents a pathway to low-carbon living. The CO2 emission of a family-sized biogas system in villages in China is only 11% of that of coal burning alternative, reducing CO2 emissions by 89% [40], see Fig. 9.11. In addition, biomass can also be utilized for electricity and heat generation, which can benefit more buildings at urban or even regional scale. For example, Salix psammophila, a desert shrub that can easily survive and grow in the desert because of its welldeveloped root system, can be used for power generation, and the estimation based on a Salix direct-fired power generation system in Inner Mongolia, China, showed that the ‘planting-to-wire’ GHG emissions were 114 g CO2e/ kWh [41], only one-ninth of the emissions of coal-fired electricity, about 980 g CO2e/kWh on average [42]. Similarly, wood pellets, a kind of biomass

FIGURE 9.10 Utilization of wood pellets for thermal-based services in building.

The utilization of renewable energy for low-carbon buildings Chapter | 9

303

FIGURE 9.11 Household biogas system.

solid fuel produced by wood residue, are favourable feedstock energy attributable to their high calorific value, low ash content and slagging rate. For per unit (1-GJ) heat generation, the ‘fuel-to-heat’ GHG emissions of wood pellets were 132 kg CO2e, whereas the footprint of coal-fired heat was 196 kg CO2e. Notably the GHG emissions of 1-GJ wood pellet-fired heat could be as low as 12 kg CO2e if carbon neutral effect (the GHG absorption during tree growth offsets the emission during tree combustion) was considered [43]. Therefore, considering the substantial energy demand resulted from the high occupant density of urban residential buildings and the large amount of electrical-driven equipment in office buildings, as well as those buildings’ limited site area, utilizing biomass energy in an off-site manner (i.e. purchasing biomass-based electricity and heat) should be the priority for urban building owners, whereas the deployment of biogas and efficient biomass boilers could be considered by rural residents. However, biomass energy utilizations also meet barriers. First, the feedstock shortage caused by other production processes and sectors might undermine the availability of biomass feedstock for energy generation. For example, paper mills also use wood residue as raw materials for their production, which reduces the supply of wood pellets for heat generation. Second, the diverse feedstock sources and lack of technical standards can result in the unstable heat value of biomass energy. Third, the economic performance of biomass energy application is mainly affected by feedstock collection, and the high labour costs for the energy crop cultivation might make biomass energy less economic attractive. Compared with coal-fired power generation, most of biomass-fired technologies are not economically competitive, requiring government subsidies to advance society-wide biomass energy utilizations. (4) Geothermal energy Geothermal energy is heat generated and stored below the earth’s surface which can be captured and harnessed as a clean and undepletable energy source for heating and power generation. Globally, the geothermal energy is

304

PART | III Applications

primarily used for providing heating service. The IPCC [6] projected that geothermal heat provision would be 13.6 EJ by 2020, accounting for 68% of the total geothermal primary energy supply (2.01 EJ). For buildings, geothermal energy is mainly used for space heating and cooling (see Fig. 9.12), water heating and cooking. The geothermal technology for building heating can be categorized into two types of system: the open and closed loop. The open loop (single pipe) system directly utilizes the water heated by geothermal underground to circulate through radiators, while the closed loop (double pipe) technology uses heat exchangers to transfer heat from the geothermal water to a closed loop that circulates heated freshwater through the radiators [6]. Both the two systems dispose the geothermal water into injection wells. The geothermal power generation technology (flashed steam or enhanced geothermal systems) directly decarbonizes grids by eliminating fossil fuel combustion for electric power generation, providing an off-site option to enabling LCBs. Comparatively, geothermal heat pumps (GHP) mitigate building GHG emissions in an indirect way, that is taking advantage the relatively constant ground or groundwater temperature to provide space heating and cooling. Practices in China showed that GHP has favourable energy, economic and environmental performances [44]. A high-rise and large-scale education building (with a floor area of 49,166 m2) in north China indicated that deploying a 750-kw GHP system can reduce building heating and cooling energy use by 84% and 83% compared to traditional municipal heating and air conditioner cooling [45]. The 8-million-yuan GHP technology could be economically cost-effective in 7.4 years, which is shorter compared to that of PV installation on buildings, about 12 years [46,47]. The energy and GHG payback times of the GHP were 0.5 and 0.3 years, respectively.

FIGURE 9.12 Geothermal-based building heating and cooling system (closed loop).

The utilization of renewable energy for low-carbon buildings Chapter | 9

9.5

305

Path forward for advancing low-carbon buildings

(1) Comprehensively and specifically assessing the energy and GHG reductions of applying renewable energy technologies in buildings. Robust decision makings and building designs associated with utilizing renewable energy technologies to a great extent depend on precise and specific understanding of the technologies’ energy and environmental performance. To avoid the narrow-view decision makings, renewable energy technologies must be assessed in a life-cycle manner, that is using the LCA approach, which holistically considers the technologies’ impact from cradle to grave. However, LCA results would be significantly affected by multiple factors, mainly including the inconsistent system boundary (inherited in the selection of different LCA modeling methods) and the temporal, spatial and technological specificity of parameters and variables used in the assessment. For example, in terms of the ‘plantingto-wire’ GHG accounting of biomass power generation technology in China, results of the process-based LCA could be 11% lower than those of hybrid models because of the truncations on services and accessory equipment; the temporary and spatial uncertainty brought by parameter settings for various time and regions/countries could be up to 10% and 16%, respectively [48]. Furthermore buildings have long use period, during which many factors might significantly vary, for example the increase in labour costs and the cancellation of government’s incentives. These variations would influence the cost-effectiveness of renewable energy technologies and thus should be considered by building owner and architect. (2) Optimizing the on-site and off-site deployment of renewable energy technologies. The applicability of renewable energy technologies in buildings depends on building function and site conditions. Although it has been advocated that renewable energy should be prioritized for building on-site use [49], buildings in urban area, especially the residential buildings in Asian cities with high occupant density, always meet difficulties in deploying the renewable energy technologies. As such, off-site renewable energy utilizations should be encouraged by levering flexible policies and regulations (such as carbon offset accredited by green power purchase) to advance the market penetration of renewable energy to maximize their GHG abatement benefits in the building sector in broader areas, for example within a region or a country. Comparatively rural residential buildings have a wide range of options to use the renewable energy technologies such as solar PV, wind turbine and biogas, especially for people living in remote regions where access to grid facilities is limited. Thus, integrating renewable energy in buildings not only helps to mitigate GHG emissions but also enhances residents’ quality of life, leading to an improved energy justice in the

306

PART | III Applications

society. Notably, advances in batteries for energy storage make microgrids feasible for rural electrification. However, reliability and financial viability should be the minimum threshold for the technology deployment [50]. Also government involvement and capital subsidies are essential for the success of microgrid projects. (3) Balancing renewable energy technology deployment for new-built buildings and existing buildings. While energy efficiency is increasingly embedded in new building designs and construction, the majority of today’s building stock do not incorporate renewable energy use, and these buildings will be in place in the foreseeable future [51]. For the same amount of energy reduction, investing in efficiency measures would cost only half of installing renewable energy generating capacity [52]. However, government and owners should not be blinded to the potential utilization of renewable energy in existing buildings. Renewable energy-related building retrofits could still be economically feasible, exerting positive externality of energy reduction and GHG mitigation on the society. In addition to the factors (such as the sufficiency of renewable energy resource, the area of building site eligible for technology installation, the cost of utilities and the availability of government incentives) that must be considered before deploying renewable energy technology in new-built buildings, existing building retrofits need to pay extra attention to owners’ willingness to retrofits. Since most of urban residential buildings have multiple owners, the cost and benefit of the retrofits should be reasonably allocated. In addition, owners’ desire to preserve or not alter existing architectural features should also be respected. For historical buildings, retrofits must be conducted under the requirements of code and regulation. (4) Unlocking LCB potentials through high-performance energy end-use devices and occupants’ green behaviours. Although the renewable technologies physically enable the delivery of LCBs, the green potentials associated with various high-performance electrical appliances and occupants’ behaviours are not neglectable. To achieve LCBs, the deployment of renewable technologies should be integrated with energy-efficient end-users and environmental responsible living style. For example, the Catalyst Building, a 1390-m2 mixed-use higher education and commercial office building newly built in Washington, USA, will be accredited as zero-energy and zero-carbon by the International Living Future Institute. Due to the near passive house building envelope, the energy use of the building’s HVAC equipment was significantly lowered, whereas plug loads and lighting energy account for about 60%
70% of the total building energy use [53], indicating the necessity to engage occupants to reduce plug load energy for low-carbon built environment creation. Furthermore, unlocking the hidden energy reduction and GHG mitigation potentials of buildings through energy-efficient end-use

The utilization of renewable energy for low-carbon buildings Chapter | 9

307

devices and occupants’ behavioural change are especially critical for those emerging economies such as China and India, considering their raising pursuit for a more comfortable indoor environment.

References [1] United Nations. 2018 Revision of world urbanization prospects. New York: United Nations; 2018. [2] EU (European Commission). EU buildings database. ,https://ec.europa.eu/energy/en/eubuildings-database.; 2016 [accessed 25.11.2019]. [3] Chang Y, Ries R, Wang YW. Life-cycle energy of residential buildings in China. Energy Policy 2013;62:656
64. [4] WBCSD (World Business Council for Sustainable Development). Energy efficiency in buildings
transforming the market. Geneva: WBCSD; 2009. [5] IEA (International Energy Agency) and UNEP (United Nations Environment Programme). 2018 Global status report: towards a zero-emission, efficient and resilient buildings and construction sector. Paris: IEA; 2018. [6] IPCC (Intergovernmental Panel on Climate Change). Special report on renewable energy sources and climate change mitigation. Cambridge, UK/New York: Cambridge University Press; 2011. [7] Luo T, Tan Y, Langston C, Xue X. Mapping the knowledge roadmap of low carbon building: a scientometric analysis. Energy Build 2019;194:163
76. [8] Crawley D, Pless S, Torcellini PA. Getting to net zero. NREL report no: NREL/JA-55046382. Golden, CO: National Renewable Energy Lab; 2009 [9] Stern N, Stern NH. The economics of climate change: the stern review. Cambridge: Cambridge University Press; 2007. [10] USGBC (U.S. Green Building Council). LEED v4 for building design and construction. Washington, DC: USGBC; 2019. [11] Natali P, Haley K. Insight brief: towards sustainable mining, ,https://d231jw5ce53gcq. cloudfront.net/wp-Content/uploads/2017/07/RMI_Insight_Brief_Toward_Sustainable_ Mining_2017.pdf.; 2017 [accessed 24.09.2019]. [12] McLellan B, Choi Y, Ghoreishi-Madiseh S, Hassani F. Emissions and the role of renewables: drivers, potential, projects and projections. chapter in mining and sustainable development. London: Routledge; 2018. [13] Maennling N, Toledano P. The renewable power of the mine. New York: Columbia Center on Sustainable Investment; 2018. [14] Neoen. DeGrussa solar hybrid project, ,https://www.oecd.org/dev/inclusivesocietiesanddevelopment/Session-4-deGrussa-Solar-hybrid-Project.pdf.; 2017 [accessed 26.11.2019]. [15] Wang M, Han J, Dunn J, Cai H, Elgowainy A. Well-to-wheels energy use and greenhouse gas emissions of ethanol from corn, sugarcane and cellulosic biomass for US use. Argonne, IL: Argonne National Laboratory; 2012. [16] EESI (Environmental and Energy Study Institute). Biofuels versus Gasoline: the emissions gap is widening, ,https://www.eesi.org/articles/view/biofuels-versus-gasoline-the-emissions-gap-is-widening.; 2016 [accessed 20.11.2019]. [17] Chang Y, Li GJ, Yao Y, Zhang LX, Yu C. Quantifying the water-energy-food nexus: current status and trends. Energies 2016;9(2):65.

308

PART | III Applications

[18] Chang Y, Li XD, Masanet E, Zhang LX, Huang ZY, Ries R. Unlocking the green opportunity for prefabricated buildings and construction in China. Resour Conserv Recycl 2018;139:259
61. [19] SCC (The State Council of the People’s Republic of China). Some opinions of the CPC central committee and the state council on further strengthening the management of urban planning and construction. Beijing: SCC; 2016. [20] Pacheco, A. Plant prefab pledges to achieve carbon neutrality by 2028, ,https://archinect. com/news/article/150161756/plant-prefab-pledges-to-achieve-carbon-neutrality-by-2028.; 2019 [accessed 18.09.2019]. [21] Musall E, Weiss T, Voss K, Lenoir A, Donn M, Cory S, et al. Net Zero energy solar buildings: an overview and analysis on worldwide building projects. In: EuroSun Conference; 2010. [22] Li X, Wagner F, Peng W, Yang J, Mauzerall DL. Reduction of solar photovoltaic resources due to air pollution in China. Proc Natl Acad Sci 2017;114(45):11867
72. [23] Sweerts B, Pfenninger S, Yang S, Folini D, van der Zwaan B, Wild M. Estimation of losses in solar energy production from air pollution in China since 1960 using surface radiation data. Nat Energy 2019;4:657
63. [24] HZB (Helmholtz-Zentrum Berlin). World record for tandem perovskite-CIGS solar cell, ,https://www.helmholtz-berlin.de/pubbin/news_seite?nid 5 20769;sprache 5 en; seitenid 5 .; 2019 [accessed 20.11.2019]. [25] USDOE (U.S. Department of Energy). Cadmium telluride, ,https://www.energy.gov/eere/ solar/cadmium-telluride.; 2019 [accessed 20.12.2019]. [26] James T, Goodrich A, Woodhouse M, Margolis R, Ong S. Building-integrated photovoltaics (bipv) in the residential sector: an analysis of installed rooftop system prices. NREL Report No: NREL/TP-6A20-53103. Golden, CO: National Renewable Energy Lab; 2011. [27] Louwen A, Schropp R, Sark W, Faaij A. Geospatial analysis of the energy yield and environmental footprint of different photovoltaic module technologies. Sol Energy 2017;155:1339
53. [28] Yao Y, Chang Y, Masanet E. A hybrid life-cycle inventory for multi-crystalline silicon PV module manufacturing in China. Environ Res Lett 2014;9:114001. [29] USDOE (U.S. Department of Energy). Energy demands on water resources. Washington, DC: DOE; 2006. [30] Maurer C, Cappel C, Kuhn T. Progress in building-integrated solar thermal systems. Sol Energy 2017;154:158
86. [31] Trenkner U, Dias P, Preiß D, Noyon X. Integrating solar thermal in buildings
a quick guide for architects and builders. Paris: UNEP; 2014. [32] Dean J, McNutt P, Lisell L, Burch J, Jones D, Heinicke D. Photovoltaic-thermal new technology demonstration. Golden, CO: National Renewable Energy Lab; 2015. [33] WBDG (Whole Building Design Guide). Wind technology, ,https://www.wbdg.org/ resources/wind-technology.; 2016 [accessed 18.12.2019]. [34] Boˇsnjakovi´c M. Wind power buildings integration. J Mech Eng Autom 2013;3 (2013):221
6. [35] Masanet E, Chang Y, Gopal A, Larsen P, Morrow. W, Sathre R, et al. Life-cycle assessment of electric power systems. Annu Rev Environ Resour 2013;38:107
36. [36] Yang JH, Chang Y, Zhang LX, Hao Y, Yan Q, Wang CB. The life-cycle energy and environmental emissions of a typical offshore wind farm in China. J Clean Prod 2018;180:316
24.

The utilization of renewable energy for low-carbon buildings Chapter | 9

309

[37] Dayan E. Wind energy in buildings: power generation from wind in the urban environment
where it is needed most. Refocus 2006;7(2):33
34, 36, 38. [38] IEA (International Energy Agency). Energy balances of non-OECD countries. 2012 Edition Paris: IEA; 2012. ¨ rge-Vorsatz D, Eyre N, Graham P, Harvey D, Hertwich E, Jiang Y, et al. Chapter 10
[39] U Energy end-use: building. In: Global energy assessment
toward a sustainable future. Cambridge, UK: Cambridge University Press; 2012. [40] Zhang LX, Wang CB, Song B. Carbon emission reduction potential of a typical household biogas system in rural China. J Clean Prod 2013;47:415
21. [41] Wang CB, Zhang LX, Chang Y, Pang MY. Biomass direct-fired power generation system in China: an integrated energy, GHG emissions, and economic evaluation for Salix. Energy Policy 2015;84:155
65. [42] Chang Y, Huang RZ, Ries R, Masanet E. Life-cycle comparison of greenhouse gas emissions and water consumption for coal and shale gas fired power generation in China. Energy 2015;86:335
43. [43] Wang CB, Chang Y, Zhang LX, Pang MY, Yan H. A life-cycle comparison of the energy, environmental and economic impacts of coal versus wood pellets for generating heat in China. Energy 2017;120:374
84. [44] Huang BJ, Mauerhofer V. Life cycle sustainability assessment of ground source heat pump in Shanghai, China. J Clean Prod 2016;119:207
14. [45] Chang Y, Gu YR, Zhang LX, Wu CY, Liang L. Energy and environmental implications of using geothermal heat pumps in buildings: an example from north China. J Clean Prod 2017;167:484
92. [46] Allouhi A, Saadani R, Kousksou T, Saidur R, Jamil A, Rahmoune M. Grid-connected PV systems installed on institutional buildings: technology comparison, energy analysis and economic performance. Energy Build 2016;130:188
201. [47] Zhao XG, Zeng YP, Zhao D. Distributed solar photovoltaics in China: policies and economic performance. Energy 2015;88:572
83. [48] Wang CB, Chang Y, Zhang LX, Chen YS, Pang MY. Quantifying uncertainties in greenhouse gas accounting of biomass power generation in China: system boundary and parameters. Energy 2018;158:121
7. [49] Becque´ R, Weyl D, Stewart E, Mackres E, Jin LT, Shen XF. Accelerating building decarbonization: eight attainable policy pathways to net zero carbon buildings for all. Washington, DC: WRI World Resource Institute; 2019. [50] Schnitzer D, Lounsbury D, Carvallo J, Deshmukh R, Apt J, Kammen D. Microgrids for rural electrification: a critical review of best practices based on seven case studies. Washington, DC: United Nations Foundation; 2014. [51] Hayter S, Kandt A. Renewable energy applications for existing buildings. NREL Report No: NREL/CP-7A40-52172. Golden, CO: National Renewable Energy Lab; 2011. [52] IEA (International Energy Agency). The world energy outlook 2006 maps out a cleaner, cleverer and more competitive energy future, ,https://www.iea.org/news/the-worldenergy-outlook-2006-maps-out-a-cleaner-cleverer-and-more-competitive-energy-future.; 2006 [accessed 15.10.2019]. [53] IES (Integrated Environmental Solutions Limited). Catalyst building, ,https://www.iesve. com/software/case-studies/5943/catalyst-building.; 2019 [accessed 2.12.2019].

This page intentionally left blank

Chapter 10

Towards a renewable-energydriven district heating system: key technology, system design and integrated planning Yi Dou1, Lu Sun2, Minoru Fujii2, Yasunori Kikuchi1,3, Yuichiro Kanematsu1 and Jingzheng Ren4 1

Presidential Endowed Chair for “Platinum Society”, The University of Tokyo, Tokyo, Japan, Center for Social and Environmental Systems Research, National Institute for Environmental Studies (NIES), Tsukuba, Japan, 3Institute for Future Initiatives, The University of Tokyo, Tokyo, Japan, 4Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China 2

Chapter Outline 10.1 Introduction 311 10.2 Key technologies and system design for renewable-energydriven district heating 314 10.2.1 Indicators and design principle for enhancement of district heating systems 314 10.2.2 System design and key technologies of renewableenergy-driven district heating system 318

10.3 Integrated urban planning for renewable-energy-based district heating 10.3.1 Urban and industrial symbiosis 10.3.2 Modelling the strategic urban renewal for promoting district heating 10.4 Conclusions Acknowledgements References

324 324

326 328 329 329

10.1 Introduction Energy consumption is one of the key issues of human society, which is strongly associated with both resource depletion issue and climate change issue. Avoiding unnecessary energy use, shifting the usage of fossil energy to renewables, and improving the efficiency of energy use are three strategies Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00010-8 © 2021 Elsevier Inc. All rights reserved.

311

312

PART | III Applications

towards a low-carbon sustainable energy system. According to the statistics by International Energy Agency (IEA), the global energy consumption was keeping increasing from 6.26 million ktoe to 9.41 million ktoe during 1990
2015, of which residential and commercial sectors possess 29.4% of the total by 2015 [1]. In this fraction, energy use for hot water, space heating and cooling is generally over the half, followed by energy use for lighting and other purposes. Particularly energy demand for space heating and cooling is much higher in cold or hot regions, where district heating and cooling systems are playing a critical role for energy supply. District heating and cooling system itself is not a quite new technology but still in development in the world. The world first district heating system was adopted in France using geothermal heat since the 14th century, followed by the district heating network constructed in the United States and Canada [2,3]. The present-day district heating networks were popularized from Europe to the rest of the world from 1930s, until now the overall district heating market was valued at 163.76 billion USD in 2017 and is projected to achieve 203.00 billion USD by 2023 [4]. According to the report by IEA, the share of district network supplies in total final energy consumption has surpassed 15% in countries such as Denmark, Finland and Russia, while the share of renewable energy in district heating networks is already over 50% in Sweden and Denmark [5]. Nowadays, introducing district heating and cooling system becomes one pillar of the strategy for urban transition towards a low-carbon sustainable urban development. On the one hand, in high density residential and commercial area, district heating and cooling system has higher energy efficiency than individual system to support energy-saving activities. This is easy to understand if we consider the economy of scale to heat supply technologies. The more compact the heat supply becomes, the lower average cost of equipment installation can be achieved. Additionally concentrated heat supply can offset the real-time demand variation between users, so as to increase the total working ratio of heat supply system. Worldwide researches have reported the high energy efficiency and environmental benefits brought from district energy supply, that is, district heating and cooling is thought as one important measure to save regional fuel purchase cost with less greenhouse gases emission to the atmosphere [3]. On the other hand, district heating and cooling system provides a platform to introduce more local resources, which helps in forming an economic circle in the region. In the ancient time, people’s life is almost supported by local natural resources including local wood usage for space heating and boiled water. Until the modern industrialization, fossil fuels became the basic energy sources fed into boilers to generate heat and driving forces. Local society has to pay for the fuel cost if the fuels from another region have been imported, that means a great loss of wealth to the society, even causes geographic disparity of wealth between nations. With the recent technology

Towards a renewable-energy-driven district heating system Chapter | 10

313

FIGURE 10.1 General view of a district heating and cooling system.

innovation, district heating systems allow to accept much lower-temperature heat than before, which means more options of local resources, such as municipal general waste, biomass, geothermal and solar heat, are available. This brings hope for the less developed and resource dependent societies to realize energy security and self-sufficiency and finally contribute to local economic growth. As successful cases, ‘Stadtwerke’ is widely developed in Germany which utilizes public finance to support complete energy service including district heating in a city, while district heating networks are developed in Japan as important component of national policy called ‘produce locally, consume locally’. Similarly the well-known Heat Roadmap Europe also provides a comprehensive perspective of technology development and policy proposal for district heating [6,7]. Fig. 10.1 is a general view of the structure of district heating and cooling system. Although district heating and cooling system provides such merits to the socioeconomy, it is still not quite popularized in many countries, even many of existing projects are suffering in market recession. Using European countries as an example, the market share of district heating (the percentage of citizens who has access to district heating networks) in Iceland, Denmark, Finland and Sweden is higher than 50% by 2012, but it in the Netherlands, Norway, Germany and France is at 10% even less [8]. Particularly Denmark and Sweden report the best successful district heating systems in energy efficiency and emission reductions [9,10]. The same as a developed country, Japan also introduced district heating systems from 1970s, but so far there are still very few market shares belonging to district heating in urban heat supply system [11]. Except for the climatic factor, many other factors including the difference in urban morphology, competition between district and individual heating, even cultural and lifestyle changes bring barriers to the popularization of district heating systems. Furthermore, the coming global trend of depopulation and energy-saving activities are also great challenges for introducing district heating system.

314

PART | III Applications

Beyond the challenges, several opportunities appear to support the development of district heating and cooling systems. On the one hand, recent technology innovations let district heating systems allow more types of energy sources and supply much lower-temperature heat. Known as the 4GDH (4th Generation District Heating), new technologies in district heating effectively reduce the heat loss rate and system operation cost, so that biomass, solar heat, river heat, underground heat and other renewables become usable as energy sources input into the heating system. Additionally the application of seasonal heat storage will significantly enlarge the usage of waste heat from industries, especially the huge amount of cooling water [12,13]. This will further extend the feasibility of introducing district heating system in various situations and reduce the fuel purchase cost. On the other hand, market and policy trend becomes more familiar with renewable energies, so that the financial burden for infrastructure investment is quite reduced while more subsidies for renewables turn to valid for district heating system. For instance, heating market has been liberalized in Japan and many European countries; at the same time specific subsidies such as Renewable Heat Incentive applied in England become a policy trend to support renewableenergy-driven district heating system [14,15]. Comprehensively summarizing the challenges and opportunities for promoting the application of district heating and cooling, it is hard to say whether its market will meet an explosive growth or not, but obviously the application will be extended globally adjusting to the local conditions. The following sections try to summarize the technical development and support tools in urban energy planning which help in transition towards a renewableenergy-driven district heating system.

10.2 Key technologies and system design for renewable-energy-driven district heating Transition towards a renewable-energy-driven district heating system is not a new story but a continuous endeavour to enhance the technoeconomic competitiveness and environmental benefits of installing district heating system. Here are not only several technical innovations but also changes on the perspective and design principle of district heating systems.

10.2.1 Indicators and design principle for enhancement of district heating systems 10.2.1.1 Energy efficiency and exergy efficiency Conventionally energy efficiency is the first and most important indicator to judge if it is suitable to install a district heating system or not. Energy

Towards a renewable-energy-driven district heating system Chapter | 10

315

efficiency is usually defined as the division of used energy with the total energy input, so energy loss is the key factor to evaluate the system performance of district heating from an economic perspective. In a fossilenergy-based heating system, energy efficiency also means the environmental performance on resource saving and emission reductions. Energy loss in a district heating system can be divided into energy conversion loss in supply side, distribution loss to the users and exchange loss in user side. Energy conversion loss is dependent on the performance of selected energy conversion technology, while energy distribution loss is mainly dependent on the scale of pipeline networks and heat insulation performance. In case of hot water transmission, not only energy is lost due to the heat conduction from pipelines to the outside but also it is lost by additional energy use due to the water pumping for keeping the pressure during long distance transmission. In case of steam supply, water pumping can be sometimes omitted, but pressure decrease will go faster than temperature decrease. When the pressure of steam decreases to the saturated vapour pressure, steam will turn to water status that loses the latent heat which is much larger than sensible heat. Heat exchange loss in user side is also critical because lower heat exchange efficiency will require higher temperature heat input and faster hot water supply. In fact, the defined energy efficiency is not so simply representing the actual performance of a district heating system. Companies usually report a rated energy efficiency assuming that heat load is at a certain level and keep stable. Actually the heat load is changing anytime in a day and quite different between seasons, this is taken into consideration by evaluating yearround energy efficiency. In comparison to individual heating systems, the rated energy efficiency of district heating is usually higher because of the economy of scale and heat load levelling by concentrated heat supply. However, heat load in summer becomes much lower than it in winter that sometimes makes individual heat supply to perform more efficiently than district heat supply. This means not only energy loss but also annual equipment operation rate are critical factors. Exergy efficiency is a similar but more specific indicator, that is often applied to measure the usefulness of a substance or energy efficiency of an energy system. It is not only possible to track the exergy flow of an energy system to find out the key process with largest exergy loss but also solve the optimization problems such as the best temperature of heat recovery and supply. This indicator has been well applied in the cases of fossil-fuel-based or geothermal-heat-based district heating systems [16
18]. In business perspective, another experiential indicator called linear heat density is well adopted for roughly evaluating the feasibility of introducing district heating systems. Defined as Eq. (10.1), linear heat density (Dl ) is the division of total heat sales (Qs ) by total pipeline length (L). This division can

316

PART | III Applications

be further decomposed into four fractions as below [19]. Dl 5

Qs P AB Q s AL 5    AL P AB L L

ð10:1Þ

Here, P is the population in an area, AL is the land area, and AB is the total building floor area. Obviously these four fractions represent the area’s population density, intensity of floor area use (per person), intensity of energy use by floor area, and pipeline network design. Interestingly this decomposition supports a demand-side perspective that intensive land use with well network design will essentially improve the feasibility of introducing district heating systems.

10.2.1.2 Cascade and upgrade use of heat energy Conversional district heating systems usually combust fossil fuels to generate heat for supply; however, fossil fuels are creating high-quality heat energy which satisfies the requirement for power generation and industrial use. Thus using fossil fuels directly for district heating means a large exergy loss. Cascade use of high-quality heat energy can maximize the total value recovered from heat use that optimizes the whole district heating system. As shown in Fig. 10.2, fossil fuels can be first used in a turbine to generate high-temperature and high-pressure steam for power generation. Then, the lower-temperature heat extracted from turbine can be used in industries while the further lowtemperature heat exhausted from chimney and condenser can be recovered for civil heat use through district heating. Similarly the cold energy extracted from liquified natural gas (LNG) also can be first used for power generation, then sent to district cooling system as a kind of waste heat. Except for cascade use, heat energy is possible to be upgraded in the case where higher-quality heat is required. For example, municipal general waste with low heat value cannot generate electric power in high efficiency, but if mixed with fossil fuels, the average quality will be adjusted to exactly fit the requirement of high-efficiency power generation. Furthermore much lowerquality heat, such as solar heat, underground heat, and river heat, is also able to be upgraded for industrial and civil heat use, or directly used for district heating if heat exchange efficiency is high enough in user side [20]. Learnt from the principle of cascade and upgrade use of heat energy, it is clear that a higher-efficiency district heating system should transit from fossilenergy-based system towards renewable-energy-based system not only for resource conservation and emission reductions but also for enhancing the overall efficiency of urban energy system. Since renewable and natural heat energies are usual in low temperature, such a supportive district heating system is also known as low-temperature district heating system. In addition, the principle also requires district heating system to be a multienergy system, in which various energy sources can be mixed for real-time adjusting the energy quality and quantity.

FIGURE 10.2 Cascade and upgrade use of heat energy.

318

PART | III Applications

10.2.2 System design and key technologies of renewable-energydriven district heating system 10.2.2.1 System composition of a renewable-energy-driven district heating system Section 10.2.1 has declared the principles of designing a high-efficiency district heating system including cascade and upgrade heat use, as well as complementary energy resource input. These principles determine the system composition for high-efficiency district heating should become more open source and spatiotemporally flexible. Fig. 10.3 summarized the trend of district heating system composition based on the concept of the fourth- and the fifth-generation district heating [6,12,21,22]. At the first-generation district heating system, fossil fuels are directly combusted in a boiler to generate steam or high-temperature hot water because the heat insulation of pipelines and heat exchange efficiency in user side are quite low. For enhancing the energy efficiency in supply side, Combined Heat and Power generation (CHP) is introduced to partly or totally substitute the boiler in the secondgeneration district heating systems, through which the fossil fuels are combusted for both heat and power generation. Here, the key of improving energy system efficiency is to make use of the excessive heat from the turbine, while the power generation rate is just around 50%. Until the thirdgeneration district heating, higher efficiency in heat distribution and exchange allows hot water as the medium, so that heat recovered from municipal waste and industrial process as well as renewable heat from biomass and solar energy can be joint into the system to partly substitute the use of fossil fuels. Most of the current district heating systems are based on these generation system design. With the introduction of advanced heat insulation materials and heat exchangers, recent heat distribution system can allow lower-temperature hot water for district heating. Therefore, advanced district heating projects in Europe have begun the demonstration of the fourth-generation district heating (4GDH). On the one hand, low-temperature heat sources such as geothermal and river heat are joint into the system, while seasonal heat storage is installed to realize an optimal technical combination and annually adjusted operation schedule. On the other hand, renewable energies like wind power and solar photovoltaic energy are introduced as electricity input to centralized heat pumps for heat supply. Since heat pumps are using electricity from renewables for driving the refrigerant to transfer natural heat, the overall energy efficiency will be improved a lot with significant reductions in fossil fuel usage and greenhouse gas emissions. However, the efficiency of heat pumps is not always positively correlated to heat supply scale while the unit heat consumption of buildings is decreasing fast, and thus many studies turn to promote distributed heat pumps as called the fifth-generation district heating system.

FIGURE 10.3 Five generation of district heating system.

320

PART | III Applications

10.2.2.2 Key technologies for a renewable-energy-driven district heating system According to the definition of five generation district heating systems, several technologies are indispensable for the transition to renewable-energydriven district heating systems. 10.2.2.2.1 Energy conversion The first technology, boiler is the most common part of a district heating system, through which fossil fuels or biomass are combusted to generate superheated steam. Here biomass can be input in a boiler as straw, chips, waste woods, biofuels or biogas. Due to the latest technology, a boiler can reach the highest thermal efficiency at 97% while it is convenient to be installed anywhere with any size for heat supply. It is also easy to turn on/off a boiler and regularly maintain the station, even realize an unattended heat supply station. However, in case of directly using superheated steam for district heating, its advantage on resource conservation is weaker than combined heat and power generation, while in case of low-temperature district heating, its total energy efficiency is much lower than heat pump using renewable energy. The second technology, combined heat and power generation (CHP) is currently one of the mainstream technologies in district heating projects. It combusts fossil fuels or biomass to generate high-pressure high-temperature steam for power generation in the turbine; then excessive heat from the turbine and chimney can be recovered for heat supply. Through this process, the overall energy efficiency can over 80% including 40%
50% for power generation and 30%
40% for heat use. Compared to heat-only heating system, CHP system can sell both heat and electricity, where the price of electricity is much higher than heat supply. Furthermore the excessive heat can also be used to provide cooling water through a set of absorption chillers, that is called Combined Cooling, Heating and Power generation (CCHP) [23,24]. However, the installation cost of CHP is much higher than it of boiler, and the same is the cost of control and maintenance. In addition, the greenhouse gas emission reductions are quite limited compared to heat pump technology. The third technology, heat pump is a desired power-to-heat technology which has a great potential of environmental benefits using renewable energies. It uses electricity for driving the continuous cycle of heat exchange through refrigerant to heating and cooling system. Thus its energy efficiency [often represented by coefficient of performance (COP)] can over 5.0 in case of civil use, and 2.0 in case of industrial use. Heat pumps are easy to be controlled due to the variation of heat load and electricity price, so that the required capacity of heat/electricity storage can be reduced a lot than the other heat supply technologies. However, the COP of heat pump is quite related to the environmental conditions, the scale of heating and cooling project, temperature requirement in user side and the type of refrigerant, decision to install a

Towards a renewable-energy-driven district heating system Chapter | 10

321

heat-pump-based district heating system should be very careful. In this case, optimization for the whole energy system including power and heat supply with maximum use of local renewables is recommended [25,26]. Renewable heat sources, such as solar energy and geothermal energy, are also available for district heating through heat exchanger. However, these are not yet popularized because of the lower temperature of heat and limited efficiency of heat exchange. For larger scale district heating projects, such renewable heat sources usually cannot support enough heat quantity, meanwhile the installation cost is still expensive [27]. Heat pumps using electricity from renewables mentioned above are one kind of upgrade heat use, in case batteries can be used for power storage instead of heat storage. As another option, Hydrogen generated from excessive renewable energies is possible to substitute fossil fuels in CHP-based district heating, at the same time Hydrogen is also a kind of energy storage, but the overall energy efficiency of this case is also limited compared to heat-pump-based district heating. 10.2.2.2.2

Heat distribution

In most of the situations, a district heating system deliveries steam or hot water through a network of insulated pipelines. Pipes can be laid aboveground or underground, while the latter is much more expensive than the former. Since the investment on pipeline network usually possesses 1/3 of the total costs of district heating projects, there are some pilot projects using phase-change materials (PCMs) as mobile energy storage which can effectively save investment cost. These materials, including organic compounds, salt hydrates and eutectic mixtures, can serve heating temperature from 0 C to 250 C that becomes a potential solution to replace conventional pipeline networks [28,29]. 10.2.2.2.3

Heat storage

Hourly variable heat load leads to a certain capacity of heat storage for realtime matching heat supply and demand. It is common to introduce a diurnal heat storage using hot water to adjust the real-time flow rate, while installing a water-based pit seasonal storage for levelling the seasonal changes of heat load. Except for water, a seasonal storage can also use rock beds or ground soil, as well as PCMs mentioned before, but the critical problems in popularizing seasonal heat storage are the expensive investment and strict regulation [30,31].

10.2.3 Optimization for a renewable-energy-driven district heating system Optimization for maximum cost efficiency is a critical problem in designing district heating system. When argue the feasibility of installing a district heating system, cost efficiency comes out as the first and most important

322

PART | III Applications

indicator to be questioned. Generally optimization can be achieved from both supply side and demand side.

10.2.3.1 Supply side optimization Supply side optimization is a common perspective in designing any energy system. It answers two questions: what kind of technology combination can satisfy the energy demand with minimum investment, and how to improve the operation schedule to achieve minimum operation costs. The most applied method in previous studies is linear programming, of which a common form of objective function is shown in Eq. (10.2) [32,33]. X X X X Sðs; tÞ 1 cbj ηij 1 ccj λij 1 cT μ 1 cwj ϕij ðs; tÞ ð10:2Þ mincF s;t

i;j

i;j

i;j;s;t

The first item represents annual fuel purchase cost, where S(s,t) represents the annual energy input in district heating and cF is the price of each type of fuels. The second item represents the initial fixed investment, where ηij is a binary variable to define whether equipment j is installed in location i or not, cbj is the investment cost of equipment j. The third item represents the variable costs, where λij is the adjustable capacity of equipment j in location i, ccj is the cost due to the capacity of equipment j. The fourth item represents the cost for energy storage, where μ is the capacity of energy storage and cT is the cost due to its capacity. The final item represents the operation costs, where ϕij ðs; tÞ is a binary variable to reveal the status of turning on or off the equipment j in location i, and cwj is its operation cost due to working schedule. Through the solution of this equation, it is possible to know the optimal combination of technologies and the best operation schedule of each equipment for minimizing the total costs of district heating system. After joining with renewable energies, especially combined power generation from renewables, district heating system becomes one part of the local multienergy system where each energy unit can be both an energy supplier and a user at the same time acting as an energy hub. For the best design of such multienergy systems, researchers developed more complex optimization method with a wider consideration in technology selection and evaluation criteria [34,35]. Recently many supporting tools are developed for automatically solving the optimization of multienergy systems in various time resolution and scale, such as the software EnergyPRO for detailed simulation of multienergy supply, and EnergyPLAN for aggregated optimization of a board energy systems cross sectors in a region [36,37]. These measures are also joined into the concept of low-temperature or low-energy district heating networks which are specific for the cases in low-density cities [38
42]. Mathematic optimization models combined with Geographic Information System (GIS) is also a tendency. Taking geographic characters into consideration, supporting tools of GIS can not only hold an inventory of potential

Towards a renewable-energy-driven district heating system Chapter | 10

323

heat sources and sinks and draw the detailed location through heat atlas [43
46] but also help in optimizing network design joint with linear programming model and technoeconomic analysis [47,48]. Some technical improvements, such as introducing twin pipelines, layout of a T-connection network, and smart flow rate control, are also studied to further increase the efficiency of district heating system [49
52]. However, without the coordination from demand side, the improvements will be quite limited.

10.2.3.2 Demand-side management Optimization on equipment installation and operation schedule is relatively an easy way for matching energy supply with variable demand in maximum cost efficiency. By contrast, demand-side management supports another perspective to optimize the district heating system. 10.2.3.2.1

Demand response

Application of demand response in district heating system at first needs the popularization of smart metre to measure real-time heat consumption. However, unlike the grid electricity, usually half or one hour is expended to reflect the flow changes from supply side to demand side. Thus prediction of hourly heat load is quite important to harmonize the variation between supply and demand. There are two approaches to manage the demand variation. One is based on Information and Communication Technologies (ICT) to inform the real-time information to the users, such as real-time price variation and energysaving result, for stimulating the response from users’ behaviour. The other one is based on Internet of Things (IoT) to directly install collected energy use equipment in user side so that energy consumption will be real-time managed by energy control centre. Particularly in the case of using solar heat or realtime priced electricity, supply side has to promptly transfer the variation information to users and require coordinated action to keep system stable [53,54]. 10.2.3.2.2 Building mix Because daily heat load curves are quite different due to the special purpose of buildings, building clusters with various purposes can partly offset the heat load variation. According to the people’s lifestyle, heat load of residential house is usually high in the night but low in the daytime. By contrast, it of workplace and commercial buildings is usually low in the night but high in the daytime. Similar with demand response, the levelled heat load can help in reducing the scale of heat supply and storage equipment [55,56]. 10.2.3.2.3

Land use change

Land use change is not an often-discussed issue in designing district heating system because the system design is usually based on the current situation where the requirement of pay-pack period is strictly limited within 20 years.

324

PART | III Applications

As one unexpected fact, the recent depopulation trend and energy-saving behaviours in buildings may reduce around half of the current energy demand in the future. Furthermore this path may be speeded up, corresponding to the popularization of net-zero-energy buildings (ZEBs). However, people have already begun the transition towards compact cities, in which citizens are encouraged to move back from suburbs to city centre while city centre is redeveloped to avoid unoccupied houses and come back to the previous prosperity. Guided land use changes can support all the measures mentioned above to enhance the competitiveness of district heating. During building renewable, smart building energy management system with building-based renewable energies is much easier for installation; meanwhile the trendy design of urban complex in city centre also helps in promoting building mix to level the regional heat load. Particularly compact city planning can significantly enhance the linear heat density that makes district heating system more efficient and easier to connect to renewables. The next section will emphatically discuss an integrated planning method considering all the factors mention above.

10.3 Integrated urban planning for renewable-energy-based district heating As discussed above, transition towards a renewable-energy-based district heating system not only means to horizontally extend the system boundary to multi sectors, but also require a long-term energy planning strategy to control the uncertainties. At building or neighbourhood level, real-time adjustment of heat load and supply is the key solution for optimal system design and operation schedule. By contrast, at city or region scale, long-term land use design becomes the most important factor for optimizing district heating system. Through the collaboration with urban designers and urban planning section of local government, district heating systems are expected to keep competitiveness in the long term while maximally make use of local renewables (Fig. 10.4). This section will introduce the recent practice in joining urban planning with district heating systems including the perspective from urban and industrial symbiosis, as well as integrated urban renewal strategy.

10.3.1 Urban and industrial symbiosis Industrial Symbiosis is currently a well-discussed concept in developing ecoindustries, which aims at maximizing the resources conservation and emission reductions through exchange of by-products between industries [57]. Compared to the similar concept like Circular Economy, industrial symbiosis is specific on designing the material/energy flow between industries from the perspectives of system engineering. Accordingly heat exchange and its cascade use become an important issue in industrial symbiosis, such as steam extracted from the turbine or condenser of thermal power

Towards a renewable-energy-driven district heating system Chapter | 10

325

plant to factories, steam extracted from incinerator to factories, heat exchange between factories, and cold heat from liquified natural gas. One of the well-known practice, Kalundborg City of Denmark established a pipeline network to exchange waste heat between power plant and industries [58,59], while another famous case, located in Ulsan City of South Korea, demonstrated a board pipeline network for waste heat exchange between incinerators and nearby industries [60,61]. Despite the investment on pipes is not a negligible amount, the actual pay-back is quite short even less than 6 months with substantial revenue. Recently the boundary of industrial symbiosis has been extended to cover urban areas called urban symbiosis, aiming to realize a larger scale of material and energy exchange. In this scope, sectors including industrial parks, urban area, agriculture and forestry are organically joined to exchange by-products as much as possible for maximum resource conservation and emission reductions. Regarding the heat supply issue, urban symbiosis particularly focuses on appropriate application of waste-to-energy technologies and district heating systems using waste heat and local renewables

FIGURE 10.4 Demand-side management as a key for promoting district energy system.

FIGURE 10.5 District heating system in urban and industrial symbiosis.

326

PART | III Applications

(Fig. 10.5). As supplements to the knowledge in the field of conventional energy engineering, urban and industrial symbiosis especially emphasizes on the process analysis of heat sources and cross-sector spatial design of district heating system through material/energy flow analysis [62
66].

10.3.2 Modelling the strategic urban renewal for promoting district heating According to the previous research and practice mentioned above, the authors try to model the strategic urban renewal joining with the social demonstration of next-generation district heating technologies, where industrial location changes and urban land use design are taken into consideration. As one previous case study, the authors proposed a waste-heat-based district heating system in Shinchi Town of Fukushima Prefecture, Japan. In this case, waste heat from thermal power plant and factories is proposed as heat sources for the industrial park and nearby urban and agricultural area. At first, waste heat exchange between thermal power plant and factories in industrial park is indicated feasible with a considerable potential on fossil fuel saving and emission reductions in current land use allocation. However, this potential will be greatly affected by land use changes according to the local government’s industrial location policies. If expansion of energy-intensive industries becomes the target of industrial location policy, increased heat demand can support a sufficient usage of waste heat sources. Oppositely although employee-intensive industries can bring more jobs, the decreased heat demand will destroy the feasibility of district heating using waste heat [67,68]. Furthermore in the case of waste heat usage for households and plant factory, it is clear that directly installing district heating system (business-asusual scenario, BAU) is not economic than individual heating system (baseline) because of the dispersed building distribution, but it will be feasible with significantly improved emission reduction effect if design the residential area in a compact pattern with supplementary heat consumption from a proposed plant factory [69]. Learnt from the changes of energy flow with land use scenarios shown in Fig. 10.6, the key point for enhancing district heating system in this case is the realization on sufficient usage of waste heat with a minimum scale of pipeline network. In order to popularize the smart joint district heating system practiced in Shinchi Town to the whole Soma Region in the north coastal area of Fukushima Prefecture, Japan, the authors discussed possible impacts from regional urban renewal strategy on the potential of introducing district heating system. Building distribution changes in past decades are collected to support urban renewal simulation through a method called 4-dimensional Geographic Information System (4d-GIS). Six factors are taken into the simulation of urban renewal and district heating network expansion: (1) current depopulation tendency in the region; (2) part of potential migrants in the

FIGURE 10.6 A case study on land use adjustment for enhancing waste-heat-based district heating network in Fukushima Prefecture, Japan.

328

PART | III Applications

suburbs come back to city centre that aims to satisfy the minimum plot ratio for district heating network expansion; (3) longer lifespan of buildings that slows down the path of urban renewal; (4) heat load reductions due to the increasing proportion of ZEBs in newly built buildings; (5) improved district heating technologies due to the boom of the 4GDH system; (6) energy price and unit CO2 emission changes in power and heating market. Despite the introduction of district heating system is currently doubted in Soma Region because of the low population density, results from the research indicated the feasibility of introducing district heating systems considering interaction between various sectors. In detailed, although the trend of depopulation and ZEBs will substantially reduce the regional total heat load, compact urban planning is possible to maintain the minimum plot ratio for expanding district heating network. However, this effect is quite variable due to geographic characters and other factors. For example, not the dense or dispersed cities but the medium-density cities are with the largest potential of promoting district heating because of the balance between heat density and pipeline construction cost. On the other hand, longer lifespan of buildings will slow down the urban renewal towards compact city, but increasing energy price may cover this negative effect. Finally CHP-based district heating is indicated gradually lose the competitiveness with renewable-energybased individual heating, but this can be opposite if district heating transits towards renewable-energy-based system. For more details, please refer to the authors’ previous publications [70].

10.4 Conclusions In general, district heating and cooling systems are recently performing more efficiently than individual heating and cooling systems, but without transition to renewable-energy-based systems, their technoeconomic competitiveness and environmental contribution will be doubted. To speed up this transition, on the one hand, it is necessary to continuously invest in technology development including high-efficiency boilers, heat pumps and seasonal heat storage; meanwhile try the best to reduce the installation cost of equipment and infrastructure. In addition, more flexible heat distribution like nonpipelinebased heat transmission can be considered in case. On the other hand, both supply side and demand side should be joint into the management system to optimize equipment composition and operation schedule. In supply side, support tools such as linear programming can be applied to optimize the system composition based on the technoeconomic analysis on technologies and the inventory of local renewables endowment. In demand side, demand response can shift the peak of heat load while building mix can help in levelling the daily variation of heat load to adjust with heat supply cost. In the long term, guided urban renewal towards compact cities is expected to keep the performance of district heating and cooling by a certain heat load density.

Towards a renewable-energy-driven district heating system Chapter | 10

329

Through the connection to renewable energies, district heating and cooling systems obtain more chances of policy and financial supports. In direct way, incentives for waste disposal and biomass utilization and subsidies for energy efficiency improvements are available. In indirect way, the electricity generated from renewables, such as solar photovoltaic and wind energy, is also subsidized during a long time. Furthermore, savings from the carbon tax probably introduced in the future can become another solid support, while this advantage will be strengthened by additional merit from environmentoriented green finance. The proposed renewable-energy-based district heating system involves various stakeholders from agriculture, industry, residential, commercial and public sectors that makes the progression of project implementation more complicated and time consumed. Therefore the third person view focusing on overall optimization is required. Urban designer can be the most appropriate position for coordination, while positive participation from citizens is absolutely necessary in negotiation. Comprehensively considering these factors, social enterprises based on public
private partnership could be the best business model to promote renewable-energy-based district heating and cooling systems.

Acknowledgements The financial support from the JSPS KAKENHI (19K23530) and the Environment Research and Technology Development Fund (3
1709, 3
1905) of Japan was gratefully acknowledged.

References [1] IEA. World energy balances (2019 edition). International Energy Agency (IEA), ,https:// www.iea.org/subscribe-to-data-services/world-energy-balances-and-statistics.; 2019. [2] Rezaie B, Rosen MA. District heating and cooling: Review of technology and potential enhancements. Appl Energy 2012;93:2
10. [3] Lake A, Rezaie B, Beyerlein S. Review of district heating and cooling systems for a sustainable future. Renew Sustain Energy Rev 2017;67:417
25. [4] MarketsandMarkets. District heating market by heat source (coal, natural gas, renewable, oil & petroleum products), plant type (boiler plant, chp), application (residential, commercial, industrial), and geography - Global Forecast to 2023. Market Research Report; 2018. [5] IEA. Share of renewable energy in district heating networks; 2018. [6] Persson U, Mo¨ller B, Werner S. Heat roadmap Europe: identifying strategic heat synergy regions. Energy Policy 2014;74:663
81. [7] Connolly D, et al. Heat roadmap Europe: combining district heating with heat savings to decarbonise the EU energy system. Energy Policy 2014;65:475
89. [8] Colmenar-Santos A, et al. District heating and cogeneration in the EU-28: current situation, potential and proposed energy strategy for its generalisation. Renew Sustain Energy Rev 2016;62:621
39.

330

PART | III Applications

[9] Di Lucia L, Ericsson K. Low-carbon district heating in Sweden
examining a successful energy transition. Energy Res Soc Sci 2014;4:10
20. [10] Chittum A, Østergaard PA. How Danish communal heat planning empowers municipalities and benefits individual consumers. Energy Policy 2014;74:465
74. [11] Werner S. International review of district heating and cooling. Energy 2017;137:617
31. [12] Lund H, et al. 4th generation district heating (4GDH): Integrating smart thermal grids into future sustainable energy systems. Energy 2014;68:1
11. [13] Lund H, et al. The status of 4th generation district heating: research and results. Energy 2018;164:147
59. [14] Connor PM, et al. The development of renewable heating policy in the United Kingdom. Renew Energy 2015;75:733
44. [15] Connor P, et al. Devising renewable heat policy: overview of support options. Energy Policy 2013;59:3
16. [16] Kec¸eba¸s A. Energetic, exergetic, economic and environmental evaluations of geothermal district heating systems: an application. Energy Convers Manag 2013;65:546
56. [17] Terhan M, Comakli K. Energy and exergy analyses of natural gas-fired boilers in a district heating system. Appl Therm Eng 2017;121:380
7. [18] Hepbasli A. A key review on exergetic analysis and assessment of renewable energy resources for a sustainable future. Renew Sustain Energy Rev 2008;12(3):593
661. [19] Persson U, Werner S. Heat distribution and the future competitiveness of district heating. Appl Energy 2011;88(3):568
76. [20] Fujii M, et al. Contribution to a low-carbon society from improving exergy of waste-to-energy system by upgrading utilization of waste. Resources, Conserv Recycling 2019;149:586
94. [21] Paiho S, Reda F. Towards next generation district heating in Finland. Renew Sustain Energy Rev 2016;65:915
24. [22] Ziemele J, et al. Combining energy efficiency at source and at consumer to reach 4th generation district heating: economic and system dynamics analysis. Energy 2017;137:595
606. [23] Li Y, Fu L, Zhang S. Technology application of district heating system with cogeneration based on absorption heat exchange. Energy 2015;90:663
70. [24] Liu M, Shi Y, Fang F. Combined cooling, heating and power systems: a survey. Renew Sustain Energy Rev 2014;35:1
22. [25] Averfalk H, et al. Large heat pumps in Swedish district heating systems. Renew Sustain Energy Rev 2017;79:1275
84. [26] Sayegh MA, et al. Trends of European research and development in district heating technologies. Renew Sustain Energy Rev 2017;68:1183
92. [27] Brand M, Svendsen S. Renewable-based low-temperature district heating for existing buildings in various stages of refurbishment. Energy 2013;62:311
19. [28] Pereira da Cunha J, Eames P. Thermal energy storage for low and medium temperature applications using phase change materials
a review. Appl Energy 2016;177:227
38. [29] Chiu JN, et al. Industrial surplus heat transportation for use in district heating. Energy 2016;110:139
47. [30] Xu J, Wang RZ, Li Y. A review of available technologies for seasonal thermal energy storage. Sol Energy 2014;103:610
38. [31] Li Y, Rezgui Y, Zhu H. District heating and cooling optimization and enhancement
towards integration of renewables, storage and smart grid. Renew Sustain Energy Rev 2017;72:281
94. [32] Togawa T, et al., Design support framework for distributed energy system considering regional characteristics. J Jpn Soc Civil Eng, Ser G (Environ Res) 2015;71(6):II_139
II_149.

Towards a renewable-energy-driven district heating system Chapter | 10

331

[33] Togawa T, et al., Desing and standards for distributed energy system in the intermountainous area. J Jpn Soc Civil Eng, Ser G (Environ Res) 2017;73(5):I_107
I_119. [34] Mancarella P. MES (multi-energy systems): an overview of concepts and evaluation models. Energy 2014;65:1
17. [35] Bazmi AA, Zahedi G. Sustainable energy systems: role of optimization modeling techniques in power generation and supply-A review. Renew & Sustain Energy Rev 2011;15 (8):3480
500. [36] Connolly D, et al. A review of computer tools for analysing the integration of renewable energy into various energy systems. Appl Energy 2010;87(4):1059
82. [37] Allegrini J, et al. A review of modelling approaches and tools for the simulation of district-scale energy systems. Renew & Sustain Energy Rev 2015;52:1391
404. [38] Broberg S, et al. Industrial excess heat deliveries to Swedish district heating networks: drop it like it’s hot. Energy Policy 2012;51:332
9. [39] Sun FT, et al. A new waste heat district heating system with combined heat and power (CHP) based on ejector heat exchangers and absorption heat pumps. Energy 2014;69:516
24. [40] Kapil A, et al. Process integration of low grade heat in process industry with district heating networks. Energy 2012;44(1):11
19. [41] Li HW, Svendsen S. Energy and exergy analysis of low temperature district heating network. Energy 2012;45(1):237
46. [42] Ostergaard PA, Lund H. A renewable energy system in Frederikshavn using lowtemperature geothermal energy for district heating. Appl Energy 2011;88(2):479
87. [43] Finney KN, et al. Developments to an existing city-wide district energy network - part I: identification of potential expansions using heat mapping. Energy Convers Manag 2012;62:165
75. [44] Finney KN, et al. Developments to an existing city-wide district energy network: part II analysis of environmental and economic impacts. Energy Convers Manag 2012;62:176
84. [45] Yeo IA, Yoon SH, Yee JJ. Development of an environment and energy Geographical Information System (E-GIS) construction model to support environmentally friendly urban planning. Appl Energy 2013;104:723
39. [46] Gils HC, et al. GIS-based assessment of the district heating potential in the USA. Energy 2013;58:318
29. [47] Nielsen S, Moeller B. GIS based analysis of future district heating potential in Denmark. Energy 2013;57:458
68. [48] Girardin L, et al. EnerGis: a geographical information based system for the evaluation of integrated energy conversion systems in urban areas. Energy 2010;35(2):830
40. [49] Chae SH, et al. Optimization of a waste heat utilization network in an eco-industrial park. Appl Energy 2010;87(6):1978
88. [50] Tol HI, Svendsen S. Improving the dimensioning of piping networks and network layouts in low-energy district heating systems connected to low-energy buildings: a case study in Roskilde, Denmark. Energy 2012;38(1):276
90. [51] Dalla Rosa A, Christensen JE. Low-energy district heating in energy-efficient building areas. Energy 2011;36(12):6890
9. [52] Dalla Rosa A, et al. District heating (DH) network design and operation toward a systemwide methodology for optimizing renewable energy solutions (SMORES) in Canada: a case study. Energy 2012;45(1):960
74. [53] Wang J, et al. Review and prospect of integrated demand response in the multi-energy system. Appl Energy 2017;202:772
82.

332

PART | III Applications

[54] O’Connell N, et al. Benefits and challenges of electrical demand response: a critical review. Renew Sustain Energy Rev 2014;39:686
99. [55] Chow TT, Chan ALS, Song CL. Building-mix optimization in district cooling system implementation. Appl Energy 2004;77(1):1
13. [56] Best RE, Flager F, Lepech MD. Modeling and optimization of building mix and energy supply technology for urban districts. Appl Energy 2015;159:161
77. [57] Chertow M, Ehrenfeld J. Organizing self-organizing systems. J Ind Ecol 2012;16(1):13
27. [58] Jacobsen NB. Industrial symbiosis in Kalundborg, Denmark - a quantitative assessment of economic and environmental aspects. J Ind Ecol 2006;10(1
2):239
55. [59] Domenech T, Davies M. Structure and morphology of industrial symbiosis networks: the case of Kalundborg. 4th 5th Uk Soc Netw Conf 2011;10:79
89. [60] Park JY, Park HS. Securing a competitive advantage through industrial symbiosis development the case of steam networking practices in Ulsan. J Ind Ecol 2014;18(5):677
83. [61] Park JM, Park JY, Park HS. A review of the national eco-industrial park development program in Korea: progress and achievements in the first phase, 2005
2010. J Clean Production 2016;114:33
44. [62] Geng Y, et al. Recent progress on innovative eco-industrial development. J Clean Prod 2016;114:1
10. [63] Fujii M, et al. Possibility of developing low-carbon industries through urban symbiosis in Asian cities. J Clean Prod 2016;114:376
86. [64] Ohnishi S, et al. Comparative analysis of recycling industry development in Japan following the Eco-Town program for eco-industrial development. J Clean Prod 2016;114:95
102. [65] Dou Y, et al. Feasibility of developing heat exchange network between incineration facilities and industries in cities: case of Tokyo Metropolitan Area. J Clean Prod 2018;170:548
58. [66] Sun L, et al. Energy-saving and carbon emission reduction effect of urban-industrial symbiosis implementation with feasibility analysis in the city. Technol Forecast Soc Change 2020;151:119853. [67] Dou Y, et al. Potential of waste heat exchange considering industrial location changes: a case of Shinchi-Soma region in Fukushima, Japan. J Jpn Soc Civil Eng, Ser. G (Environ. Res.) 2017;73(6):II_353
II_363. [68] Togawa T, et al. Feasibility assessment of the use of power plant-sourced waste heat for plant factory heating considering spatial configuration. J Clean Prod 2014;81:60
9. [69] Dou Y, et al. Innovative planning and evaluation system for district heating using waste heat considering spatial configuration: a case in Fukushima, Japan. Resour, Conserv Recycl 2018;128:406
16. [70] Dou Y, et al. Proliferation of district heating using local energy resources through strategic building-stock management: a case study in Fukushima, Japan. Front. Energy 2018;12 (3):411
25.

Chapter 11

Renewable energy-driven desalination for more water and less carbon Aamir Mehmood and Jingzheng Ren Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China

Chapter Outline 11.1 Introduction 333 11.2 Desalination technology 336 11.2.1 Thermal desalination techniques 337 11.2.2 Membrane desalination techniques 339 11.2.3 Desalination installed capacity and trends 340 11.3 Energy and desalination 345 11.3.1 Renewable energy resources for desalination 345 11.4 Renewable energy integrated desalination: technical, economic and social development aspects 347 11.4.1 Solar desalination 347 11.4.2 Nuclear energy-driven desalination 350

11.4.3 Wind energy-driven desalination 353 11.4.4 Geothermal energy-driven desalination 354 11.4.5 Ocean/wave energy-driven desalination 357 11.5 Barriers, issues and opportunities in desalination technology development 357 11.5.1 Brine production 359 11.5.2 Desalination cost and CO2 emissions 359 11.6 Outlook 361 Abbreviations 361 References 362

11.1 Introduction The whole world is struggling for 17 global goals set on the 2030 agenda for sustainable development [1,2]. SDG-15 is about ‘life on land’, for which the mankind is in attempt of making its existence on earth planet sustainable for the first day. The future sustainability is underpinning on ‘Energy
Water
Environment’ nexus [3
5]. This nexus directly contributes Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00011-X © 2021 Elsevier Inc. All rights reserved.

333

334

PART | III Applications

to three sustainable development goals (SDGs): goal-6 (clean water and sanitation), goal-7 (affordable and clean energy) and goal-13 (climate action) [2]. Water and energy are independent variables but are closely interlinked and environment is indirectly dependent on both. Water is an abundant element of the nexus that covers B71% of the earth planet surface, equivalent to around 326 million trillion gallons of water [6,7]. According to WHO, water with total dissolved solids value less than 500 ppm is drinkable or up to maximum 1000 ppm in certain cases [8]. Only 3% of the total water content is available as fresh water, around 66% (2/3 parts) of which exists in the form of glaciers and caps [7]. Total global water distribution in the form of saltwater and fresh water mix by source on earth surface is portrayed in Fig. 11.1. The available one-third portion of the total fresh water content is being polluted and wasted due to improper consumption at industrial, domestic and agriculture level activities [7,11]. 5.3 billion people (i.e. B71% of global population) have safe access to uncontaminated drinking water. While 2.2 billion people do not have safely managed drinking water services [12]: G G G G G

1.4 billion count has access to basic level services. 206 million people have limited access. 435 million people take water from unprotected springs and wells. 144 million drink water collected from lakes, ponds, rivers and streams. Even 2 billion people are drinking faeces contaminated water.

FIGURE 11.1 Global water distribution and water mix on earth planet by source [9,10].

Renewable energy-driven desalination Chapter | 11

335

In-short, 41% of the population count is facing water stress, and by 2025 this number will increase to the half of total count [12,13] and will become more intense by 2040 as reported by World Economic Forum [14]. In such a scary situation when water is not a problem of any certain region but is everywhere and everyone’s concerned issue, desalination seems a potential answer to the water quality and quantity problems [15]. It is a process adopted to remove the extra amount of salts/minerals from available water to make it drinkable [16
18]. In this process, water is separated into two streams: one with permissible dissolved salts, and other with high percentage of dissolved salts (called as brine). Basic schematic of desalination process is shown in Fig. 11.2. Desalination techniques are applied to process the available feed water content that is categorized based on its salinity content: G G

Saline water Brackish water

Water with 30% salinity is labelled as saline water [19
21] where main content is sodium chloride (NaCl). Saline water has three categories: (1) moderate saline water with 3000
10,000 ppm salinity; (2) high saline water with 10,000
35,000 ppm salinity concentration and (3) sea water that has B35,000 ppm salinity [22]. Salinity concentration mainly depends on the percentage of NaCl content. As NaCl content increase, freezing point of the water decreases, while density, viscosity and refractive index increases. Water with salinity content higher than that of freshwater but less than that of saline water, that is lies in 0.5%
30% range is called as brackish water [19,23,24]. The removal of excessively present salinity content and to make it drinkable for humanity has become very necessary to fulfill the basic water needs

FIGURE 11.2 Desalination process schematic [9].

336

PART | III Applications

of mankind in concern with SDGs [2]. Desalination techniques can help in achieving these goals for which they require input thermal or electrical energy. These power demands are mostly fulfilled by combustion of fossil fuels [25]. The extensive combustion of fossil fuels is also producing hazardous environmental emissions [25] and disturbing the nature. The renewable energy (RE) resources are the possible answer that should be integrated to drive the desalination techniques for achieving sustainable water production goals. This chapter is aimed to highlight the different aspects of RE driven desalination techniques. In the start, the desalination and its different techniques are defined, then focus remained on answering the following questions: G

G

G

What is the current status of desalination installed capacity globally and how much research is focused on different desalination techniques to make them technically, economically, environmentally and sociopolitically favourable? What kind of relation exists between desalination techniques and their driving forces, that is energy sources? Then attention was given to elaborate the attempts that are made on making the different RE resources driven desalination techniques successful. How much brine waste is being produced through different desalination activities and how could it be managed? Possible reduction in CO2 emission due to integration of RE sources is also discussed.

The chapter is organized as: desalination technology categorization, its different techniques, their global installed capacity and research trends are discussed in Section 11.2. Section 11.3 defines the input energy required to drive desalination techniques and introduces the RE sources and their percentage share involved in driving desalination. Section 11.4 elaborates the technical, economic and social aspects of desalination technology when driven by different RE sources, that is solar [photovoltaic (PV) and thermal], nuclear, wind, geothermal and ocean/wave. Certain barriers (like brine production and CO2 emissions) and opportunities (like brine waste to energy, and emission reduction) associated with desalination technology development pathway are highlighted in Section 11.5. Section 11.6 represents the concluding remarks along with some suggestions to handle the produced waste.

11.2 Desalination technology Desalination techniques are categorized into two groups based on the input power source either thermal or electrical energy. There is a third category that involves ion-exchange methodology for desalination. Each desalination category is coupled with multiple techniques. Desalination technology categorization is defined in Fig. 11.3.

Renewable energy-driven desalination Chapter | 11

337

FIGURE 11.3 Desalination technology categorization [9,13,26,27].

11.2.1 Thermal desalination techniques In thermal desalination process, saline/brackish water is evaporated, and fresh water is gathered through the condensation of vapours. Thermal heat energy is consumed to evaporate the water. This is heat energy recovery process-based technique. The amount of heat energy being recovered depends upon the number of stages in which desalination process is completed. Around 25% of total produced fresh water is the outcome of thermal energy-driven desalination techniques [28]. Thermal desalination techniques include multieffect distillation (MED), multistage flash (MSF), vapour compression (VC) and adsorption desalination (AD).

11.2.1.1 Multieffect distillation MED is the first of its kind and more than 150 years older desalination technique exists since 1840 [29]. MED process is based on evaporation
condensation principle. The capacity of the system exists in 600
30,000 m3/day range [30]. The performance of MED process depends upon the number of the stages/effects that vary from eight to seventeen. At the first stage, vapours are produced that provide evaporating media for the next stages/effects. As pressure drops progressively among multieffects,

338

PART | III Applications

water evaporates at lower temperature. This process continues till last stage where vapours are condensed in separate cooled condenser [30,31].

11.2.1.2 Multistage flash desalination MSF desalination system was first designed in 1957 and installed in Kuwait [32]. MSF technique dominated the desalination industry soon after its invention because of its characteristic higher performance compared to MED, and currently accounts for 40% of the total desalination market [33]. The reason associated with higher performance was the utilization of heat of condensation for preheating of the feed water before flashing into the chamber [34]. Additionally its low corrosion and fouling effects make it more attractive option than MED. MSF system operates on number of vacuum stages/chambers. Pressure at each stage decreases progressively compare to the last one. Preheated feed water is introduced into the first vacuum chamber/stage where it evaporates rapidly (called flashing) due low pressure. The flashing of feed water process continues as pressure drops progressively at each next stage compared to the last one. Produced vapours got condensed on the surface of preheating tubes results in producing distillate and heat energy that is transferred inside the tubes to preheat the feed water [35]. 11.2.1.3 Vapour compression desalination VC desalination setup is usually used in attachment with MED system to improve the performance ratio and efficiency of MED process. Vapour compression is done thermally (thermal vapour compression, TVC) or mechanically (mechanical vapour compression, MVC). In this process, vapours produce at certain stage/effect are recompressed to reuse the vapour heat. Production capacity size of TVC system may range to 20,000 m3/day, while that of MVC ranges up to 3000 m3/day [9]. 11.2.1.4 Adsorption desalination AD was first introduced in 1881 [36]. It has a capacity of producing good quality drinkable water with lowest energy consumption  1.5 KWh/m3, because 45 C
85 C temperature is the requirement of sorption process that is attained by utilizing solar/geothermal energy or low-grade waste heat [11,37]. This thermal desalination process has capability to overcome the limitations of MED and MSF processes [9,11,38]. Thermal desalination is one of the oldest and much developed technology. It has multiple techniques to process the feed water. One of the thermal desalination techniques (i.e. MSF) has ability to reduce the maximum salinity concentration [27]. Thermal desalination demands less input electrical energy compared to membrane techniques [27]. Overall they are less energy efficient where energy demand (including both electrical and thermal

Renewable energy-driven desalination Chapter | 11

339

equivalent) is much higher and have lower desalinated water product yield [39]. At the same time, thermal desalination techniques are comparatively expensive in terms of per unit water production cost, and have higher CO2 emissions values [27].

11.2.2 Membrane desalination techniques There are multiple membrane desalination technologies: reverse osmosis (RO), electrodialysis (ED/EDR), forward osmosis (FO) and nano-filtration (NF). Electrical energy input is used to drive the RO, ED and FO processes [13].

11.2.2.1 Reverse osmosis The addition of RO process was a breakthrough in the field of science and technology where evaporation was not required. In this technique, semipermeable membranes made of cellulose acetates, polysulfones and polyamides are used. Feed water is pumped through the membranes at high pressure. Membranes filter out the drinkable water from high concentrated solution [40]. RO technology is more favourable for brackish water where 10
15 bar pressure is required to process the feed water, while seawater RO desalinators operate in 50
80 bar range [9]. 11.2.2.2 Electrodialysis Although ED principle was first introduced in late 19th century, its implementation at industrial scale started around 1960s [41]. ED system in comparatively simple comprised of ion-exchange membranes (IEMs), ED stack, input electrical energy supply and some accessories. ED stack serve as system’s structure consists of two end plates. In-between the ED stack plates, there is a series of IEMs. Other accessory elements such as electrodes, spacers, gasket gel and feed and concentrate compartments are also designed in-between ED stack plates. During ED process, ionic current is converted into electrons’ current passes through IEMs, electrolyte and external electric circuit [42]. The research progress on ED desalination system remained much slow. Even RE sources got integrated with ED desalination system just 5 year back in 2015 [41] that resulted in making this technique environmental friendly. 11.2.2.3 Forward osmosis FO is one of the emerging desalination techniques. It operates on natural phenomena of osmotic pressure gradient in which saline water flows from low osmotic pressure region towards high osmotic pressure region [43]. A semipermeable membrane separates these two different pressure regions through which water passes and salinity concentration decreases. Desalinated

340

PART | III Applications

water has high osmotic pressure compared to saline water [44]. FO desalination is a novel one and getting attention because certain limitations associated with other membrane desalination techniques can be overcome. Advantages associated with FO technique are low membrane fouling, less energy intensive [43] and high drinkable water content recovery [45]. Membrane desalination is the most researched and commercially developed technology evident from the fact that RO and ED desalination techniques account for 68% of the global desalination installed capacity [42]. Membrane desalination techniques demand less input energy value, have higher water production rate with less unit water production cost [27,39], and are environment friendly with reduced CO2 emissions [27]. There are certain limitations associated with membrane desalination technology like membrane fouling, scaling, etc. [42].

11.2.3 Desalination installed capacity and trends 11.2.3.1 Global status of desalination It is expected to increase in water demand by .55% till 2050 [5]. To address this increasing demand, there is an exponential increase in the number of installed desalination facilities worldwide since the start of 21st century to provide the sustainable solution to the thirsty humans in concern with the SDG-15 the most [28,46]. There are around 20,000 desalination plants in operation worldwide, producing 126 million m3/day drinking water [19,47]. Around 300 million people of 150 countries are relying on desalinated water [47]. All the produced desalinated water is not consumed by humans, but also used in agriculture and industrial sectors [19]. On regional basis, middle east and north Africa account for 47.5% of the total installed capacity [28]. Regional percentage share in global desalination installed capacity is shown in Fig. 11.4. In middle east, Gulf Cooperation Council countries especially Kingdom of Saudi Arabia leads the lane. List of few countries having operational desalination plants and are at the forefront in this technology race, is given in Table 11.1 [48]. In terms of desalination technologies (evident from Fig. 11.5) [28], RO is the most dominating type accounts for 84% of total installed units, and producing 69% of the total produced desalinated water. Following the RO type, the most developed technologies are of thermal type, that is MSF and MED, accounts for 18% and 7% share of the total produced desalinated water. Salinity content of the desalinated water is one of the most important quality parameters to be considered while deciding a certain technique for processing the feed water. Table 11.2 shows the capacity of each desalination technique of removing the salts from feed water to make it drinkable. It is concluded that MSF and ADs handle the feed water with highest salinity content and produce good quality water with lowest salinity percentage [27,49
51].

Renewable energy-driven desalination Chapter | 11

341

FIGURE 11.4 Regional share in global desalination capacity [28].

TABLE 11.1 Countries with operational desalination plants [48]. Country name

No. of plants

Country name

No. of plants

KSA

32

Algeria

15

United States

2000

UAE

13

Spain

13

Australia

10

Oman

08

Israel

08

South Africa

07

Maldives

07

Singapore

05

Egypt

04

Bahrain, Norway, Grand Cayman,

03 each

Hong Kong, India, Kazakhstan, UK

02 each

Germany, Cyprus, Chile, Malta, Qatar, Pakistan, Aruba, China

01 each

KSA, Kingdom of Saudi Arabia; UAE, United Arab Emirates.

For desalination processes (evident from Fig. 11.6) [28], sea water is the most abundantly used feed water type accounts for 61% of the total produced desalinated water. Brackish water and river water are used to produce 21% and 8% of the total desalinated water, respectively, and are in queue following the sea water.

342

PART | III Applications

FIGURE 11.5 Global status of produced desalinated water capacity from different desalination techniques [28].

TABLE 11.2 Capability of the desalination technologies of reducing salinity content from feed water to produce desalinated water [27,49]. Desalination technique

Feed water salinity ( 3 103 ppm)

Produced water salinity (ppm)

RO

45

10

MSF

70

10

MED

45

10

MVC

42

10

HDH

35

400

SD

55

80

Freezing

37

100

FO

45

6

NF

1

90

ED

6

250

ADs

67

10

I.Ex

1.5

13

G.Hyd

35

200

RO, Reverse osmosis; MSF, multistage flash; MED, multieffect distillation; MVC, mechanical vapour compression; HDH, humidification de-humidification; FO, forward osmosis; NF, nanofiltration; ED, electrodialysis; ADs, adsorption desalinations; I.Ex, ion-exchange; G.Hyd, gas hydrate.

Renewable energy-driven desalination Chapter | 11

343

FIGURE 11.6 Global status of produced desalinated water capacity from different feed water types [28].

11.2.3.2 Research trends in desalination A number of researchers are focusing on this domain with the aim of improving the water quality and quantity throughout the world as ‘Water is not only a developing world issue, it is related to everyone and everywhere’ [14]. Around 16,500 papers have been published on desalination subject since 1980
2018. This increase is exponential after 2010, when number became double during each five-year period [28]. In the start, total focus was on technical aspects of the desalination. There was no appreciable focus on economic, energy and environmental aspects of the desalination till 2000. In recent decade, the economics and energy aspect of the desalination got attention, and number of published articles increased from ,400 (in 2000) to .5000 in 2018. Similarly environmental aspect was neglected till start of 21st century. Then this aspect got pace and B2000 manuscripts have been published since 2018 addressing the environmental aspect of the desalination processes [28]. Socio-political aspect of this field is still not in focus. Supply of the clean water is not only associated with social demand, but there are a number of political aspects serve as barriers in the progress of this technology [28,52,53]. There are multiple desalination technologies on which research is being carried out and papers are getting published. Number of publication on different desalination techniques during 1980
2018 period are shown in Fig. 11.7 [28]. Among them, RO is the most researched technology with maximum number of published papers and an exponential increase is

FIGURE 11.7 Number of research publications on different desalination techniques during 1980
2018 time period [28].

Renewable energy-driven desalination Chapter | 11

345

observed in this number after 2000. Along with RO, emerging technologies (including FO, MD, NF) got attention during the last decade especially. Term ‘other desalination technologies’ mentioned in Fig. 11.7 include humidification de-humidification (HDH) and VC that are also in consideration but much less compared to the emerging and RO techniques [28,54]. In-short, desalination in mature and well proven technology. It is also a need of time to provide the basic requirement of mankind for the existence on earth planet. There are multiple available techniques that can be applied to get portable water. One can choose any certain desalination technique by considering certain factors and making compromises between advantages and disadvantages associated with each technique. Factors to be considered are as follows [55]: G G G G

Required water production rate, Available feed water type, Available and suitable input energy source (either electrical or thermal), Capital cost of the system, especially required land area and transportation cost, etc.

11.3 Energy and desalination Desalination is an energy intensive process that requires thermal (for thermal techniques such as MSF, MED, ADs, HDH) or electrical (for membrane techniques especially RO, ED) energy input for the process [27,56,57]. Input energy is also one of the most important criteria to be considered while choosing a certain desalination technique in concern with SDG 6, 7 and 13 [2]. Input energy accounts for B36% of the total desalination process operational expenditures [57]. In past, most of the input energy requirement was fulfilled by combusting fossil fuels and still the majority portion of fuel mix accounts for conventional liquified fuel products [56,57]. Among desalination techniques, each require a different amount of thermal/electrical energy values as tabulated in Table 11.3 [27,58]. It is concluded that all emerging techniques such as ADs, ion-exchange, G.Hyd and solar distillation require much less amount of input energy comparatively. While among developed technologies, RO process needs much more electrical energy compared to the thermal technologies such as MSF and MED. But on the other end, MSF and MED need both electrical and thermal input energies in comparison with RO that needs only electrical power. The most expensive type of the desalination is HDH that needs maximum amount of thermal energy along with appreciable input electric power.

11.3.1 Renewable energy resources for desalination With the start of 21st century, RE resources got attention in the field of desalination. And this energy transition is moving with pace to make this

346

PART | III Applications

TABLE 11.3 Input energy required by different desalination techniques for per cubic meter water desalination operation [27,58]. Desalination technique

Input electrical energy (kWh/m3)

Input thermal energy (kWh/m3)

RO

8.2




MSF

5.2

19.4

MED

3.8

16.4

MVC

11.1




HDH

03

120

Freezing

11.9




FO

05




NF

4.49




ED

5.5




ADs

1.38




I.Ex

1.1




G.Hyd

1.58




RO, Reverse osmosis; MSF, multistage flash; MED, multieffect distillation; MVC, mechanical vapour compression; HDH, humidification de-humidification; FO, forward osmosis; NF, nanofiltration; ED, electrodialysis; ADs, adsorption desalinations; I.Ex, ion-exchange; G.Hyd, gas hydrate.

technology reliable, affordable and sustainable. Efforts are made to run the desalination techniques by harnessing the renewables to the maximum [5,59
61]. Around 131 installed desalination plants are powered by RE resources, contributing for 1% of the total world produced desalinated water [5]. RE resources that can be harnessed to empower the desalination processes are: solar (including solar thermal; solar PV; CSP); wind; nuclear; geothermal and Ocean/wave. Among these, solar PV is the leading one with 43% contribution towards total RE share in energy mix for desalination, while solar thermal accounts for 27%, wind for 20%, and 10% share is of hybrid RE technologies (as shown in Fig. 11.8) [5,23]. RE share in energy mix in increasing with pace to minimize the threat of extinction of fossil fuels. It can play an important role in improving energy security leading towards sustainable future and environment friendliness. But the issue is: renewables can serve as an add on only, they can never replace the fossil fuels [62]. Because there are certain limitations associated with them and the most important one is their unpredictable nature. No one can predict the incident solar irradiance power or wind speed exactly. That’s why there is always a need of backup.

Renewable energy-driven desalination Chapter | 11

347

FIGURE 11.8 Percentage of RE share to drive the desalination techniques [5]. RE, Renewable energy.

11.4 Renewable energy integrated desalination: technical, economic and social development aspects 11.4.1 Solar desalination Solar desalination is categorized into two main types based on the mode of harnessing the solar incident irradiance power, that is thermal or photovoltaic [63]. Percentage contribution of solar thermal and PV energy towards different desalination applications is shown in Fig. 11.9.

11.4.1.1 Solar photovoltaic desalination Solar PV technology has the highest installation capacity share among the renewables globally [25]. Solar PV technology outcome is electricity only, so it empowers only those desalination techniques that require electrical power input such as RO, ED and MVC. RO is the most mature and

348

PART | III Applications

FIGURE 11.9 Current status of solar energy integrated with different desalination techniques [64,65].

developed desalination technology among all available options [28,65]. Inshort, PV-RO desalination systems combination is the leading one accounts for 32% of the total RE driven desalination combinations [65,66] as depicted in Fig. 11.9 too. Solar PV powered RO desalination systems are considered the most for producing fresh water for domestic, agriculture and remote regions [67]. The success of this solar PV driven RO desalination systems is associated with four factors [68]: G

G G

G

Flexibility in modularity of PV system, that is number of modules can be increased in the system after the installation, low maintenance cost and long-life of around 20 years of PV system, Fresh water demand usually increases with increase in temperature. This increase in temperature is associated with increase in incident solar irradiance intensity that favours the solar PV system, Predictable bell-shaped diurnal solar irradiance curve.

A number of studies have been performed on solar PV driven RO systems to find out its technical and economic feasibility in different configurations/schemes [68
77]. As RE technologies got mature, prices of PV cells decreased. This reduction in prices made this combination more favourable [72]. Even the location of the PV-RO system matters a lot as PV is too much site dependent technology alongside with weather dependent. In this concern, multiple feasibilities have been discussed to make this combination more viable [67,70
72,78,79]. Performance of the solar PV driven RO plant was

Renewable energy-driven desalination Chapter | 11

349

evaluated by providing battery storage backup against directly connected RO system, and concluded that RO system connected directly to the solar PV modules performs better [71]. Another comparative study of PV-RO system was performed against the diesel engine driven RO system with aim of providing the full day uninterrupted power, and concluded that PV-RO system produced fresh water at a very competitive cost [67]. Overall, solar PV-RO systems are found techno-economo-environmentally more feasible compared to the diesel engine driven in those areas which are rich with incident solar irradiance power [72,78,79]. Manolakos et al. [70] evaluated the economic feasibility of RO plant, connected directly with solar PV array, against the desalination system operated by organic Rankine cycle, and concluded that the water production cost of directly connected PV-RO plant was appreciably lower with 7.77 h/m3 value compared to 12.53 h/m3. The feasibility of PV-RO system largely depends upon the performance of PV array. Number of the researches have been done to improve the outcome of PV array through solar tracking either single axis or dual axis, and concluded that single axis may lead to 33% increased annual water production while dual axis by 36% [69,75,76,80
82]. Another factor associated with PV-RO system performance is accumulation or scattering of dust on the surface [83]. The dust particles can lead to the increase in panel temperature that could cause physical damage [84]. In this concern, multiple cleaning techniques such as mechanical cleaning methods, PV coatings and electrostatic methods have been suggested [85,86]. Ambient temperature is also an important factor that affects the PV panel conversion efficiency. A number of cooling methodologies have been suggested for PV-RO system configuration to keep the PV panel temperature low [84,87]. Hybrid RE systems driven desalination techniques can perform technically and economically better. Studies have been performed to check the reliability and feasibility of RO desalination process driven by wind-PV hybrid RE system [88,89]. On the other side, RO membranes are very important part of desalination process. Normal lifetime of RO membrane is five years under standard operating conditions. Number of efforts have been made to improve the performance and lifetime of RO membranes [90
93]. Earlier PV driven RO plants are designed to operate at constant conditions such as flow, pressure and power level. Now efforts are made to propose different strategies that could enable the PV driven desalination system to operate under variable conditions, and to make the systems more economical, and be able to operate for full day [65,94,95].

11.4.1.2 Solar thermal desalination Solar thermal desalination (STD) is a combination of two systems in which solar thermal heat energy is used as input power for different desalination techniques like MED, MSF, AD, HDH, VC, etc. [96]. Among these

350

PART | III Applications

desalination techniques, MED and MSF are the most developed and commercially deployed types. STD is the second most developed technology after solar PV driven RO desalination system [64,65]. STD is potentially economic and sustainable option for providing high quality fresh water when the world is worried about water and energy future [39]. STD technology uses thermal collectors for small/medium scale applications where up to 100 C
120 C temperature will be required, while concentrated collectors are used for large-scale applications where required temperature will be above 120 C [96,97]. Solar thermal driven desalination is much older technology evolved during late 19th century. Primarily MSF and MED desalination systems were installed and evaluated for research and development purpose till the start of 21st century in comparison with PV-RO technology that was much mature and commercialized at that time. Hence, direct solar thermal powered desalination systems were limited to small scale applications [96,98
102]. STD systems got attention and became tremendously interesting from developing and improvement in performance point of views since last few years [103,104]. The performance analysis of STD system should include the evaluation of the two parameters [105]: G

G

The performance ratio (PR), is a ratio between water enthalpy of phase change and specific energy consumption of the process itself, The gain output ratio (GOR) is a ratio between the obtained product (kg of condensed vapour) and energy required for the process in terms of kg of condensed vapour of the thermal source.

An extensive quantity of literature focused on solar thermal driven desalination is available that deals with qualitative and quantitative analysis of STD systems for making this technology economically and technically feasible [96,98
102,106
110]. Most of the researchers focused on developing high quality materials for efficient photo-thermal conversion [111
117]. Despite of these efforts, there is lack of understanding of the fact that how these material-based innovations can enhance the overall performance of STD [111
113,117
121]. Efficient condensation and the recovery of latent heat of condensation related knowledge gap is still a part of the literature [122
124] that should be addressed to make the solar thermal driven desalination systems techno-economically more feasible.

11.4.2 Nuclear energy-driven desalination The International Atomic Energy Agency put a lot of efforts on integrating the nuclear energy with desalination techniques since 1960s [125,126]. Number of reports have been published in this concern that conclude the nuclear energy as a viable alternative option for empowering the desalination process due to certain associated features like environmental protection due

Renewable energy-driven desalination Chapter | 11

351

FIGURE 11.10 Economic comparison of power generation between nuclear and other mostly used energy resources [125].

to reduced GHG carbon emissions, economically feasible for remote areas where fossils fuels are not available or not economically feasible to combust, and ultimate conservation of fossil fuels [127,128]. Nuclear energy production cost is lower than many available energy sources such as gas, coal, peat, wood and even that of wind (evident from Fig. 11.10) [125]. Nuclear energy can be used to produce fresh water from seawater through desalination process either as thermal energy input or electrical energy input [129]. Among desalination techniques, MSF and MED require thermal energy input while RO rely on electrical energy input. The energy requirements of both technology categories can be fulfilled at a reasonable cost compared to fossil fuel desalination [130,131]. At the first in 1968 [132,133], a dual-purpose nuclear power plant was evaluated technically and economically for providing electricity and fresh water to agro-industrial complexes in the Middle East. Then in late 1970s [133], Libya signed a contract to develop a dual-purpose nuclear power plant with seawater desalination and power generation purposes. In the start of 21st century, nuclear desalination transition got attention. Faibish and Ettouney [134] evaluated the multiple schemes of MSF desalination process powered by nuclear energy and concluded the process for heating nuclear reactor and nuclear power plant. Wu [135] designed 200 MW(th) nuclear heating reactor coupled with MED process and evaluated the economics of the system. It was concluded that average energy cost of the nuclear plant may reach to 5.44 $/t of steam, while the cost of producing water might be 0.72
0.76 $/m3. Hanra [136] studied the RO-MSF hybrid

352

PART | III Applications

system powered by nuclear heat energy with aim of proposing the reduced maintenance and operation cost of the produced desalinated water. Al-Mutaz [137] studied the hybrid RO-MSF process coupled with nuclear reactor and concluded that this coupling increases the overall availability factor of the desalination process. Likewise, several studies have been carried out on hybrid desalination (RO-MSF and RO-MED) systems empowered by nuclear energy [138
143] to analyze their thermal and economic aspects. Some famous hybrid nuclear powered desalination plants in the world are: Shevchenko (MED-MSF) in Kazakhstan, Karachi nuclear power plant (ROMED) and NDDP (MSF-RO) in India [142]. There are 11 main nuclear reactor coupled desalination projects that were launched with R&D approach to evaluate and optimize their performance. These projects are as follows [125,127,144]: G

G

G

G

G

G

G

G

G

CANDESAL in Canada [145]: Pressurized heavy water reactor (PHWR) was coupled with RO desalination system and 20%
40% desalinated water production efficiency improvement was achieved. BARC in India [146,147]: RO and MSF desalination system were coupled with PHWR with aim of producing 425 m3/day fresh water from MSF and 90 m3/day fresh water from RO process. INVAP in Argentina [148]: In this project MSF/MED/RO based simulation models were developed for safety assessment. KAERI in Republic of Korea [149,150]: System integrated modular advanced reactor (SMART) pressurized water reactor (PWR) was coupled with RO and MSF with the aim of getting 40,000 ton of water/day production, additionally MED-TVC process was coupled with SMART. IPPE, OKBM in Russia [151]: 20 test facilities were developed to evaluate, small modular (SMR) nuclear reactor was coupled with different desalination plants. CNSTN/TUNDESAL in Tunisia [152,153]: PWR was coupled with MED and RO desalination processes to assess its feasibility and to optimize them. INET in China [152,154]: In this project, nuclear heating reactor of 200 MW capacity was coupled with MED system to investigate two different MED desalination processes, that is low temperature horizontal tune desalination system of 120,000 m3/day capacity, and high temperature stack desalination system with 160,000 m3/day capacity. NPPA in Egypt [155]: Nuclear Power Plants Authority (NPPA) made a request in 1997 to investigate the RO desalination process coupled with PWR considering certain conditions such as pressure on RO membrane, feed water temperature and resultant time function for fresh water production of 140,000 m3/day. CNESTEN in Morocco [156]: nuclear heating reactor powered MED and RO desalination processes were evaluated economically.

Renewable energy-driven desalination Chapter | 11 G

G

353

BATAN in Indonesia [157]: SMR PWR reactor was coupled with desalination system to assess its feasibility and safety. EURODESAL in Southern Europe [158,159]: Hybrid MED-RO desalination system powered by nuclear energy is developed to evaluate its technical, economic and safety analysis.

In addition to these, there are many countries, such as Albania, Algeria, Croatia, Chile, DR Congo, Kuwait, Sri Lanka, Saudi Arabia, Peru, Thailand, Uruguay, United Arab Emirates (UAE), Uganda, Vietnam and Zambia, considering this nuclear transition. Some have cancelled their plans due to certain concerns associated with nuclear plants, while Middle East countries (UAE, Saudi Arabia and Kuwait) are the best options for nuclear desalination. It is a need of the time to utilized nuclear energy for desalination purposes but there are certain concerns that should get addressed for fast nuclear desalination development. These concerns are: hazardous nuclear waste, possible radioactive contamination in produced water and water supply continuation during reactor shutdown period [125].

11.4.3 Wind energy-driven desalination Wind energy and desalination are two different technologies. There is need of an interface between these two technologies to transfer the wind generated power to the desalination system. Wind energy can empower the desalination process through four possible ways: electrical energy, thermal energy, gravitational potential and kinematic energy (shaft power). Electrical is the mode of interface mostly used till now to integrate the wind power with desalination system. RO, ED and MVC are desalination processes that need electrical input energy [160
162]. Modern history of wind energy started around 1975 when wind turbines were commercialized, and Europe put focus on producing power from nonnuclear RE sources. Wind energy integration with desalination technology was in experimentation phase since the start of 21st century [106]. Rodriguez et al. first carried out preliminary economic analysis of RO plant powered by wind energy [163]. Weiner et al. designed a RO plant powered by wind and solar both for remote area locality. System was designed to operate at 33% service factor, with two days battery storage autonomy [164]. Veza et al. designed and tested ED desalination system empowered by wind energy input with the aim of choosing the most suitable desalination system for connecting to medium scale off-grid wind form. They concluded that EDR system can easily be empowered by wind energy [165]. Pestana et al. [166] evaluated the operation of RO plant connected to wind energy system without any storage. A distiller was proposed by Nakatake and Tanaka [167] empowered by fractional thermal energy produced by windmill. Performance

354

PART | III Applications

of the distiller evaluated theoretically and predicted that 1.5 kg/day fresh water can be produced when 6 m/s wind speed is available. Forstmeier et al. [168] developed the physics based models of RO and MVC desalination systems empowered by wind energy to check their technical and economic feasibility, and concluded that wind powered desalination systems are in competitive environment with others. RO desalination systems powered by gravitational and kinematical energy have been installed and studied [101,169]. Literature is evident that wind energy is mostly used to drive and evaluate the RO desalination systems as shown in Fig. 11.9 [64,65,93,170
173]. Since last few years, efforts are made to drive the RO system with the combination of wind and solar PV technologies along with storage to make the system reliable [65,174
176]. Because solar PV installed capacity is highest among all renewables and growing pace is also at the top, and RO desalination system and research trend on its operation improvement are also in the most attention. There are some concerns associated with wind driven desalination technology due to which it is used to power the desalination system comparatively less than solar PV: appreciable fluctuations in produced power as wind speed varies continuously, wind energy potential is mostly available in remote areas especially on coastal line [65,173].

11.4.4 Geothermal energy-driven desalination Geothermal energy is well proven and mature form of technology that can be used at commercial level for multiple applications such as electric power production, cooling, heating and industrial applications. There are some features associated with geothermal energy source that make it more reliable and sustainable source for desalination process compared to some other renewables. These features include [177
181]: G

G

G

G

G

Geothermal production technology, that is hot water extraction form underground aquifers is not affected by seasonal changes. Geothermal energy has high capacity factor (evident from Fig. 11.11) and can be used to fulfil the demand of smallest to largest energy consuming utilities. Geothermal resources ensure reliable and stable heat energy supply to both major desalination techniques, that is thermal desalination and RO. Geothermal resources with 70 C
90 C temperature range are available abundantly around the world that can be an ideal option to empower the MED desalination (as it needs low temperature) process. Geothermal resources available with 1100 C can be utilized for power generation and certain heating applications, and can lead to the conservation of 46%
89% energy savings (as shown in Table 11.4) [182].

Renewable energy-driven desalination Chapter | 11

355

FIGURE 11.11 Capacity factor (in percentage) of different renewable energy resources [182,183].

TABLE 11.4 Potential energy savings for geothermal driven desalination applications [182]. Desalination technology

Energy required (MJ/m3)

Saving in energy required amount (MJ/m3)

Conventional MED-TVC

194




Hybrid MED-SWRO

105

46%



28

85%



23

88%



21

89%

Geothermal MED at 100 C Geothermal MED at 115 C Geothermal MED at 130 C

MED, Multieffect distillation; SWRO, sea-water reverse-osmosis; TVC, thermal vapour compression.

G

G

G

Integration of geothermal energy with desalination process is costeffective Simultaneous production of water and power is possible from geothermal resources. Geothermal desalination is environment friendly as no emissions are produced.

356 G

G G G G G

PART | III Applications

Geothermal resources require lower surface per unit energy production compared to all other renewables, for example 10 m 3 10 m well size is enough for 20 MWth energy production. The hottest geothermal resources are mostly available in few deserts of the world like in United States, Mexico, central America, Caribbean Islands, Middle Eastern countries and North African Regions. Cogeneration schemes that require 120 C
200 C or higher temperature can be integrated with these resources. Low temperature desalination techniques such as solar stills, LTMED (low temperature MED) and HDH processes that require 40 C
70 C can be integrated with geothermal sources available abundantly on the earth surface [184]. In start 21st century, geothermal was not so much focused source to empower the desalination techniques. Later, researchers and developers analyzed its effective in comparison with other resources. Geothermal energy resource for desalination process is still relatively unexplored and limited number of studied have been performed in this concern [185
189]. Awerbuch et al. [190] studied the geothermal desalination for the first time and proposed it for power and water production. A study on integration of MSF and HTED (high temperature ED) with geothermal energy was performed [191]. A seawater desalination system driven by low enthalpy geothermal energy was studied in Greece [192]. Bouchekima [193] performed a study to analyze the performance of solar still by providing brackish geothermal feed water. Gude [194] studied the different configurations of RO desalination system integrated with renewables, and concluded that the increase in feed water temperature from 25 C to 35 C for RO process integrated with geothermal energy source will enhance the permeate flux rate by 34%. Multiple efforts have been made by different researcher to make this integration technically and economically optimized, to aware the society and to bring this geothermal driven desalination technology in competition with other renewables’ driven [195
198]. As a result of these efforts, some geothermal driven desalination plants have been installed around the world: MED and MSF desalination plant in Baja California, Mexico [199]; A two stage MED in Kimolos, Greece [192]; A HDH in Tunisia [200]; MD coupled with multiple effect distiller in Tunisia [185]; MED/VTE (2 effects) in Salton sea/ Imperial; MED/VTE (15 effects) Valley, United States [181,201].

There are some challenges due to which geothermal driven desalination application is not growing in a pace as others are now-a-days [180,182]: large quantity of heat for long-term operation, easy access near to the point of application, low elevation and easy access to roads, low tendency of pipeline and wells scaling, proximity of electrical loads on transmission lines, etc.

Renewable energy-driven desalination Chapter | 11

357

11.4.5 Ocean/wave energy-driven desalination Ocean has multiple forms of the energy such as thermal gradient, tides, salient gradient and waves [202] that can be used for power and water production. Ocean/wave energy is the least developed type and has least contribution among renewables’ driven desalination applications. Douglas et al. firstly developed a device DELBOUY to use the pressurized water for desalination purpose through RO membranes [203]. Verduzco-Zapata et al. did numerical modelling for increasing the buoy size that resulted in production of enough force required for pressurizing the seawater. Later on, they developed that system where desalination process was driven by low energy of the ocean surface waves [204,205]. The ocean energy-driven desalination technology was not in practice till start of 21st century [96]. Davies [206] developed wind-wave powered distiller with the focus on the fact that an uninterrupted energy supply will be available due to wind power source. The use of ocean thermal gradient energy for fresh water production through vacuum desalination was reported by Mani et al [207]. Kumar et al. [208] designed ocean thermal gradient energy-driven desalination system where brackish water was input feed and, concluded that the quality of the produced water was drinkable. In spite of the fact that an estimated power potential of 8000
80,000 TWH/year can be harnessed from wave energy, the integration of wave energy with desalination technology is least developed. It is evident from the literature that it was more than a decade ago when wave powered desalination application was discussed considerably [209]. Now researchers are focusing on this application due to increasing demand of sustainable, reliable and cost-effective drinkable water supply [209]. In this concern, some wave powered desalination configurations/schemes have been proposed that might be potentially developed systems in future. Proposed schemes are as follows [209
211]: G G G

Oscillating water column driven RO, VC and ED desalination processes; wave activated body driven RO, VC and ED desalination processes and overtopping wave energy system driven RO, VC and ED desalination processes.

Overall, RE energy integration with different desalination techniques is dependent on different factors that are summarized in Table 11.5 [55].

11.5 Barriers, issues and opportunities in desalination technology development Water and energy production from sustainable and reliable sources and their integration is a need of the time for securing the future. Existence of improvement options in each type of desalination techniques reflects that no

358

PART | III Applications

TABLE 11.5 Comparative suitability evaluation of RE resources for desalination technology [55]. Parameters Energy Source

Desalination techniques can be driven

Source availability continuity

Power availability security

Site requirements

Solar PV

RO, ED, MVC

Output is not continuous (storage/ backup is required)

Output is unpredictable

Good match and available almost everywhere

Solar thermal

Through heat: TVC, MED, MSF Through shaft power: MVC, RO, ED Through electricity: RO, MVC

Output is not continuous (storage/ backup is required)

Output is unpredictable

Good match and available almost everywhere

Nuclear

Heat energy: TVC, MED, MSF Electricity: RO, ED, MVC

Output is continuous

Output is predictable

Source is limited to location

Wind

Shaft power: MVC, RO Electricity: RO, MVC, ED

Output is not continuous (storage /backup is required)

Output is unpredictable

Source is location dependant

Geothermal

Heat energy: TVC, MED, MSF Electricity: RO, ED, MVC

Output is continuous

Output is predictable

Source is limited to location

Ocean/wave

Heat energy: TVC Electricity: RO, ED, MVC

Output is not continuous (storage /backup is required)

Output is unpredictable

limited to location

RE, Renewable energy; RO, reverse osmosis; ED, electrodialysis; MVC, mechanical vapour compression; TVC, thermal vapour compression; MED, multieffect distillation; MSF, multistage flash.

Renewable energy-driven desalination Chapter | 11

359

one is fully efficient that is because of presence of associated limitations. Similarly there are different barriers and issues associated with each type of RE resources like destabilized output power of wind energy due to continuous variations in air speed, fluctuations in output solar power due to variant incident irradiance intensity, etc. But there are certain common issues associated with all desalination techniques: ‘brine’ production; cost effectiveness and environmental emissions.

11.5.1 Brine production Brine is a highly concentrated salt that is removed from feed water to make it drinkable. Amount of the brine targeted to remove depends upon the type of feed water, either sea water or brackish water [212]. Maximum possible production of brine during desalination process lead to higher quality of produced water. How much amount of brine needs to be removed directly leads to the amount of input energy required [213]. Current global brine production is around 141.5 million m3/day (total 51.7 billion m3/year). This value is 50% higher than the value of produced desalinated water against which this brine get produced (it means brine production is 1.5 times of the produced water [214]). Maximum share of this brine is produced by Middle East and North African countries as majority number of desalination systems are installed in these regions. Global regional brine production figures are shown in Fig. 11.12 [28]. This produced brine waste amount is directly associated with desalination technology growth. More the desalination technology will grow, greater will be the amount of the brine waste. As this waste seems an issue on one side, while on the other hand this is an opportunity for producing energy from waste. We can reduce the amount of brine waste from a certain system but not overall. Because water demand is increasing with pace that needs growth in desalination technology development. Ultimately there will be more brine waste that provides an opportunity to utilize this waste to produce energy that can be consumed again for desalination process or other application.

11.5.2 Desalination cost and CO2 emissions The whole world is struggling to achieve the defined SDGs. The success in achieving these SDGs depends upon the economic status of the country [1]. This financial status affects and being affected by the development of technologies. In-short, the cost of the water production process is very important parameter to make a certain technique successful, and a lot of focus is being diverted towards this direction to make the fresh water availability economical and feasible [215]. Water production cost values of different desalination techniques are presented here in Table 11.6 [27].

360

PART | III Applications

FIGURE 11.12 Global region wise brine production [28].

TABLE 11.6 Desalinated water production cost for different desalination techniques [27]. Desalination technique

Water production cost ($/m3)

Desalination technique

Water production Cost ($/m3)

RO

0.75

MSF

0.96

MED

0.86

MVC

0.92

HDH

3.93

Freezing

0.34

FO

0.75

NF

1.12

ED

0.83

ADs

0.2

I.Ex

1.12

G.Hyd

0.63

RO, Reverse osmosis; MED, multieffect distillation; HDH, humidification de-humidification; FO, forward osmosis; ED, electrodialysis; I.Ex, ion-exchange.

Another important concern of the world is environment. That’s why, Kyoto protocols have been defined and one of the SDGs is especially focused on environmental emissions. CO2 emission values produced by developed desalination technologies are tabulated in Table 11.7 [27]. Technical and economic improvement of the desalination systems and their integration with renewables will lead to reduction in these emissions further [216].

Renewable energy-driven desalination Chapter | 11

361

TABLE 11.7 Amount of CO2 released by desalination techniques [27]. Desalination technique

CO2 released amount (kg/m3)

Desalination technique

CO2 released amount (kg/m3)

RO

3.8

MSF

6.9

MED

5.5

MVC

5.1

HDH

29.1

Freezing

5.5

FO

2.3

NF

2.1

ED

2.5

ADs

0.6

I.Ex

0.5

G.Hyd

0.7

RO, Reverse osmosis; MED, multieffect distillation; HDH, humidification de-humidification; FO, forward osmosis; ED, electrodialysis; I.Ex, ion-exchange.

11.6 Outlook Importance of desalination to address the water shortage issue, faced by 41% population, and its different types are discussed. Global installed capacity and techno-economo-environmental research being carried out to improve the desalination techniques concludes that reverse osmosis is the most developed and researched desalination type. Both electrical and thermal energy resources are needed to drive the desalination techniques, and these energy requirements can be fulfilled by the integration of RE resources. Among RE resources, each source has certain operational limitations but solar seems the most promising one. Another major issue associated with increasing produced drinking water quantity through desalination is the brine production that can be resolved by applying waste to energy principles on this content. In-short, integration of RE resources towards desalination technology would be a milestone to address the energy
water
environment nexus pointing towards bright, healthy and eco-friendly future.

Abbreviations AD CSP ED EDR GCC G.Hyd HDH I.Ex LLE

Adsorption desalination Concentrated solar power Electro dialysis Electro dialysis reversal Gulf Cooperation Council Gas hydrate Humidification de-humidification Ion-exchange Liquid
liquid extraction

362 MD MED MSF MVC NF ppm RO RE SDG STD SW TDS TVC VC WHO

PART | III Applications Membrane desalination Multieffect distillation Multistage flash Mechanical vapour compression Nano-filtration Parts per million Reverse osmosis Renewable energy Sustainable development goal Solar thermal desalination Sea water Total dissolve solids Thermal vapour compression Vapour compression World Health Organization

References [1] Yadav A, et al. Strategic planning and challenges to the deployment of renewable energy technologies in the world scenario: its impact on global sustainable development. Environ Dev Sustain 2020;22(1):297
315. [2] Sustainable Development Goal Impact: WBCSD Water Stewardship. ,https://www. wbcsd.org/Programs/Food-and-Nature/Water/Water-stewardship. [3] Chen C, et al. Planning an energy
water
environment nexus system in coal-dependent regions under uncertainties. Energies 2020;13(1):208. [4] Chen L, et al. Demonstration of a feasible energy-water-environment nexus: waste sulfur dioxide for water treatment. Appl Energy 2019;250:1011
22. [5] Shahzad MW, et al. Energy
water
environment nexus underpinning future desalination sustainability. Desalination 2017;413:52
64. [6] Hoffman A. Water, energy, and environment: a primer. IWA Publishing; 2019. [7] NASA. Ocean worlds: oceans on earth, ,https://www.nasa.gov/specials/ocean-worlds/. [8] World Health Organization. Guidelines for drinking-water quality [electronic resource]: incorporating first addendum, vol. 1, recommendations. 3rd ed. ISBN 92 4 154696 4. [9] Shahzad MW. The hybrid multi-effect desalination (MED) and the adsorption (AD) cycle for desalination. National University of Singapore, PhD Thesis, 2013. ,https://core.ac.uk/ download/pdf/48679046.pdf . . [10] Table of world water distribution. ,http://serc.carleton.edu/details/images/12447.html.; [20 July 2008]. [11] Son HS, et al. Pilot studies on synergetic impacts of energy utilization in hybrid desalination system: Multi-effect distillation and adsorption cycle (MED-AD). Desalination 2020;477:114266. [12] ‘Drinking Water’: World Health Organization (WHO). ,https://www.who.int/en/newsroom/fact-sheets/detail/drinking-water. [13] Wankhade J, Gutte P, Mote S. Intelligent solar desalination system on sea water. 2020, EasyChair. [14] Sarni W. Water inequality is a global issue
here’s what we must do to solve it. World Economic Forum. ,https://www.weforum.org/agenda/2019/10/water-inequality-developing-world-usa-west.; [October 2019].

Renewable energy-driven desalination Chapter | 11

363

[15] Fountain H. The World can make more water from the sea, but at what cost? The New York Times. ,https://www.nytimes.com/2019/10/22/climate/desalination-water-climate-change.html? fbclid 5 IwAR0cBFft7rv2CoAV0Eanxcz5ml1sMuSkQSfbA2pnwTwBJLQN5ytP-sKKqIU.; [October 2019]. [16] Kucera J. Desalination: water from water. Hoboken, NJ: John Wiley & Sons; 2019. [17] Spiegler KS. Principles of desalination. Elsevier; 2012. [18] Vickers JL. Desalination unit with electricity generation. Google Patents, 2020. [19] Is water desalination the best answer to the water crisis, innovation: bluevision. ,https:// bluevisionbraskem.com/en/innovation/is-water-desalination-the-best-answer-to-the-watercrisis/.; [April 2019]. [20] Ahn J, et al. High performance electrochemical saline water desalination using silver and silver-chloride electrodes. Desalination 2020;476:114216. [21] Hoque A, Abir AH, Shourov KP. Solar still for saline water desalination for low-income coastal areas. Appl Water Sci 2019;9(4):104. [22] Rumble JR, Lide DR, Bruno TJ. CRC handbook of chemistry and physics: a readyreference book of chemical and physical data, ninety nine ed., CRC Press, Boca Raton, 2017. isbn: 9781138561632. ,https://olin.tind.io/record/1640176/ . [23] Alsarayreh AA, et al. Evaluation and minimisation of energy consumption in a mediumscale reverse osmosis brackish water desalination plant. J Clean Prod 2020;248:119220. [24] da Silva GDP, Sharqawy MH. Techno-economic analysis of low impact solar brackish water desalination system in the Brazilian Semiarid region. J Clean Prod 2020;248:119255. [25] World Energy Outlook- 2019. International energy agency. ,https://www.iea.org/reports/ world-energy-outlook-2019/renewables#highlights. [26] Mohammed RH, Askalany AA. Productivity improvements of adsorption desalination systems, in solar desalination technology. Springer; 2019. p. 325
57. [27] Youssef P, Al-Dadah R, Mahmoud S. Comparative analysis of desalination technologies. Energy Procedia 2014;61:2604
7. [28] Jones E, et al. The state of desalination and brine production: a global outlook. Sci Total Environ 2019;657:1343
56. [29] Al-Shammiri M, Safar M. Multi-effect distillation plants: state of the art. Desalination 1999;126(1
3):45
59. [30] Al-hotmani O, et al. Optimisation of multi effect distillation based desalination system for minimum production cost for freshwater via repetitive simulation. Comput Chem Eng 2020;106710. [31] Kalogirou SA. Seawater desalination using renewable energy sources. Prog Energy Combust Sci 2005;31(3):242
81. [32] El-Dessouky HT, Ettouney HM. Fundamentals of salt water desalination. Amsterdam; New York: Elsevier; 2002. [33] Sellami A, et al. Once-through multistage flash desalination installation combined with thermal vapour compression MSF-OT/TVC. Desalin Water Treat 2019;161:48
55. [34] Lv H, et al. Numerical simulation and optimization of the flash chamber for multi-stage flash seawater desalination. Desalination 2019;465:69
78. [35] Alsehli M, Choi J-K, Aljuhan M. A novel design for a solar powered multistage flash desalination. Sol Energy 2017;153:348
59. [36] Tien C. Adsorption calculations and modeling. Butterworth-Heinemann; 1994. [37] Thu K, et al. Performance investigation of advanced adsorption desalination cycle with condenser
evaporator heat recovery scheme. Desalin Water Treat 2013;51(1
3):150
63.

364

PART | III Applications

[38] Ng K, et al. Apparatus and method for desalination, 2010, Patent, ID Code: 109540 ,http://ecite.utas.edu.au/109540.. [39] Wang Z, et al. Pathways and challenges for efficient solar-thermal desalination. Sci Adv 2019;5(7) eaax0763. [40] Fitzsimons L. A detailed study of desalination energy models and their application to a semi-conductor ultra-pure water plant. 2011, PhD Thesis, Dublin City University. , http:// doras.dcu.ie/16388/ . [41] Campione A, et al. Electrodialysis for water desalination: A critical assessment of recent developments on process fundamentals, models and applications. Desalination 2018;434:121
60. [42] Sajjad A-A, et al. Electrodialysis desalination for water and wastewater: a review. Chem Eng J 2019;122231. [43] Xu S, et al. Novel graphene quantum dots (GQDs)-incorporated thin film composite (TFC) membranes for forward osmosis (FO) desalination. Desalination 2019;451:219
30. [44] Ray C, Jain R. Drinking water treatment: focusing on appropriate technology and sustainability. Springer Science & Business Media; 2011. [45] Martinetti CR, Childress AE, Cath TY. High recovery of concentrated RO brines using forward osmosis and membrane distillation. J Membr Sci 2009;331(1
2):31
9. [46] Robbins J. ‘As water scarcity increases, desalination plants are on the rise’: Yale environment 360. Published by Yale School of Foresty and Environmental studies. ,https://e360. yale.edu/features/as-water-scarcity-increases-desalination-plants-are-on-the-rise.; [June 2019]. [47] The IDA Water Security Handbook 2019
2020, Water desalination report , https:// www.desalination.com/publications/catalogue/ida-handbook.. [48] Wendorf M. The fresh water crisis and desalination plants. Interesting engineering, ,https://amp.interestingengineering.com/the-fresh-water-crisis-and-desalination-plants.; [April 2019]. [49] Brandt M. J., et al. The IDA Water Security Handbook 2019
2020, Water desalination report , https://www.desalination.com/publications/catalogue/ida-handbook.. [50] McGovern RK, et al. Performance limits of zero and single extraction humidificationdehumidification desalination systems. Appl Energy 2013;102:1081
90. [51] Shen J, et al. Analysis of a single-effect mechanical vapor compression desalination system using water injected twin screw compressors. Desalination 2014;333(1):146
53. [52] Lucier KJ, Qadir M. Gender and community mainstreaming in fog water collection systems. Water 2018;10(10):1472. [53] Qadir M, et al. Fog water collection: challenges beyond technology. Water 2018;10 (4):372. [54] Subramani A, Jacangelo JG. Emerging desalination technologies for water treatment: a critical review. Water Res 2015;75:164
87. [55] Eltawil MA, Zhengming Z, Yuan L. Renewable energy powered desalination systems: technologies and economics-state of the art. In: Proceedings of the 12th international water technology conference. Alexandria, Egypt; 2008. [56] Shahzad MW, et al. Renewable energy-driven desalination hybrids for sustainability. Desalin Water Treat 2018;95. [57] ‘Desalination’, U.S. Department of energy, office of energy efficiency and renewable energy, 2019 ,https://www.energy.gov/sites/prod/files/2019/03/f61/Chapter%207.pdf . .

Renewable energy-driven desalination Chapter | 11

365

[58] Al-Karaghouli A, Kazmerski LL. Energy consumption and water production cost of conventional and renewable-energy-powered desalination processes. Renew Sustain Energy Rev 2013;24:343
56. [59] Moser M, et al. Renewable desalination: a methodology for cost comparison. Desalin Water Treat 2013;51(4
6):1171
89. [60] Papapetrou M, Biercamp C, Wieghaus M. Roadmap for the development of desalination powered by renewable energy: promotion for renewable energy for water production through desalination. Stuttgart: Fraunhofer Verlag; 2010. p. 79. [61] Shahzad MW, et al. Desalination with renewable energy: a 24 hours operation solution, in water and wastewater treatment. IntechOpen; 2019. [62] Mehmood A, et al. Performance evaluation of solar water heating system with heat pipe evacuated tubes provided with natural gas backup. Energy Rep 2019;5:1432
44. [63] Rizzuti L, Ettouney HM, Cipollina A. Solar desalination for the 21st century: a review of modern technologies and researches on desalination coupled to renewable energies. Netherlands: Springer Science & Business Media; 2007. [64] Ghaffour N, et al. Renewable energy-driven desalination technologies: a comprehensive review on challenges and potential applications of integrated systems. Desalination 2015;356:94
114. [65] Mito MT, et al. Reverse osmosis (RO) membrane desalination driven by wind and solar photovoltaic (PV) energy: state of the art and challenges for large-scale implementation. Renew Sustain Energy Rev 2019;112:669
85. [66] Abdelkareem MA, et al. Recent progress in the use of renewable energy sources to power water desalination plants. Desalination 2018;435:97
113. [67] Helal A, Al-Malek S, Al-Katheeri E. Economic feasibility of alternative designs of a PVRO desalination unit for remote areas in the United Arab Emirates. Desalination 2008;221 (1
3):1
16. [68] Thomson M, Infield D. Laboratory demonstration of a photovoltaic-powered seawater reverse-osmosis system without batteries. Desalination 2005;183(1
3):105
11. [69] Thomson M, Infield D. A photovoltaic-powered seawater reverse-osmosis system without batteries. Desalination 2003;153(1
3):1
8. [70] Manolakos D, et al. Technical and economic comparison between PV-RO system and RO-solar Rankine system. Case study: Thirasia island. Desalination 2008;221 (1
3):37
46. [71] Mohamed ES, et al. A direct coupled photovoltaic seawater reverse osmosis desalination system toward battery based systems
a technical and economical experimental comparative study. Desalination 2008;221(1
3):17
22. [72] Bilton AM, Kelley LC, Dubowsky S. Photovoltaic reverse osmosis
feasibility and a pathway to develop technology. Desalin Water Treat 2011;31(1
3):24
34. [73] Soric A, et al. Eausmose project desalination by reverse osmosis and batteryless solar energy: design for a 1 m3 per day delivery. Desalination 2012;301:67
74. [74] Clarke DP, Al-Abdeli YM, Kothapalli G. The effects of including intricacies in the modelling of a small-scale solar-PV reverse osmosis desalination system. Desalination 2013;311:127
36. [75] Kelley LC, Dubowsky S. Thermal control to maximize photovoltaic powered reverse osmosis desalination systems productivity. Desalination 2013;314:10
19. [76] Kumarasamy S, Narasimhan S, Narasimhan S. Optimal operation of battery-less solar powered reverse osmosis plant for desalination. Desalination 2015;375:89
99.

366

PART | III Applications

[77] Ntavou E, et al. Experimental evaluation of a multi-skid reverse osmosis unit operating at fluctuating power input. Desalination 2016;398:77
86. [78] Forstmeier M, Feichter W, Mayer O. Photovoltaic powered water purification
challenges and opportunities. Desalination 2008;221(1
3):23
8. [79] Bilton AM, et al. On the feasibility of community-scale photovoltaic-powered reverse osmosis desalination systems for remote locations. Renew Energy 2011;36(12):3246
56. [80] Richards BS, Scha¨fer A. Design considerations for a solar-powered desalination system for remote communities in Australia Desalination 2002;144 (1-3):193
199. ,https://doi. org/10.1016/S0011-9164(02)00311-9.. [81] Cheah S-F. Photovoltaic reverse osmosis desalination system. 2004: US Department of the Interior, Bureau of Reclamation, Denver Office. [82] Ahmad N, et al. Modeling, simulation and performance evaluation of a community scale PVRO water desalination system operated by fixed and tracking PV panels: a case study for Dhahran city, Saudi Arabia. Renew Energy 2015;75:433
47. [83] Almarshoud A, Adam E. Towards VLS-PV deployment in Saudi Arabia: challenges, opportunities and recommendations. Energy Policy 2018;114:422
30. [84] Zaihidee FM, et al. Dust as an unalterable deteriorative factor affecting PV panel’s efficiency: why and how. Renew Sustain Energy Rev 2016;65:1267
78. [85] Moharram K, et al. Influence of cleaning using water and surfactants on the performance of photovoltaic panels. Energy Convers Manag 2013;68:266
72. [86] Syafiq A, et al. Advances in approaches and methods for self-cleaning of solar photovoltaic panels. Sol Energy 2018;162:597
619. [87] Vyas H, et al. Modus operandi for maximizing energy efficiency and increasing permeate flux of community scale solar powered reverse osmosis systems. Energy Convers Manag 2015;103:94
103. [88] Hossam-Eldin A, El-Nashar A, Ismaiel A. Investigation into economical desalination using optimized hybrid renewable energy system. Int J Electr Power Energy Syst 2012;43 (1):1393
400. [89] Mokheimer EM, et al. Modeling and optimization of hybrid wind
solar-powered reverse osmosis water desalination system in Saudi Arabia. Energy Convers Manag 2013;75:86
97. [90] Al-Bastak NM, Abbas A. Periodic operation of a reverse osmosis water desalination unit, Separation Science & Technology 1998;33(16):2531
40. ,https://doi.org/10.1080/ 01496399808545317 . [91] Al-Bastaki N, Abbas A. Use of fluid instabilities to enhance membrane performance: a review. Desalination 2001;136(1
3):255
62. [92] Latorre FJG, B´aez SOP, Gotor AG. Energy performance of a reverse osmosis desalination plant operating with variable pressure and flow. Desalination 2015;366:146
53. [93] Cabrera P, et al. Wind-driven SWRO desalination prototype with and without batteries: a performance simulation using machine learning models. Desalination 2018;435:77
96. [94] Karavas C-S, et al. A novel autonomous PV powered desalination system based on a DC microgrid concept incorporating short-term energy storage. Sol Energy 2018;159:947
61. [95] Ganora D, et al. An assessment of energy storage options for large-scale PV-RO desalination in the extended Mediterranean region. Sci Rep 2019;9(1):1
10. [96] Garcia-Rodriguez L. Seawater desalination driven by renewable energies: a review. Desalination 2002;143(2):103
13. [97] Hamed OA, et al. Concentrating solar power for seawater thermal desalination. Desalination 2016;396:70
8.

Renewable energy-driven desalination Chapter | 11

367

[98] Delyannis E-E. Status of solar assisted desalination: a review. Desalination 1987;67:3
19. [99] Belessiotis V, Delyannis E. Water shortage and renewable energies (RE) desalination
possible technological applications. Desalination 2001;139(1
3):133
8. [100] Delyannis E. Historic background of desalination and renewable energies. Sol energy 2003;75(5):357
66. [101] Garc´ıa-Rodr´ıguez L. Renewable energy applications in desalination: state of the art. Sol energy 2003;75(5):381
93. [102] Garc´ıa-Rodr´ıguez L. Assessment of most promising developments in solar desalination, in solar desalination for the 21st century. Berlin: Springer; 2007. p. 355
69. [103] Tao P, et al. Solar-driven interfacial evaporation. Nat energy 2018;3(12):1031
41. [104] Wang P. Emerging investigator series: the rise of nano-enabled photothermal materials for water evaporation and clean water production by sunlight. Environ Sci Nano 2018;5 (5):1078
89. [105] Pouyfaucon AB, Garc´ıa-Rodr´ıguez L. Solar thermal-powered desalination: a viable solution for a potential market. Desalination 2018;435:60
9. [106] Belessiotis V, Delyannis E. The history of renewable energies for water desalination. Desalination 2000;128(2):147
59. [107] Subiela VJ, et al. Canary Islands Institute of Technology (ITC) experiences in desalination with renewable energies (1996
2008). Desalin Water Treat 2009;7(1
3):220
35. [108] Li C, Goswami Y, Stefanakos E. Solar assisted sea water desalination: a review. Renew Sustain Energy Rev 2013;19:136
63. [109] Garc´ıa-Rodr´ıguez L. Current trends and future prospects of renewable energy-driven desalination (RE-DES). In: Renewable energy technologies for water desalination. CRC Press; 2017. p. 261
78. [110] Shalaby S. Reverse osmosis desalination powered by photovoltaic and solar Rankine cycle power systems: a review. Renew Sustain Energy Rev 2017;73:789
97. [111] Bae K, et al. Flexible thin-film black gold membranes with ultrabroadband plasmonic nanofocusing for efficient solar vapour generation. Nat Commun 2015;6(1):1
9. [112] Zhou L, et al. Self-assembly of highly efficient, broadband plasmonic absorbers for solar steam generation. Sci Adv 2016;2(4):e1501227. [113] Zhou L, et al. 3D self-assembly of aluminium nanoparticles for plasmon-enhanced solar desalination. Nat Photonics 2016;10(6):393. [114] Zielinski MS, et al. Hollow mesoporous plasmonic nanoshells for enhanced solar vapor generation. Nano Lett 2016;16(4):2159
67. [115] Ye M, et al. Synthesis of black TiOx nanoparticles by Mg reduction of TiO2 nanocrystals and their application for solar water evaporation. Adv Energy Mater 2017;7(4):1601811. [116] Yi L, et al. Scalable and low-cost synthesis of black amorphous Al-Ti-O nanostructure for high-efficient photothermal desalination. Nano Energy 2017;41:600
8. [117] Zhang L, et al. Plasmonic heating from indium nanoparticles on a floating microporous membrane for enhanced solar seawater desalination. Nanoscale 2017;9(35):12843
9. [118] Ren H, et al. Hierarchical graphene foam for efficient omnidirectional solar
thermal energy conversion. Adv Mater 2017;29(38):1702590. [119] Shi Y, et al. A 3D photothermal structure toward improved energy efficiency in solar steam generation. Joule 2018;2(6):1171
86. [120] Shi Y, et al. A robust CuCr2O4/SiO2 composite photothermal material with underwater black property and extremely high thermal stability for solar-driven water evaporation. Adv Sustain Syst 2018;2(3):1700145.

368

PART | III Applications

[121] Zhu M, et al. Plasmonic wood for high-efficiency solar steam generation. Adv Energy Mater 2018;8(4):1701028. [122] Gordon JM, Chua HT. The merits of plasmonic desalination. Nat Photonics 2017;11 (2):70. [123] Chiavazzo E, et al. Passive solar high-yield seawater desalination by modular and lowcost distillation. Nat Sustain 2018;1(12):763
72. [124] Xue G, et al. Highly efficient water harvesting with optimized solar thermal membrane distillation device. Glob Chall 2018;2(5
6):1800001. [125] Al-Othman A, et al. Nuclear desalination: a state-of-the-art review. Desalination 2019;457:39
61. [126] International Atomic Energy Agency. Desalination of water using conventional and nuclear energy. Technical Reports Series No. 24, Vienna, 1964. [127] Mansouri NY, Ghoniem AF. Does nuclear desalination make sense for Saudi Arabia? Desalination 2017;406:37
43. [128] International Atomic Energy Agency. New technologies for seawater desalination using nuclear energy. Technical Report Series No. 1753, Vienna, 2015. [129] Faibish RS, Konishi T. Nuclear desalination: a viable option for producing freshwater. Desalination 2003;157(1
3):241
52. [130] Khamis I. A global overview on nuclear desalination. Int J Nucl Desalin 2009;3 (4):311
28. [131] Tian L, Tang Y, Wang Y. Economic evaluation of seawater desalination for a nuclear heating reactor with multi-effect distillation. Desalination 2005;180(1
3):53
61. [132] Oak Ridge National Laboratory. Middle East study: application of large water-producing energy centers-the study area, ORNL-4481-Vol. I, USA, 1971. [133] Megahed MM. Nuclear desalination: history and prospects. Desalination 2001;135 (1
3):169
85. [134] Faibish RS, Ettouney H. MSF nuclear desalination. Desalination 2003;157 (1
3):277
87. [135] Wu S. Analysis of water production costs of a nuclear desalination plant with a nuclear heating reactor coupled with MED processes. Desalination 2006;190(1
3):287
94. [136] Hanra M. Desalination of seawater using nuclear heat. Desalination 2000;132 (1
3):263
8. [137] Al-Mutaz IS. Coupling of a nuclear reactor to hybrid RO-MSF desalination plants. Desalination 2003;157(1):259
68. [138] Ansari K, Sayyaadi H, Amidpour M. Thermoeconomic optimization of a hybrid pressurized water reactor (PWR) power plant coupled to a multi effect distillation desalination system with thermo-vapor compressor (MED-TVC). Energy 2010;35(5):1981
96. [139] Chen J, et al. Operations optimization of nuclear hybrid energy systems. Nucl Technol 2016;195(2):143
56. [140] Dincer S, Dincer I. Comparative evaluation of possible desalination options with various nuclear power plants. In: Exergetic, energetic and environmental dimensions. Elsevier; 2018. p. 569
82. [141] Khamis I, El-Emam R. IAEA coordinated research activity on nuclear desalination: the quest for new technologies and techno-economic assessment. Desalination 2016;394:56
63. [142] Khan SU-D, Khan SU-D. Karachi Nuclear Power Plant (KANUPP): as case study for techno-economic assessment of nuclear power coupled with water desalination. Energy 2017;127:372
80.

Renewable energy-driven desalination Chapter | 11

369

[143] Misra B. Seawater desalination using nuclear heat/electricity
prospects and challenges. Desalination 2007;205(1
3):269
78. [144] Pioro I, et al. Introduction: a survey of the status of electricity generation in the world. In: Handbook of generation IV nuclear reactors. Elsevier; 2016. p. 1
34. [145] Humphries J, Middleton E. Candesal: a Canadian nuclear desalination system. Desalination 1994;99(2
3):345
65. [146] Tewari P, Rao I. LTE desalination utilizing waste heat from a nuclear research reactor. Desalination 2002;150(1):45
9. [147] Tewari PK, Misra BM, Technological innovations in desalination, BARC Newsl. 1
7. ,http://www.barc.gov.in/publications/nl/2001/200109-01.pdf.; 2001. [148] International Atomic Energy Agency. Optimization of the coupling of nuclear reactors with desalination systems. Vienna: IAEA; 2005. [149] Chang M, Hwang Y. Coupling of MED
TVC with SMART for nuclear desalination. Int J Nucl Desalin 2003;1(1):69
80. [150] Chang MH, Kim S. Approach for SMART application to desalination and power generation. Advisory group meeting on materials and equipment for the coupling interfaces of nuclear reactors with desalination and district heating plants, Vienna. ,http://www.iaea. org/inis/collection/NCLCollectionStore/_Public/29/067/29067717.pdf.; 1998. [151] Zverev K, et al. Status and prospect of R&D aimed at application of nuclear reactors for seawater desalination in Russia. Int J Nucl Desalin 2004;1(3):281
97. [152] Abdallah AAHA. Innovative development of nuclear desalination technologies and cost improvement approaches. Ann Fac Eng Hunedoara 2018;16(2):31
41. [153] Nisan S, Dardour S. Economic evaluation of nuclear desalination systems. Desalination 2007;205(1
3):231
42. [154] Haijun J, Yajun Z. Nuclear seawater desalination plant coupled with 200 MW heating reactor. in: International Symposium on the Peaceful Applications of Nuclear Technology in the GCC Countries, Nov 3
5, 2008. ,https://www.kau.edu.sa/Files/320/ Researches/47532_18975.pdf . . [155] Megahed MM. Feasibility of nuclear power and desalination on El-Dabaa site. Desalination 2009;246(1
3):238
56. [156] Tabet M. Prospects of nuclear desalination in Morocco. Int J Nucl Desalin 2005;1 (4):411
15. [157] Udiyani P, Husnayani I, Sunaryo G. Analysis of radiation safety for small modular reactor (SMR) on PWR-100 MWe type. In: Journal of Physics: Conference Series. IOP Publishing; 2018. [158] Alessandroni C, et al. Safety aspects of nuclear desalination with innovative systems. The EURODESAL Project, International Congress on Advances in Nuclear Power Plants; Hollywood, FL, United States, 2002, ISBN 0-89448-663-2. ,https://inis.iaea.org/ search/search.aspx?orig_q 5 RN:40044700.. [159] Nisan S, et al. Sea-water desalination with nuclear and other energy sources: the EURODESAL project. Nucl Eng Des 2003;221(1
3):251
75. [160] Gude VG, Nirmalakhandan N, Deng S. Renewable and sustainable approaches for desalination. Renew Sustain Energy Rev 2010;14(9):2641
54. [161] Ma Q, Lu H. Wind energy technologies integrated with desalination systems: review and state-of-the-art. Desalination 2011;277(1
3):274
80. [162] Mahmoudi H, et al. Assessment of wind energy to power solar brackish water greenhouse desalination units: a case study from Algeria. Renew Sustain Energy Rev 2009;13 (8):2149
55.

370

PART | III Applications

[163] Garcia-Rodriguez L, Romero-Ternero V, Go´mez-Camacho C. Economic analysis of wind-powered desalination. Desalination 2001;137(1
3):259
65. [164] Weiner D, et al. Operation experience of a solar-and wind-powered desalination demonstration plant. Desalination 2001;137(1
3):7
13. [165] Veza JM, Pen˜ate B, Castellano F. Electrodialysis desalination designed for wind energy (on-grid tests). Desalination 2001;141(1):53
61. [166] de la Nuez Pestana I, et al. Optimization of RO desalination systems powered by renewable energies. Part I: wind energy. Desalination 2004;160(3):293
9. [167] Nakatake Y, Tanaka H. A new maritime lifesaving distiller driven by wind. Desalination 2005;177(1
3):31
42. [168] Forstmeier M, et al. Feasibility study on wind-powered desalination. Desalination 2007;203(1
3):463
70. [169] Fadigas EAFA, Dias J. Desalination of water by reverse osmosis using gravitational potential energy and wind energy. Desalination 2009;237(1
3):140
6. [170] Ali E, Ajbar A, Boumaza M. Performance assessment of a wind driven membrane desalination unit in Saudi Arabia. J Eng Res 2017;5(2). [171] Alsairafi A, Al-Shehaima M. Wind driven reverse osmosis desalination for concrete factory application in Kuwait. J Clean Energy Technol 2016;4(2). [172] Lai W, et al. Effects of wind intermittence and fluctuation on reverse osmosis desalination process and solution strategies. Desalination 2016;395:17
27. [173] Smits, R., Analysis of a wind driven reverse osmosis desalination system: experimental study using a pressure exchanger energy recovery device. Delft University of Technology, Master’s thesis, 2019. ,http://resolver.tudelft.nl/uuid:94973538-dd8f-4a938ddd-bb6c7b04522e . [174] Maleki A, Khajeh MG, Rosen MA. Weather forecasting for optimization of a hybrid solar
wind-powered reverse osmosis water desalination system using a novel optimizer approach. Energy 2016;114:1120
34. [175] Wu B, et al. Optimal design of stand-alone reverse osmosis desalination driven by a photovoltaic and diesel generator hybrid system. Sol Energy 2018;163:91
103. [176] Zhang G, et al. Simulated annealing-chaotic search algorithm based optimization of reverse osmosis hybrid desalination system driven by wind and solar energies. Sol Energy 2018;173:964
75. [177] Barbier E. Geothermal energy technology and current status: an overview. Renew Sustain Energy Rev 2002;6(1
2):3
65. [178] Goosen M, Mahmoudi H, Ghaffour N. Water desalination using geothermal energy. Energies 2010;3(8):1423
42. [179] Зуй, В.И., Great expectations for geothermal to 2100
messages for now, GRC Transactions 2011;35:1175
1183 ,https://www.isor.is/sites/isor.is/files/Great%20Expectations%20for% 20Geothermal%20to%202100%20-%20Messages%20for%20Now_0.pdf.. [180] Li K, et al. Comparison of geothermal with solar and wind power generation systems. Renew Sustain Energy Rev 2015;42:1464
74. [181] Gude VG. Geothermal source potential for water desalination
current status and future perspective. Renew Sustain Energy Rev 2016;57:1038
65. [182] Gude VG. Geothermal source for water desalination
challenges and opportunities. In: Renewable energy powered desalination handbook. Elsevier; 2018. p. 141
76. [183] Chamorro CR, et al. World geothermal power production status: energy, environmental and economic study of high enthalpy technologies. Energy 2012;42(1):10
8.

Renewable energy-driven desalination Chapter | 11

371

[184] Sanner B. Geothermal energy opportunities for desert regions. In: Proceedings of the global conference on renewable energy approaches for desert regions [GCREADER]. Le Royal Hotel Amman, Jordan; 2006. [185] Bouguecha S, Dhahbi M. Fluidised bed crystalliser and air gap membrane distillation as a solution to geothermal water desalination. Desalination 2003;152(1
3):237
44. [186] Mohamed A, El-Minshawy N. Humidification
dehumidification desalination system driven by geothermal energy. Desalination 2009;249(2):602
8. [187] Gutie´rrez H, Esp´ındola S. Using low enthalpy geothermal resources to desalinate sea water and electricity production on desert areas in Mexico. GHC Bull 2010;29:19
24. [188] Mahmoudi H, et al. Application of geothermal energy for heating and fresh water production in a brackish water greenhouse desalination unit: a case study from Algeria. Renew Sustain Energy Rev 2010;14(1):512
17. [189] Davies PA, Orfi J. Self-powered desalination of geothermal saline groundwater: technical feasibility. Water 2014;6(11):3409
32. [190] Awerbuch L, et al. Geothermal energy recovery process. Desalination 1976;19 (1
3):325
36. [191] Boegli W, Suemoto S, Trompeter K. Geothermal desalting at the East Mesa test site. Desalination 1977;22(1
3):77
90. [192] Ehmann H, Cendagorta M. Mediterranean conference on renewable energy sources for water production. Santorini, Greece: European Commission, EURORED Network, CRES, EDS; 1996. p. 10
2. [193] Bouchekima B. Solar desalination plant for small size use in remote arid areas of South Algeria for the production of drinking water. Desalination 2003;156(1
3):353
4. [194] Gude VG. Energy consumption and recovery in reverse osmosis. Desalination Water Treat 2011;36(1
3):239
60. [195] Missimer TM, Ghaffour N, Ng KC. Geothermal energy/desalination concepts. In: Renewable energy technologies for water desalination. CRC Press; 2017. p. 107
30. [196] Mohammadi A, Mehrpooya M. Energy and exergy analyses of a combined desalination and CCHP system driven by geothermal energy. Appl Therm Eng 2017;116:685
94. [197] Ozbey-Unal B, et al. Boron removal from geothermal water by air gap membrane distillation. Desalination 2018;433:141
50. [198] Capocelli M, et al. Reuse of waste geothermal brine: process, thermodynamic and economic analysis. Water 2020;12(2):316. [199] Rodr´ıguez-Girone´s M, Perez J, Veza J. A systematic approach to desalination powered by solar, wind and geothermal energy sources. In: Proceedings of the Mediterranean conference on renewable energy sources for water production. Santorini, Greece: European Commission, EURORED Network, CRES, EDS; 1996. [200] Bourouni K, Deronzier JC, Tadrist L. Experimentation and modelling of an innovative geothermal desalination unit. Desalination 1999;125(1
3):147
53. [201] Sephton Water Technology. VTE geothermal desalination pilot/demonstration project. Project summary. ,http://www.sephtonwatertech.com/DocumentsPDF/VTE_Geothermal_ Desalination_Project_Summary_2012_02_05.pdf. [accessed 02.2020]. [202] Franzitta V, et al. The desalination process driven by wave energy: a challenge for the future. Energies 2016;9(12):1032. [203] Hicks DC, et al. Delbouy: ocean wave-powered seawater reverse osmosis desalination systems. Desalination 1989;73:81
94.

372

PART | III Applications

[204] Verduzco-Zapata M, et al. Development of a wave-powered desalination device: numerical modelling. In: Proceedings of the 12th European wave and tidal energy conference, 2017. [205] Verduzco-Zapata MG, et al. Development of a desalination system driven by low energy ocean surface waves. J Coast Res 2018;85(sp1):1321
5. [206] Davies P. Wave-powered desalination: resource assessment and review of technology. Desalination 2005;186(1
3):97
109. [207] Mani A, Kumaraswamy S, Kumar RS. Utilisation of ocean thermal energy for desalination of brackish water. In: Technical report. National Institute of Technology Madras; 2002. [208] Kumar RS, Mani A, Kumaraswamy S. Experimental studies on desalination system for ocean thermal energy utilisation. Desalination 2007;207(1
3):1
8. [209] Leijon J, Bostro¨m C. Freshwater production from the motion of ocean waves
a review. Desalination 2018;435:161
71. [210] Nolan G, Ringwood J. Control of a heaving buoy wave energy converter for potable water production, Irish Signal and Systems Conference, Dublin, June 28-30, 2006. ,http://eprints.nuim.ie/4422/1/JR_Control_Heaving_Buoy.pdf.. [211] Lekka A, Turner MC, Ringwood JV. A class of globally stabilising controllers for the control of wave energy devices for potable water production. In: 2012 IEEE international conference on control applications. IEEE; 2012. [212] Xu P, et al. Critical review of desalination concentrate management, treatment and beneficial use. Environ Eng Sci 2013;30(8):502
14. [213] Ghaffour N, Missimer TM, Amy GL. Technical review and evaluation of the economics of water desalination: current and future challenges for better water supply sustainability. Desalination 2013;309:197
207. [214] Root T. Desalination plants produce more waste brine than thought. National Geographic. ,https://www.nationalgeographic.com/environment/2019/01/desalinationplants-produce-twice-as-much-waste-brine-as-thought/.; 2019. [215] Garrick DE, Hanemann M, Hepburn C. Rethinking the economics of water: an assessment. Oxford Rev Econ Policy 2020;36(1):1
23. [216] Osorio-Aravena J, et al. Transition toward a fully renewable based energy system in Chile by 2050 across power, heat, transport and desalination sectors. Int J Sustain Energy Plan Manag 2020;25.

Part IV

Sustainability

This page intentionally left blank

Chapter 12

The environmental performance of hydrogen production pathways based on renewable sources Eskinder Demisse Gemechu and Amit Kumar Department of Mechanical Engineering, 10-263 Donadeo Innovation Centre for Engineering, University of Alberta, Edmonton, AB, Canada

Chapter Outline 12.1 Introduction 376 12.2 H2 production pathways and applications 377 12.2.1 Water electrolysis 379 12.2.2 Biomass to H2 380 12.3 Method 384 12.3.1 Life cycle assessment 384 12.3.2 Goal and scope definition 384 12.3.3 Inventory analysis of windbased water electrolysis 385 12.3.4 Inventory analysis of solarbased water electrolysis 386 12.3.5 Inventory analysis of the thermal gasification of biomass 386 12.3.6 Inventory analysis of bio-oil reforming 391 12.3.7 Inventory analysis of supercritical water gasification of algae 392

12.3.8 Sensitivity and uncertainty analyses 394 12.4 Greenhouse gas footprints of H2 production pathways 395 12.4.1 Greenhouse gas footprint of water electrolysis 395 12.4.2 Greenhouse gas footprint of gasification 395 12.4.3 Greenhouse gas footprint of bio-oil reforming 397 12.4.4 Greenhouse gas footprint of supercritical water gasification 398 12.4.5 Comparative assessment incorporating sensitivity and uncertainty analyses 399 12.5 Conclusions 401 Acknowledgements 401 References 401

Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00012-1 © 2021 Elsevier Inc. All rights reserved.

375

376

PART | IV Sustainability

12.1 Introduction The scientific consensus on the existence of global warming and the attribution to human activities has been increasing [1
3]. Studies provide strong evidence that the recently observed changes in climate are anthropogenic [4
6]. Greenhouse gas (GHG) emissions, mainly driven by a fossil fuelbased global economy, are the main contributor [7]. The international community acknowledged the need for collective and ambitious efforts to combat climate change by responding to the Paris Agreement, a global agreement that aims to limit the average earth temperature rise well below 2 C above preindustrial levels [8]. Achieving such an ambitious target requires the rebuilding of the global energy system through a transition from a fossil fuel-based to a low-carbon economy. This could be accomplished through the combined effect of an increase in energy efficiency, carbon capture and storage/use, switching to zero emission energy carriers and the use of renewables such as biomass, geothermal, solar and wind to replace fossil fuels [9]. Global electrification with 100% renewable sources is achievable and economically viable in 2050, but it requires evolutionary changes [10]. Hydrogen (H2) could play a critical role in the transition to a low-carbon global economy [9,11,12]. An alternative to carbon-intensive fuels, H2 could provide a sustainable solution for a range of applications [12]. H2 from renewable electricity sources such as hydro, solar and wind can help to store and distribute clean energy in buildings and in the heating and power and transportation sectors where decarbonization is otherwise difficult [9]. Total global industrial H2 use reached 7.7 exajoules (EJ) in 2015 and demand is projected to rise to 10 EJ by 2050 [9]. As a clean fuel feedstock, H2 helps decarbonize energy and GHG emission-intensive industries. Ammonia production, oil refineries and methanol production accounted for 3.9, 2.4 and 0.6 EJ, respectively, in 2015 [9]. Yet around 96% of global H2 production is from fossil fuel-based feedstocks. Natural gas steam methane reforming took the largest share (48%), while coal and oil gasification processes together contributed 48% [13]. The share from renewable-based technologies such as water electrolysis was only 4% [13]. However, there is growing interest in developing and deploying technologies for renewablebased H2 production. Depending on the source of feedstock and energy and the conversion technology, the environmental benefit from H2 production and use can vary significantly [14
16]. Hence, it is important to systematically quantify the environmental performance of alternative technologies. This can be done through life cycle assessment, a widely used environmental assessment tool. There are many life-cycle assessments (LCAs) on H2 production; most of them are comparative assessments of different conversion technologies [17
23]. However, discrepancies in methods, system boundaries and other factors affect the robustness of the results, hence making a quality decision

The environmental performance of hydrogen production Chapter | 12

377

is challenging [24,25]. In this context, the focus of this chapter is to define a common system boundary to evaluate and compare the environmental performance of H2 production from renewable sources [wind-H2, solar-H2, bio-H2 (via gasification and bio-oil reforming) from straw, whole tree and forest residue and supercritical water gasification (SCWG) of algae]. The specific objectives are as follows: G

G G

G

To establish a common system boundary for nine H2 production pathways, To develop a bottom-up process-specific LCA model, To evaluate and compare the life cycle GHG emission performances of H2 production from nine pathways and To perform sensitivity and uncertainty analyses to identify the key parameters that influence the overall results and to provide the range of possible values for each pathway.

The chapter has five sections. The second section presents a general description of each renewal pathway considered. The third section discusses the LCA framework, that is, it defines the goal and scope, the main life cycle stages included for each pathway and the key data and assumptions used. The fourth section presents and discusses the main results and the last section draws the conclusion.

12.2 H2 production pathways and applications Fig. 12.1 shows different H2 production, distribution and use pathways both from renewable and nonrenewable energy sources. Depending on the conversion technology, H2 production is broadly categorized as thermal, electrolysis and photolytic processes. In thermal processes, feedstocks from both fossil fuel (coal, natural gas and oil) and renewable (biomass) sources are used as energy sources to break down the feedstock and generate the H2. Steam methane reforming of natural gas is the traditional H2 conversion pathway from fossil-based fuels. H2 can also be produced from coal gasification or petroleum oil partial oxidation. Coal gasification is the partial combustion of coal to produce syngas [a mixture of carbon dioxide (CO2), carbon monoxide (CO) and H2] at high pressure and temperature (at or above 700 C) [26]. The process is characterized by its faster reaction but higher cost than steam methane reforming of natural gas. In the electrolytic process, electricity is used to split water into H2 and oxygen. The environmental performance of the electrolytic process is highly dependent on the electricity source (i.e. hydro, nuclear, solar, wind). In the photolytic process, light is used to split water into H2 and oxygen. Biological or photobiological conversion pathways such as the biophotolysis of water using algae, biological water
gas shift reaction and photofermentation are promising renewable H2 technologies [27]. Because biological or photobiological technologies operate at

378

PART | IV Sustainability

FIGURE 12.1 H2 production alternative pathways.

ambient temperature and pressure, they require less energy for conversion [27]. Solar- and wind-based water electrolysis, biomass gasification, bio-oil reforming and SCWG are discussed in this section. As depicted in Fig. 12.1, there are several key application areas for H2 in a low-carbon economy: transportation, heat and power and industry [9,11,12]. Along with biofuel and electrification, H2 can help decarbonize the transportation sector, for example in a fuel cell for transportation both on and off road. The heating sector is among the energy-intensive sectors that need to be decarbonized. There have been different proposals made to lower the GHG emissions from the sector such as reducing the heat demand through high-efficiency devices, replacing natural gas with low-carbon fuel, electrification, district heating and the use of on-site renewable energy sources. H2 has the potential to decarbonize the heating sector, for example by replacing conventional natural gas boilers and furnaces with H2, fuel cell combined heat and power to coproduce heat and electricity and gas-driven heat pumps. In the power sector, H2 technology enables utility-scale integration of renewable electricity generation storage and distribution. H2 is used as a clean feedstock to decarbonize the ammonia, methanol, refining and steel industries.

The environmental performance of hydrogen production Chapter | 12

379

12.2.1 Water electrolysis Electrolysis is the process of splitting water into H2 as the main product and oxygen as the byproduct using electricity. H2 production from water electrolysis provides an energy storage solution for the utility-scale deployment of intermittent renewable energy sources such as solar and wind [28,29]. H2 has compelling characteristics that make it an excellent energy carrier alternative to fossil fuels [29]. Water electrolysis powered by renewable energy sources such as hydro, wind or solar is considered by many to be one of the promising renewable pathways [30,31]. It is economically competitive compared to technologies that use fossil fuel sources [32] in some jurisdictions, mainly as a result of the recent and projected decline in the cost of wind [33] and solar [34] energy. The electrolysis process occurs in an electrolyzer, which is comprised of electrolytes, an anode and a cathode. As presented in Fig. 12.2, depending on the type of electrolyte used, water electrolysis technology is broadly categorized as alkaline electrolysis (AEC), polymer electrolyte membrane electrolysis (PEM) or solid oxide electrolysis (SOEC). AEC is a wellestablished technique that has been widely used at the industrial scale since the 1920s [35,36]. A direct current is applied between the anode and the cathode immersed in a liquid alkaline solution. The two electrodes are separated by a diaphragm. Electrons flowing to the cathode are consumed by the hydrogen ion (H1) to form H2. The alkaline solution transports the hydroxide ions (OH) from the cathode to the anode. AEC technology is characterized by its

FIGURE 12.2 Water electrolysis process.

380

PART | IV Sustainability

low capital cost but low current density and limited operating range, which affect the capacity and associated cost of the H2 production [37]. PEM was developed to overcome some of the drawbacks of AEC technology [37]. PEM uses a solid sulfonated polystyrene membrane as polymer electrolyte. The polymer electrolyte offers a high proton conductivity, low gas crossover (high H2 purity), compact system design and high pressure and current density. PEM requires a novel metal catalyst that is highly corrosion resistant, which makes it more expensive than AEC [38]. Nevertheless, PEM can provide a more sustainable solution for future H2 production [31,38,39]. SOEC, also known as high temperature electrolysis, is less mature than AEC and PEM. SOEC uses zirconium dioxide (ZrO2) as an electrolyte. ZrO2 is a solid ceramic that operates at high temperature (700 C
900 C) with high conductivity and chemical stability [40]. These properties consequently lower the electricity demand of the SOEC system [41]. Material degradation and additional process requirements to remove the produced steam (a result of the high operating temperature) are the main challenges of SOEC technology [42].

12.2.2 Biomass to H2 Biomass is animal- or plant-based organic hydrocarbon. Because of its abundance, it is considered as significant source of renewable energy. Biomass includes agricultural and forest residues, organic municipal solid waste, sewage sludge and industrial waste [43,44]. The traditional use of biomass as a source of energy for cooking and heating accounted for 7.5% of global energy consumption in 2017 [45]. This use included burning woody biomass and agricultural residues in developing and emerging countries [45]. Biomass can also be converted to various energy forms such as electricity, thermal energy and transportation fuel through different energy conversion processes, some are mature and are at full commercial scale and others are in their early stage of development. Modern bioenergy contributed to around 5% of total global energy demand in 2017 [45]. Biomass can be used as a renewable feedstock to produce H2 via either biological or thermochemical conversion pathways (gasification and bio-oil reforming). In the thermochemical process, heat and chemical reactions are involved to produce H2 from biomass. Four biomass feedstocks for hydrogen production are discussed in this chapter: agricultural residue, forest residue, whole tree and algae. A case study for Canada has been performed. Agricultural residues are byproducts of agricultural crop processing such as rice husk, coconut and nutshells and straw from barley or wheat. Straw from wheat and barley production are among the main sources of agricultural residues in western Canada [44]. Whole tree biomass comes from a forest, usually one dedicated to pulp and lumber, and it can be used for energy production when there is low demand in the pulp and lumber sectors [44]. More than 60% of Alberta, a

The environmental performance of hydrogen production Chapter | 12

381

province in western Canada, is forested. Boreal-mixed forests account for 75% of the forested land in the province [46]. These forests are home to several species such as trembling aspen, white spruce, balsam poplar, lodgepole pine and paper birch [47]. The Government owns and controls most of the forest area, thus ensuring sustainable management practices [44], and the right to harvest is allocated to companies and individuals under the Forest Act [48]. In this study, it is assumed that excess forest capacity from pulp mills and lumber operations can be used for biomass processing [44]. Forest residues include tree bark and shrubs as well as wastes from forest logging and clearing operations such as wood chips, branches and treetops. Recovered tree limbs and tops from logging make up 15%
25% of the total forest biomass [44].

12.2.2.1 Thermal gasification Biomass gasification is a more advanced H2 production technology than steam reforming [49]. It is the partial combustion of biomass with a controlled amount of oxygen at high pressure and temperature (greater than 700 C) to produce syngas [50,51]. Biofuels, biochar, heat and power can also be produced in the absence of oxygen [50]. The process involves the use of gasifying media in the form of heat, oxygen and steam. Depending on the type of gasification agent used and the application of steam, different biomass gasification techniques are used. Steam biomass gasification is a well-established and efficient method of H2 production [51] as it provides higher H2 yield than air biomass gasification and avoids oxygen separation [49,52]. Fig. 12.3 illustrates a simplified process flow of steam biomass gasification. The process comprises drying wet biomass, pyrolysis, char gasification and the pyrolysis volatiles reaction. Low temperature, up to 150 C, is applied to evaporate the moisture in the wet biomass. The heat required for drying could be derived from other stages of the gasification process. The dried biomass undergoes thermal degradation through the pyrolysis process at a temperature between 200 C and 650 C. Char and pyrolysis volatiles (condensable and gases) are the main outputs. The char is gasified between 700 C and 1000 C to produce syngas [27]. Temperature and the steam-tobiomass ratio are the key parameters that influence the performance of biomass gasification [53,54]. Different reactor types (gasifiers) are used for biomass gasification. The nature of the contact between the feed and gasifying agent, heat transfer rate and residence time are the distinct features of each technology [55]. The entrained flow reactor, fixed bed, fluidized bed, rotary kiln reactor and plasma reactor are the main reactors used in the biomass gasification process [55]. Temperature affects char kinetics. For example increasing the temperature from 800 C to 900 C improves the char conversion residence time by a factor of six [56]. The hydrogen from the gasified biomass can be separated through processes such as membrane separation,

382

PART | IV Sustainability

FIGURE 12.3 An example of an updraft fixed bed reactor used in biomass gasification.

reforming and water
gas shift reaction. The water
gas shift reaction unit converts CO and steam to H2 and CO2. Pyrolysis volatiles are cracked or reformed to produce a more stable tar, a solid byproduct from biomass gasification [57]. The tar in the syngas poses a serious challenge as it could corrode downstream equipment and block pipelines [57,58]. Catalytic, mechanical and thermal techniques are used to enhance the cracking process to reduce the tar in produced gas [59
61]. Biomass gasification is generally considered one of the most cost-effective H2 production technologies [62].

12.2.2.2 Supercritical water gasification of biomass SCWG is a promising thermochemical technology that can be used to convert high moisture content biomass to syngas [63]. Water reaches supercritical conditions at 374.12 C and 221.2 bar above its critical point, at which the distinct liquid and gas phases do not exit. Supercritical water combines both the dissolution of liquid water and the diffusion of its gaseous phase. It

The environmental performance of hydrogen production Chapter | 12

383

is also characterized by high sensitivity towards changes in temperature, pressure and low dielectric constant, which make supercritical water to act as a solvent or catalyst depending on the conditions. Temperature, pressure, feedstock concentration (biomass-to-water ratio) and residence time are the key parameters that affect the H2 yield of the SCWG process [64]. The operating temperature ranges from 350 C to 700 C depending on the feedstock type, reactor configuration, catalyst and desired outputs [63,65]. SCWG offers several advantages over thermochemical gasification. Biomass with high moisture content can be gasified without drying [66] and hence reduces feedstock drying cost. High reaction rate, high H2 and low CO yields and low char and tar formation [23] are other advantages of SCWG. Unlike other thermochemical gasification techniques, water is used as a gasification reaction medium and catalyst in SCWG. Hydrolysis, steam reforming, water
gas shift reaction, methanization and hydrogenation are some of the subreactions in the SCWG process [67]. In steam reforming, the main biomass feedstocks and intermediate products from hydrolysis lignin such as phenolics are broken down to CO, CO2 and H2. The water
gas shift reaction in H2 production is between CO and water. Methanization and hydrogenation are secondary reactions that consume CO, CO2 and H2 to produce CH4. There are different reactor configurations for SCWG, i.e. batch, continuous stirred tank, tubular, fluidized bed and diamond anvil cell. Cellulosic [68,69], hemicellulosic [70,71] and ligneous [72] feedstock are among the widely investigated model compounds for SCWG.

12.2.2.3 Bio-oil reforming Reforming bio-oil produced from fast pyrolysis is an alternative renewable H2 production pathway [73]. Fast pyrolysis is the decomposition of biomass into bio-oil at high temperature (between 400 C and 500 C) in the absence of oxygen with a high heating rate and low residence time [74]. The bio-oil is characterized by its low heating value, high corrosiveness, high oxygen and water content and high viscosity [75]. Upgrading is required to improve the chemical and physical properties of the bio-oil. Alternative bio-oil treatment options exist [75,76], but catalytic steam reforming appears to be the most widely studied. The bio-oil undergoes steam reforming to produce either H2 or syngas. Temperature, space time (the ratio of the mass of the catalyst to the molar flow rate of bio-oil) and steam-to-carbon ratio are among the key parameters that affect H2 yield in bio-oil reforming [49]. Biooil reforming requires less energy (because of the low operational temperature requirement) than biomass gasification. The other advantage of bio-oil reforming is the lower transportation cost, which is due to the higher energy density characteristics of bio-oil compared to corresponding biomass feedstocks. Fig. 12.4 presents a simplified schematic of the bio-oil reforming process.

384

PART | IV Sustainability

FIGURE 12.4 A simplified diagram of bio-oil reforming.

12.3 Method 12.3.1 Life cycle assessment LCA is a stand-alone environmental performance assessment tool that involves the compilation and evaluation of material and energy inputs, outputs and the associated environmental impacts of a product system along its life cycle [77,78]. Here ‘life cycle’, according to ISO 14040, refers to ‘consecutive and interlinked stages of a product system, from raw material acquisition or generation from natural resources to final disposal’ [77,78]. LCA is based on the notion of life cycle thinking, a way of thinking that looks at a product system holistically and refers to the extraction of resources from nature, component production and manufacturing, and product distribution, use and final disposal. As a system and holistic approach, LCA attempts to understand the role of each life cycle stage and process on the overall environmental performance of the product system under investigation. LCA thus avoids any burden shifting among the life cycle stages or processes. LCA is a harmonized and standardized analytical tool that follows the principles and frameworks (ISO 14040:2006) [77] and requirements and guidelines (ISO 14044:2006) [78] of the International Organization for Standardization (ISO). The four phases of LCA framework are the goal and scope definition, inventory analysis, impact assessment and interpretation. Each phase is discussed in this section in the context of H2 production from the renewable energy sources considered in this chapter.

12.3.2 Goal and scope definition Goal and scope definition is the first stage of an LCA. The goal definition unambiguously sets the reasons to perform LCA, the audiences to whom the study outcome is communicated and the applications [77,78]. In LCA, the scope is defined in accordance with the goal statements. The scope definition involves describing each life cycle stage and process, specifying the main functions of the product system, defining the functional unit, setting the allocation procedure, determining the modelling approach used, identifying the data requirement and so on. The functional unit is a reference unit that clearly reflects the measurable performance of a product system. The functional unit is used as a reference to which all the input and output requirements and the associated

The environmental performance of hydrogen production Chapter | 12

385

environmental impacts are related. With the defined goal, a system boundary is used to specify the subprocesses or subsystems that are included and excluded in the assessment. Any exclusion of subprocesses/subsystems should be justified; that is their relative contribution to the overall impact should be minimal [77,78]. The main purpose of this chapter is to evaluate and compare the life cycle GHG emission performances of H2 production from renewable energy sources. Nine alternative pathways are assessed, namely, solar- and windbased water electrolysis; thermal gasification of agricultural residues, forest residues and whole trees; SCWG of algae and bio-oil reforming of agricultural residues, forest residues and whole trees. The functional unit is 1 kg of H2 produced from a specific pathway. The following sections provide the system boundary for each H2 production pathway along with process descriptions, the key data sources used in the compilation and quantification of energy and material inputs and the corresponding GHG emissions.

12.3.3 Inventory analysis of wind-based water electrolysis Fig. 12.5 shows the system boundary for western Canada-specific windbased water electrolysis. The life cycle includes material extraction, wind plant construction, electricity generation, electrolysis process, H2 production and distribution and plant decommissioning. The similar system boundary is valid for other jurisdictions with minor modification as needed.

FIGURE 12.5 System boundary diagram—wind-based water electrolysis.

386

PART | IV Sustainability

GHG emissions from wind energy generation are due to life cycle material and energy consumption in the production and transportation of components, operation and maintenance and plant end-of-life including metal recovery. The wind power plant data is from the Summer view 2 Wind Facility in southern Alberta [79]. The plant has a capacity of 66 megawatts (MW) (22 turbines each with 3 MW) and 80 m towers with 45 m blade lengths [79]. A capacity factor of 30% was considered. The water splitting is based on AEC technology, with an electricity-to-H2 efficiency of 74% at 70 C temperature and 15 bar pressure. A 15-year plant lifetime and a maximum H2 production rate of 60 Nm3/hour were considered. The electricity consumption per kg of H2 is 53 kWh. H2 is compressed using a compressor at 70% efficiency and a 22-year lifetime and transported by pipeline to a consumption facility assumed to be located 500 km away. The energy and material consumption for compressor manufacturing and operation were also considered (Table 12.1).

12.3.4 Inventory analysis of solar-based water electrolysis The life cycle stages of H2 production from solar-based water electrolysis are shown in Fig. 12.6. Electricity generation from a utility-scale solar power plant with a capacity of 5 MW is assessed [80]. The main components considered to estimate GHG emissions associated with electricity generation are solar photovoltaic (PV) production, system integration, operation and end-oflife. The solar PV production includes silica extraction and upgrading to solar-grade silicon, cell processing and module assembly. The system integration stage comprises mounting the structures, cables, inverter and transformer. Electricity generation is determined using Alberta’s monthly average solar insulation (186.3 kWh/m2), a performance ratio of 0.8, a 16.72% solar efficiency, a lifetime of 10 years and a total area of 1.95 m2 [80]. The parameters used to model energy and material requirements for the electrolysis process and H2 distribution are assumed to be similar to those considered in the wind case: the AEC water splitting, a 74% electricity-to-H2 efficiency, 53 kWh electricity consumption per kg of H2, and a pipeline length of 500 km. A detailed process description, model assumptions and data sources for a utility-scale solar power plant can be consulted in the paper by Mehedi et al. [80].

12.3.5 Inventory analysis of the thermal gasification of biomass Fig. 12.7 shows the system boundary for the biomass-to-H2 pathways. This section presents the feedstock production, transportation to the H2 production plant, gasification process and H2 distribution stages.

The environmental performance of hydrogen production Chapter | 12

387

TABLE 12.1 Inventory of material for wind-based water electrolysis. Wind turbine technical specifications Capacity (MW) (22 turbines)

66

Number of blades

3

Blade length (m)

45

Tower height (m)

80

Tower weight (tonne)

160

Material for plant construction (kg/kg H2) Aluminum and aluminum alloy

2.47E-3

Ceramic

2.92E-3

Concrete

2.55E-1

Copper

1.69E-3

Electronics

5.06E-4

Iron and steel

7.25E-2

Lubricants

3.93E-4

Plastics

4.04E-3

Electrolysis Energy consumption (kWh/kg H2)

53

Electricity-to-H2 conversion efficiency (%)

74

Plat lifetime (years)

15

Pressure (bar)

15 

Temperature ( C)

70

Energy and material consumption for H2 compression (per kg H2) Electricity (kWh)

7.05E-4

Aluminum (kg)

4.23E-5

Copper (kg)

3.17E-5

Iron and steel (kg)

1.76E-3

Lubricant (kg)

1.27E-5

Source: Adapted from Ghandehariun S, Kumar A. Life cycle assessment of wind-based hydrogen production in western Canada. Int J Hydrog. Energy 2016;41(22):9696
704.

12.3.5.1 Feedstock production Straw left over from a wheat grain harvest is the assumed source in our study. The main processes involved in wheat straw preparation are harvesting and fertilizer application. Harvesting includes swathing, raking, balling

388

PART | IV Sustainability

FIGURE 12.6 System boundary diagram—solar-based water electrolysis.

FIGURE 12.7 System boundary diagram
biomass gasification.

The environmental performance of hydrogen production Chapter | 12

389

and roadsiding. Diesel fuel consumption in each process is used to estimate corresponding GHG emissions. The removal of the residues from the soil affects the soil nutrient content; hence, we assume that synthetic fertilizer (nitrogen, phosphoric acid and potassium oxide) is applied to compensate for nutrient loss. Fuel consumption, capacity and lifetime data for each piece of equipment are summarized in Table 12.2. The GHG emissions for whole trees are mainly due to equipment fuel use in felling (cutting of the trees), skidding (dragging of the trees to the roadside) and chipping operations on the roadside. Emissions from material and energy use in equipment manufacturing, operations and final disposal were also considered. Nutrient content in forest biomass is used to estimate the amount of fertilizer required and the associated GHG emissions from fertilizer production and application. For each hectare, we assumed that 258, 41 and 125 kg of nitrogen, phosphorus and potassium were applied, respectively [81]. We also assumed a 100-year rotation time and a tree yield of 84 dry tonnes/hectare [44]. Nitrogen, phosphorus and potassium contents in the forest biomass were assumed to be 0.31%, 0.05% and 0.15%, respectively [44]. GHG emissions from fertilizer production are based on data from Johnson et al. [82]. A forest residue yield of 0.247 dry tonnes/hectare was calculated assuming that 20% of whole trees become residue over a 100-year rotation [44]. Forwarding and chipping operations are considered for forest residue production. A forwarder with diesel fuel use of 26 L/hour, productivity of 50 m3/ hour and a lifetime of 16,000 hours was considered [81]. For the chipper, a 110 L/hour diesel consumption rate, productivity of 28 dry tonne/hour and lifetime of 9000 hours were assumed [81]. Conventional diesel from the oil sands was assumed as the main source of fuel to run the forwarder and chipper. Table 12.2 summarizes the main characteristics of each feedstock and the inventory data used to estimate GHG emissions associated with fuel use for different operations.

12.3.5.2 Biomass transportation A high capacity trailer truck with a carrying capacity of 23 wet tonnes/trip and fuel efficiency of 6 miles/gallon (5 miles/gallon without load) were assumed to determine GHG emissions from biomass transportation. The distance between the biomass collection sites and biohydrogen production plants was estimated based on the net biomass yields and plant capacities [83]. For agricultural residues, the harvesting and baling area is assumed be square, with the H2 plant (gasification or bio-oil reforming) located at the center of the square. For forest residues and whole trees, the biomass harvesting area is assumed to be a circle with the plant positioned at the center [83
85]. Average distances of 112, 92 and 24 km between the collection

TABLE 12.2 Feedstock characteristics and equipment used in feedstock production. Biomass feedstock characteristics [44] Moisture content (wt.%)

Wet bulk density (kg/m3)

Fuel density (dry kg/m3)

High heating value (MJ/dry kg)

Biomass yield (dt/ha)

Ash content (%)

Agricultural residue

16

230

140

18

0.333

4

Whole tree

50

250

350

20

84

1

Forest residue

45

175

350

20

0.247

3

Fuel consumption rate (L/h)

Capacity/productivity

Lifetime (h)

Embodied energy factor (MJ/ha)

Harvester

43

12 (ha/h)

Raking

14

30 (dt/h)

Baling

58

20 (dt/h)

1500

Stacking

58

70 (dt/h)

20,000

Bale wrapper

3.3

60 (bales/h)

10,000

Bale loader

56

170 (dt/h)

10,000

Equipment used in feedstock production [81]

Agricultural residue

Whole tree

Forest residue

3

38

Feller

47

70 (m /h)

10,950

Chipper

100

30 (dt/h)

9000

3

Skidder

45

60 (m /h)

12,000

Forwarder

26

50 (m3/h)

16,000

Chipper

110

28 (dt/h)

9000

Source: Adapted from Kumar A, Cameron JB, Flynn PC. Biomass power cost and optimum plant size in western Canada. Biomass Bioenergy 2003;24(6):445
64; Adapted from Nagy CN. Energy coefficients for agriculture inputs in western Canada. Centre for Studies in Agriculture, Law and the Environment, University of Saskatchewan, 1999.

The environmental performance of hydrogen production Chapter | 12

391

sites and H2 production facilities were considered for agricultural residues, forest residues and whole trees, respectively [22]. Further details on these assumptions are available in an earlier study [22].

12.3.5.3 Gasification process For the conversion of biomass to H2, thermal gasification was considered. The process involves the following unit operations: biomass drying, syngas production and cleaning, water
gas shift reaction and the H2 purification process. Biomass feedstocks are dried to a moisture content of 12% using flue gas from char combustion. Steam is supplied at a rate of 0.4 kg/kg of biomass. A cyclone separator is used to remove the ash, a byproduct of gasification. We calculated GHG emissions based on the natural gas and electricity consumption from each unit operation in the conversion process. Around 1.14 and 1.50 kWh of electricity are required per kg H2 production for the forest residue and whole tree pathways, respectively [86]. The natural gas consumption for drying is assumed to be 14 MJ/kg of H2 [86]. Emissions from plant construction and maintenance were also considered. We assumed that the ash is used as soil nutrient compliment. A high capacity (23 wet tonnes) truck and a 50 km transportation distance were used to model the GHG emissions from ash disposal. The plant size for each feedstock type was determined based on the availability of each biomass type and location. An optimum capacity of 3000 dry tonnes/day was considered for all feedstock types [83,85]. A pipeline transportation distance of 500 km was considered in the delivery of the hydrogen to a H2 consumption facility. 12.3.6 Inventory analysis of bio-oil reforming Fig. 12.8 shows the system boundary considered for H2 production via biooil reforming. Three biomass feedstocks were used for this assessment: agricultural residues, forest residues and whole trees. Biomass production and transportation are discussed in Sections 3.5.1 and 3.5.2. The following are the key operating parameters for the fast pyrolysis biomass to bio-oil conversion process: 2
3 mm feedstock size, less than 10% moisture content, 425 C
500 C operating temperature, 50 W/cm2 heat transfer rate, and around 2 seconds residence time [84]. Heat for biomass drying and the pyrolysis reaction can be generated from char combustion and the use of noncombustible exhaust gas, respectively. Hence, there is no external energy requirement. Pyrolysis vapour, at 500 C, is cooled down to 20 C in the heat recovery steam generator to form bio-oil. 10 wt.% methanol was added to the bio-oil in order to reduce the viscosity of the bio-oil and stabilize it at room temperature [84]. The bio-oil and methanol mixture was transported to a reforming facility via high-density polyethylene pipeline. We assumed the

392

PART | IV Sustainability

FIGURE 12.8 System boundary diagram—bio-oil reforming.

reforming facility (oil sands upgrading facility) is located 500 km from the bio-oil production site [22,84].

12.3.7 Inventory analysis of supercritical water gasification of algae Fig. 12.9 shows the main processes involved in the conversion of algae biomass to H2 via SCWG.

12.3.7.1 Algae cultivation Microalgae are rich in carbohydrates, lipids and protein, which make them a promising renewable feedstock for biofuel production [87]. Open pond systems and photobioreactors (PBRs) are the most common microalgae cultivation methods [88]. Open pond systems are the oldest method. They operate in large water bodies (i.e. lakes, seas, wastewater and artificial ponds). Open pond systems offer high algae biomass yield and are the most cost-effective ways to cultivate algae [89
93]. In a PBR system, microalgae are cultivated in a closed and controlled environment. A PBR can operate indoors under artificial light or outdoors in sunlight. The controlled environment avoids

The environmental performance of hydrogen production Chapter | 12

393

FIGURE 12.9 H2 production via the supercritical water gasification of algae biomass.

contamination and also allows key cultivation parameters such as CO2 and nutrient concentrations, light intensity, pH level and temperature to be easily regulated [94]. Photobioreactor systems require higher capital and operational costs than open pond systems, but they are preferred in pharmaceutical and health applications of microalgae, in which high quality is desirable [94]. The algae production opportunities and challenges particular to Canada’s northern climate have been extensively studied by Pankratz et al. [95
97]. In this study, we assumed an open pond microalgae cultivation system with a production capacity of 2000 dry tonnes/day (which requires around 82,000 ha). Detailed design and operating parameters can be found in the studies by Pankratz et al. [95,98]. Microalgae require water, nutrients and sunlight to grow in an open pond system. In the study, diammonium phosphate and ammonia were used as sources of phosphorous and nitrogen, respectively. GHG emissions from microalgae cultivation are from the production of the corresponding fertilizer, direct land use change and direct energy consumption from the use of equipment in various operations [98]. The algae was assumed to be cultivated in a location close to the H2 production facilities; therefore transportation emissions were not included [98].

12.3.7.2 Process conversion H2 yield from the SCWG of biomass varies depending on the feedstock type and operational parameter, that is temperature, pressure, feed concentration,

394

PART | IV Sustainability

reaction time and reactor configuration [63]. A H2 yield of 10.5% was considered [23]. The prehydrolysis process breaks down the nonconventional components of the algae biomass; this process is followed by mineral separation, which removes salts at a temperature of around 380 C. The presence of salts may cause plugging and could lead to material corrosion in the long term. The supercritical water reactor operates at 600 C to produce syngas. The syngas undergoes operations using Selexol to remove sulfur and produce pure H2. The stream passes through the steam reforming unit and a water
gas shift reaction. A detailed process model description can be found in Kumar et al. [23].

12.3.8 Sensitivity and uncertainty analyses LCA has had a prominent role in the evaluation of the environmental performance of alternative product systems and in supporting decision-making. However, there are concerns among LCA users over the robustness of results. The system boundary, temporal and spatial variabilities, method, functional unit definition, differences in allocation and cut-off criteria are among the main sources of variability in LCA results [99]. Therefore it is important to perform sensitivity and uncertainty analyses to determine the validity of LCA results, especially in comparative assessments. Sensitivity analysis allows us to evaluate the sensitivity of results to selected parameters. LCA results are sensitive to those parameters such that a small variation in them will induce substantial changes in the overall results. Uncertainty analysis measures the variability of the output results by considering a range of inputs. It is important to quantify and present the sensitivity of the model output and uncertainty of the results to better interpret them. A Regression, Uncertainty and Sensitivity Tool (RUST), a spreadsheetbased model, developed by Di Lullo et al., was used for sensitivity and uncertainty analyses [100]. RUST uses the Morris method to screen the sensitive input parameters. The Morris method has a shorter processing time than other global sensitivity analyses; this is helpful in a model with large input parameters is run [101]. We used a Monte Carlo simulation using triangular distributions to quantify the probable range of GHG emissions values for each H2 production pathway. The grid electricity emissions factor, diesel emissions factor, fertilizer use, biohydrogen yield, bio-oil yield, feedstock transportation distance, hydrogen transportation distance, equipment energy consumption rate, methanol blend ratio and straw-to-grain ratio are among the key factors used in sensitivity analysis. Only the sensitive parameters from the Morris screening were considered for the uncertainty analysis. Because we assumed that the plant is located in Alberta (a western Canadian province), we used the grid electricity mix emissions factor for the year 2017, 713 g CO2 eq/kWh, in the calculation for Alberta. The emissions factors in 2010 and 2018, 851 and 654 g CO2 eq/kWh, the upper and lower

The environmental performance of hydrogen production Chapter | 12

395

ends, reflect the temporal variability in the electricity emissions factor [102]. For other jurisdiction, same methodology can be used with appropriate adjustment to the emissions factor. The emissions factor for diesel consumption was derived using the Canadian in situ bitumen extraction as reported by Di Lullo et al. [103
105], 118 g CO2 eq/MJ. The minimum and maximum possible values, 95
138 g CO2 eq/MJ, were used for sensitivity and uncertainty analyses [103
105]. A variation of 6 10% was considered for the other parameters. These could be adjusted as appropriate for other jurisdictions.

12.4 Greenhouse gas footprints of H2 production pathways 12.4.1 Greenhouse gas footprint of water electrolysis Fig. 12.10 presents the life cycle GHG emissions from solar and wind energy-based electrolysis. The global emissions from wind are estimated to be 0.69 kg CO2eq/ kg H2. The largest contribution is from wind electricity generation, which accounts for 64% of the life cycle GHG emissions. A significant portion of the electricity generation emissions is due to the energy and material requirements to construct the wind plant. H2 compression and transportation together account for 29% of the total GHG emissions. Emissions from plant operation and maintenance have a minimal contribution, only 7%. Solar-water electrolysis has 5.7 times higher GHG emissions than wind. Around 94% of the total GHG emissions are due to solar electricity generation. Solar PV panel production phase is the key driver, mainly due to energy-intensive processes such as the upgrading of silicon to industrial and solar grade. Module assembly, mounting and cell processing also contribute significantly to emissions from solar PV production. Emissions from hydrogen compression and transportation, as well as hydrogen production, account for only 5% and 1%, respectively.

12.4.2 Greenhouse gas footprint of gasification As shown in Fig. 12.11, the thermal gasification pathway results in life cycle GHG emissions of 2.03, 2.88 and 3.05 kg CO2 eq/kg H2 for agricultural residues, forest residues and whole trees, respectively. Hydrogen production is the largest contributor to the life cycle GHG emissions in all three feedstocks. The high requirement of electricity and natural gas during plant operation are the main drivers in both the forest residue and whole tree pathways. In the agricultural residue pathway, the GHG emissions are due to the consumption of natural gas. Thermal gasification involves an energyintensive drying process to reduce the high moisture contents of the feedstocks. The contribution from the infrastructure requirement and

396

PART | IV Sustainability

FIGURE 12.10 Life cycle GHG emissions of solar and wind-based H2 production. GHG, Greenhouse gas.

FIGURE 12.11 Life cycle GHG emissions of H2 production from biomass gasification. GHG, Greenhouse gas.

The environmental performance of hydrogen production Chapter | 12

397

decommissioning is small, less than 1% in all cases. Next to H2 conversion, feedstock production has considerable impact. For the agricultural residue pathway, biomass production accounts for 29% of the total GHG emissions. Fertilizer application is the main source of GHG emissions in this stage. Considerable amounts of nitrogen-, phosphorous- and potassium-based fertilizer are required in order to make up for the nutrient loss due to straw removal. Diesel use in the straw collection process (in raking, baling and roadsiding) also contributes to feedstock production GHG emissions. Biomass production in the forest residue and whole tree pathways account for 9% and 17%, respectively. In the forest residues preparation, chipping and forwarding contribute 70% and 40%, respectively. In the case of whole tree, the contribution from chipping is 40%, while felling and skidding each account for around 30% of biomass production emissions. Biomass transportation has a large contribution in the forest residue pathway, unlike in the other two pathways, mainly due to the high moisture content of the biomass and the distance between the harvesting area and the H2 plant. Biohydrogen transportation accounts for 12%
17% of the total emissions from electricity demand in booster stations. The overall GHG emissions from the gasification pathway are comparable with those in existing literature [16,20,106]. A comparative LCA of H2 production from renewable and nonrenewable sources by Acar and Dincer estimated GHG emissions of biomass gasification to be around 5.25 kg/kg H2 [20]. Recent studies by Li et al. [106] and Mehmeti et al. [16] reported values of 5.40 and 2.67 kg/kg H2 for biomass to H2 conversion via thermal gasification, respectively.

12.4.3 Greenhouse gas footprint of bio-oil reforming GHG emissions from the bio-oil reforming of agricultural residues, forest residues and whole trees are 3.43, 2.83 and 1.92 kg CO2 eq/kg H2, respectively (Fig. 12.12). Feedstock production contributes considerably to the life cycle GHG emissions in the case of agricultural residues (45%). Bio-oil transportation has a massive impact on GHG emissions in all cases; it accounts for 36%
64% of the total. The high impact of bio-oil transportation is due to the high density of bio-oil, which is around 1200 kg/m3 compared with H2 gas with a density of 0.089 kg/m3. Pump operation, which accounts for more than 95% of the pipeline energy use, is the key driver of GHG emissions from transportation. The source of electricity, hence, has a large impact on overall emissions. Sourcing renewable-based electricity such as hydro would reduce pipeline emissions. An extensive study comparing pipeline and truck transport for large-scale bio-oil production was carried out by Pootakham and Kumar [107]. The results suggest that in a region like Alberta, where the grid mix is dominated by coal energy, truck transportation has considerably fewer GHG emissions than pipeline.

398

PART | IV Sustainability

FIGURE 12.12 Life cycle GHG emissions of H2 production via the bio-oil reforming of biomass. GHG, Greenhouse gas.

The contribution from plant operations is negligible, as the energy required to dry the biomass is provided from the combustion of char generated in the process.

12.4.4 Greenhouse gas footprint of supercritical water gasification We estimated GHG emissions from the SCWG of algae biomass be 11.1 kg/ kg H2. As Fig. 12.13 shows, algae feedstock production is responsible for 70% of the GHG emissions in SCWG, and electricity consumption is the main contributor. Electricity is required in key equipment such as pumps, centrifuge and membrane filtration to reduce the water content of algae and produces a dry biomass with 20% moisture content for the subsequent process. Upstream emissions associated with nutrient supply for algae growth make up 17% of the life cycle GHG emissions. Nitrogen and phosphorus are the most important nutrients for algae growth. Agricultural fertilizers such as ammonia and diammonium phosphate are the main sources of nitrogen and phosphate. It is worth mentioning that like any other photosynthetic organism, algae use CO2 as a carbon source. The carbon content in microalgae biomass is around 45%
50% [108]. CO2-rich industrial exhaust gases could be sources of carbon for microalgae, which makes them a carbon fixation alternative. Approximately 1.8 tonnes of CO2 are required for each tonne of algae biomass [88]. However, the carbon sequestered during algae growth is not considered in this study. The H2 conversion stage accounts for 30% of the GHG emissions. Energy consumption in the compressor and pumps in

The environmental performance of hydrogen production Chapter | 12

399

FIGURE 12.13 Life cycle GHG emissions of H2 production from algae via SCWG. GHG, Greenhouse gas; SCWG, supercritical water gasification.

various unit operations (such as gasification and the gas purification units) are the main sources. Hydrogen production via SCWG uses biomass with a high moisture content; therefore the biomass does not require drying, and overall energy requirements and associated GHG emissions are lower than for the thermal gasification of similar feedstock.

12.4.5 Comparative assessment incorporating sensitivity and uncertainty analyses Fig. 12.14 shows the life cycle GHG emissions in all nine pathways. Several factors are behind the variation in GHG emission results
the energy and feedstock sources, conversion technologies and transportation and distribution efforts. H2 production from algae biomass via SCWG has the highest impact, which ranges from 10.14 to 12.72 kg CO2 eq/kg H2. Algae cultivation is the stage that contributes the most emissions. It is important to note that CO2 fixation during algae growth is not considered in this study. Depending on the source of CO2, from the atmosphere or from flue gas from industrial process, algae consumes approximately 1.8 tonnes of CO2 per tonne of biomass. This could lead to a net GHG impact of 21 kg/kg algae biomass cultivation, which could significantly reduce overall GHG emission results. Wind-based water electrolysis was found to have the lowest GHG emissions among the pathways considered, 0.69 6 0.04. Wind power generates the cleanest electricity among the sources and therefore has the lowest impact. The source of electricity is the main concern in the GHG emissions of water electrolysis, as the electrolyzer’s contribution is negligible. Bandhari et al. confirm that wind electrolysis is one of the most

400

PART | IV Sustainability

FIGURE 12.14 Life cycle GHG emissions of H2 production. GHG, Greenhouse gas.

environmentally sustainable H2 production alternatives to fossil fuel-based feedstocks electrolysis or conventional grid mix electricity-based electrolysis [24]. Moreover, hydrogen production from wind and solar helps to store excess renewable power to be used in different sectors. As most wind farms are located some distance from their use, hydrogen produced from electrolysis can help wind energy to become more accessible to end-users. GHG emissions from gasification and bio-oil reforming are comparable. For gasification, agricultural residues offer the lowest GHG emissions (1.812.25 kg CO2 eq/kg H2); values for forest residues and whole trees range from 2.62 - 3.17 and 2.91 - 3.23, respectively. Wheat straw is a byproduct from wheat grain production. There are different ways of allocating the emissions. Assuming that wheat grain is the primary product, straw is considered as a residue with low economic value. The GHG emissions from residue production are due to harvesting and the application of fertilizer to avoid soil nutrient disturbance from straw removal. For agricultural residues, the biohydrogen yield, electricity emissions factor, straw-to-grain ratio and fertilizer requirement are the most sensitive parameters. It is worth mentioning that the GHG emissions from agricultural residue would be different if emissions were allocated between the wheat grain and the straw based on their mass fractions or economic values. GHG emissions from forest residue gasification are highly sensitive to the biomass transportation distance, biohydrogen yield, diesel fuel emission factor, electricity emission factor, chipper productivity and chipper fuel consumption rate. GHG emissions from bio-oil reforming are 2.80 - 4.33, 2.33 - 3.46, and 1.57- 2.60 kg CO2 eq/kg H2, for agricultural residues, forest residues and whole trees, respectively. The biohydrogen yield, methanol blend ratio, pipeline distance and emission factor and diesel emission factor are the most sensitive parameters. It is critical to look at those parameters when interpreting the results.

The environmental performance of hydrogen production Chapter | 12

401

12.5 Conclusions The critical role of H2 in supporting a low-carbon global economy has been well acknowledged in scientific literature. Several alternative production pathways offer environmental or cost advantages depending on the type of feedstock used, conversion technology and other factors. Hence, there is a growing interest in systematically quantifying the economic viability and environmental sustainability performances of alternative pathways to support quality decision-making. Life cycle assessment, an internationally harmonized and standardized framework, has been widely applied for this purpose. However, results from comparative assessments are not robust, given the inconsistencies in methods, system boundaries, allocation criteria, spatial and temporal variations and so on. In this context, this study aims to address those challenges by developing a common system boundary to reasonably compare the GHG footprints of nine renewable H2 production pathways. These are wind-H2, solar-H2, bio-H2 (via gasification and bio-oil reforming) from straw, whole trees and forest residue and the SCWG of algae. A Morris sensitivity was performed using a Monte Carlo simulation to screen the sensitive parameters for uncertainty analysis. At 0.69 6 0.04 kg CO2 eq/kg H2, wind-based water electrolysis appears to be the most environmentally sustainable pathway. For the gasification and bio-oil reforming pathways, agricultural feedstock has lower GHG emissions than forest residue and whole tree. Dominated by the emissions from cultivation, H2 from SCWG of algae biomass appears to have the greatest impact of the nine pathways considered. The biohydrogen yield, electricity emissions factor, straw-to-grain ratio, fertilizer requirement, biomass transportation distance, diesel fuel emission factor, chipper productivity and chipper fuel consumption rate are the most sensitive parameters.

Acknowledgements The authors are grateful to the NSERC/Cenovus/Alberta Innovates Associate Industrial Research Chair Program in Energy and Environmental Systems Engineering and the Cenovus Energy Endowed Chair in Environmental Engineering at the University of Alberta for financial support for this research. As a part of the University of Alberta’s Future Energy Systems (FES) research initiative, this research was made possible in part thanks to funding from the Canada First Research Excellence Fund (CFREF). Astrid Blodgett is thanked for editing this paper.

References [1] Anderegg WRL, Prall JW, Harold J, et al. Expert credibility in climate change. Proc Natl Acad Sci 2010;107(27):12107
9. [2] Cook J, Nuccitelli D, Green SA, et al. Quantifying the consensus on anthropogenic global warming in the scientific literature. Environ Res Lett 2013;8(2):024024.

402

PART | IV Sustainability

[3] Cook J, Oreskes N, Doran PT, et al. Consensus on consensus: a synthesis of consensus estimates on human-caused global warming. Environ Res Lett 2016;11(4):048002. [4] Bro¨nnimann S, Franke J, Nussbaumer SU, et al. Last phase of the Little Ice Age forced by volcanic eruptions. Nat Geosci 2019;12(8):650
6. [5] Neukom R, Barboza LA, Erb MP, et al. Consistent multidecadal variability in global temperature reconstructions and simulations over the Common Era. Nat Geosci 2019;12 (8):643
9. [6] Neukom R, Steiger N, Go´mez-Navarro JJ, et al. No evidence for globally coherent warm and cold periods over the preindustrial Common Era. Nature 2019;571(7766):550
4. [7] UNEP. The emissions gap report 2017. Nairobi: United Nations Environment Programme (UNEP); 2017. [8] UNFCC. Report of the conference of the parties on its twenty-first session FCCC/CP/ 2015 2015/10/Add.1. ,https://unfccc.int/resource/docs/2015/cop21/eng/10a01.pdf.; [cited 19 November 2019]. [9] Hydrogen Council. Hydrogen scaling up: a sustainable pathway for the global energy transition, ,https://hydrogencouncil.com/wp-content/uploads/2017/11/Hydrogen-scaling-upHydrogen-Council.pdf.; 2017 [cited 10 September 2019]. [10] Bogdanov D, Farfan J, Sadovskaia K, et al. Radical transformation pathway towards sustainable electricity via evolutionary steps. Nat Commun 2019;10(1):1077. [11] Brandon NP, Kurban Z. Clean energy and the hydrogen economy. Philos Trans R Soc A Math Phys Eng Sci 2017;375(2098):20160400. ´ . The role of hydrogen in low carbon energy futures
a [12] Hanley ES, Deane JP, Gallacho´ir BPO review of existing perspectives. Renew Sustain Energy Rev 2018;82:3027
45. [13] IRENA. Hydrogen from renewable power: technology outlook for the energy transition. Abu Dhabi: International Renewable Energy Agency; 2018. [14] Serrano DP, Dufour J, Iribarren D. On the feasibility of producing hydrogen with net carbon fixation by the decomposition of vegetable and microalgal oils. Energy Environ Sci 2012;5(3):6126
35. [15] Mart´ın-Gamboa M, Iribarren D, Susmozas A, et al. Delving into sensible measures to enhance the environmental performance of biohydrogen: a quantitative approach based on process simulation, life cycle assessment and data envelopment analysis. Bioresour Technol 2016;214:376
85. [16] Mehmeti A, Angelis-Dimakis A, Arampatzis G, et al. Life cycle assessment and water footprint of hydrogen production methods: from conventional to emerging technologies. Environments 2018;5(2):24. [17] Hacatoglu K, Rosen MA, Dincer I. Comparative life cycle assessment of hydrogen and other selected fuels. Int J Hydrog. Energy 2012;37(13):9933
40. [18] Ozbilen A, Dincer I, Rosen MA. Exergetic life cycle assessment of a hydrogen production process. Int J Hydrog. Energy 2012;37(7):5665
75. [19] Rosner V, Wagner H-J. Life cycle assessment and process development of photobiological hydrogen production
from laboratory to large scale applications. Energy Procedia 2012;29:532
40. [20] Acar C, Dincer I. Comparative assessment of hydrogen production methods from renewable and non-renewable sources. Int J Hydrog. Energy 2014;39(1):1
12. [21] Ghandehariun S, Kumar A. Life cycle assessment of wind-based hydrogen production in western Canada. Int J Hydrog. Energy 2016;41(22):9696
704. [22] Kabir MR, Kumar A. Development of net energy ratio and emission factor for biohydrogen production pathways. Bioresour Technol 2011;102(19):8972
85.

The environmental performance of hydrogen production Chapter | 12

403

[23] Kumar M, Oyedun AO, Kumar A. A comparative analysis of hydrogen production from the thermochemical conversion of algal biomass. Int J Hydrog. Energy 2019;44 (21):10384
97. [24] Bhandari R, Trudewind CA, Zapp P. Life cycle assessment of hydrogen production via electrolysis
a review. J Clean Prod 2014;85:151
63. [25] Valente A, Iribarren D, Dufour J. Life cycle assessment of hydrogen energy systems: a review of methodological choices. Int J Life Cycle Assess 2017;22(3):346
63. [26] Mishra A, Gautam S, Sharma T. Effect of operating parameters on coal gasification. Int J Coal Sci Technol 2018;5(2):113
25. [27] M. Binder, M. Kraussler, M. Kuba, M. Luisser, Hydrogen from biomass gasification, In R. Rauch (ed.), IEA Bioenergy, 2018. [28] Abbasi T, Abbasi SA. ‘Renewable’ hydrogen: prospects and challenges. Renew Sustain Energy Rev 2011;15(6):3034
40. [29] Mazloomi K, Gomes C. Hydrogen as an energy carrier: prospects and challenges. Renew Sustain Energy Rev 2012;16(5):3024
33. [30] McKone JR, Lewis NS, Gray HB. Will solar-driven water-splitting devices see the light of day? Chem Mater 2014;26(1):407
14. [31] Barbir F. PEM electrolysis for production of hydrogen from renewable energy sources. Sol Energy 2005;78(5):661
9. [32] Glenk G, Reichelstein S. Economics of converting renewable power to hydrogen. Nat Energy 2019;4(3):216
22. [33] Wiser R, Jenni K, Seel J, et al. Expert elicitation survey on future wind energy costs. Nat Energy 2016;1(10):16135. [34] Comello S, Reichelstein S, Sahoo A. The road ahead for solar PV power. Renew Sustain Energy Rev 2018;92:744
56. [35] Zhang X, Chan SH, Ho HK, et al. Towards a smart energy network: the roles of fuel/electrolysis cells and technological perspectives. Int J Hydrog. Energy 2015;40 (21):6866
919. [36] Ursua A, Gandia LM, Sanchis P. Hydrogen production from water electrolysis: current status and future trends. Proc IEEE 2012;100(2):410
26. [37] Nuttall LJ, Fickett AP, Titterington WA. Hydrogen generation by solid polymer electrolyte water electrolysis. In: Veziro˘glu TN, editor. Hydrogen energy. Boston, MA: Springer; 1975. [38] Carmo M, Fritz DL, Mergel J, et al. A comprehensive review on PEM water electrolysis. Int J Hydrog. Energy 2013;38(12):4901
34. [39] Shiva Kumar S, Himabindu V. Hydrogen production by PEM water electrolysis
a review. Mater Sci Energy Technol 2019;2(3):442
54. [40] Brisse A, Schefold J, Zahid M. High temperature water electrolysis in solid oxide cells. Int J Hydrog. Energy 2008;33(20):5375
82. [41] Laguna-Bercero MA. Recent advances in high temperature electrolysis using solid oxide fuel cells: a review. J Power Sources 2012;203:4
16. [42] Go¨tz M, Lefebvre J, Mo¨rs F, et al. Renewable power-to-gas: a technological and economic review. Renew Energy 2016;85:1371
90. [43] Demirbas MF. Nitrogenous chemicals from carbon based materials. Energy Explor Exploit 2005;23(3):215
24. [44] Kumar A, Cameron JB, Flynn PC. Biomass power cost and optimum plant size in western Canada. Biomass Bioenergy 2003;24(6):445
64. [45] REN21. Renewables 2019 global status report. Paris: REN21 Secretariat; 2019.

404

PART | IV Sustainability

[46] Jiang X, Huang J-G, Cheng J, et al. Interspecific variation in growth responses to tree size, competition and climate of western Canadian boreal mixed forests. Sci Total Environ 2018;631-632:1070
8. [47] Stadt KJ, Huston C, David KC, et al. Evaluation of competition and light estimation indices for predicting diameter growth in mature boreal mixed forests. Ann For Sci 2007;64 (5):477
90. [48] Government of Alberta. Forests Act. ,http://www.qp.alberta.ca/1266.cfm?page 5 F22. cfm&leg_type 5 Acts&isbncln 5 9780779752508.; 2014 [cited 20 October 2019]. [49] Arregi A, Amutio M, Lopez G, et al. Evaluation of thermochemical routes for hydrogen production from biomass: a review. Energy Convers Manag 2018;165:696
719. [50] Sikarwar VS, Zhao M, Clough P, et al. An overview of advances in biomass gasification. Energy Environ Sci 2016;9(10):2939
77. [51] De Lasa H, Salaices E, Mazumder J, et al. Catalytic steam gasification of biomass: catalysts, thermodynamics and kinetics. Chem Rev 2011;111(9):5404
33. [52] Dincer I. Green methods for hydrogen production. Int J Hydrog. Energy 2012;37 (2):1954
71. [53] Parthasarathy P, Narayanan KS. Hydrogen production from steam gasification of biomass: influence of process parameters on hydrogen yield
a review. Renew Energy 2014;66:570
9. [54] Kaushal P, Tyagi R. Steam assisted biomass gasification
an overview. Can J Chem Eng 2012;90(4):1043
58. [55] Molino A, Chianese S, Musmarra D. Biomass gasification technology: the state of the art overview. J Energy Chem 2016;25(1):10
25. [56] Lopez G, Alvarez J, Amutio M, et al. Assessment of steam gasification kinetics of the char from lignocellulosic biomass in a conical spouted bed reactor. Energy 2016;107:493
501. [57] Font Palma C. Modelling of tar formation and evolution for biomass gasification: a review. Appl Energy 2013;111:129
41. [58] Sansaniwal SK, Rosen MA, Tyagi SK. Global challenges in the sustainable development of biomass gasification: an overview. Renew Sustain Energy Rev 2017;80:23
43. [59] Peng WX, Wang LS, Mirzaee M, et al. Hydrogen and syngas production by catalytic biomass gasification. Energy Convers Manag 2017;135:270
3. [60] Anis S, Zainal ZA. Tar reduction in biomass producer gas via mechanical, catalytic and thermal methods: a review. Renew Sustain Energy Rev 2011;15(5):2355
77. [61] Woolcock PJ, Brown RC. A review of cleaning technologies for biomass-derived syngas. Biomass Bioenergy 2013;52:54
84. [62] Ahmad AA, Zawawi NA, Kasim FH, et al. Assessing the gasification performance of biomass: a review on biomass gasification process conditions, optimization and economic evaluation. Renew Sustain Energy Rev 2016;53:1333
47. [63] Okolie JA, Rana R, Nanda S, et al. Supercritical water gasification of biomass: a state-ofthe-art review of process parameters, reaction mechanisms and catalysis. Sustain Energy Fuels 2019;3(3):578
98. [64] Lu Y, Guo L, Zhang X, et al. Hydrogen production by supercritical water gasification of biomass: explore the way to maximum hydrogen yield and high carbon gasification efficiency. Int J Hydrog. Energy 2012;37(4):3177
85. [65] Reddy SN, Ding N, Nanda S, et al. Supercritical water gasification of biomass in diamond anvil cells and fluidized beds. Biofuel Bioprod Biorefin 2014;8(5):728
37. [66] Kruse A. Supercritical water gasification. Biofuel Bioprod Biorefin 2008;2(5):415
37.

The environmental performance of hydrogen production Chapter | 12

405

[67] Reddy SN, Nanda S, Dalai AK, et al. Supercritical water gasification of biomass for hydrogen production. Int J Hydrog. Energy 2014;39(13):6912
26. [68] Demirbas A. Hydrogen-rich gas from fruit shells via supercritical water extraction. Int J Hydrog. Energy 2004;29(12):1237
43. [69] Resende FLP, Neff ME, Savage PE. Noncatalytic gasification of cellulose in supercritical water. Energy Fuels 2007;21(6):3637
43. [70] Nanda S, Dalai AK, Kozinski JA. Supercritical water gasification of timothy grass as an energy crop in the presence of alkali carbonate and hydroxide catalysts. Biomass Bioenergy 2016;95:378
87. [71] Yanik J, Ebale S, Kruse A, et al. Biomass gasification in supercritical water: part 1. Effect of the nature of biomass. Fuel 2007;86(15):2410
15. [72] Nanda S, Isen J, Dalai AK, et al. Gasification of fruit wastes and agro-food residues in supercritical water. Energy Convers Manag 2016;110:296
306. [73] Trane R, Dahl S, Skjøth-Rasmussen MS, et al. Catalytic steam reforming of bio-oil. Int J Hydrog. Energy 2012;37(8):6447
72. [74] Arregi A, Lopez G, Amutio M, et al. Hydrogen production from biomass by continuous fast pyrolysis and in-line steam reforming. RSC Adv 2016;6(31):25975
85. [75] Gollakota ARK, Reddy M, Subramanyam MD, et al. A review on the upgradation techniques of pyrolysis oil. Renew Sustain Energy Rev 2016;58:1543
68. [76] Tanksale A, Beltramini JN, Lu GM. A review of catalytic hydrogen production processes from biomass. Renew Sustain Energy Rev 2010;14(1):166
82. [77] ISO. ISO 14040 International standard
environmental management
life cycle assessment
principles and framework. Geneva, Switzerland: International Organisation for Standardization; 2006. [78] ISO. ISO 14044 International standard
environmental management
life cycle assessment
requirements and guidelines. Geneva, Switzerland: International Organisation for Standardization; 2006. [79] TransAlta Corporation. Summerview 2. ,https://www.transalta.com/facilities/plants-operation/summerview-2/.; 2019 [cited 21 October 2019]. [80] Mehedi TH, Gemechu E, Kumar A. Life cycle greenhouse gas emissions and energy footprints of utility-scale solar energy systems. Edmonton: University of Alberta; 2019. [81] Nagy CN. Energy coefficients for agriculture inputs in western Canada. Centre for Studies in Agriculture, Law and the Environment, University of Saskatchewan, Saskatchewan, 1999. [82] Johnson MC, Palou-Rivera I, Frank ED. Energy consumption during the manufacture of nutrients for algae cultivation. Algal Res 2013;2(4):426
36. [83] Sarkar S, Kumar A. Biohydrogen production from forest and agricultural residues for upgrading of bitumen from oil sands. Energy 2010;35(2):582
91. [84] Sarkar S, Kumar A. Large-scale biohydrogen production from bio-oil. Bioresour Technol 2010;101(19):7350
61. [85] Sarkar S, Kumar A. Techno-economic assessment of biohydrogen production from forest biomass in Western Canada. Trans ASABE 2009;52(2):519
30. [86] Spath P, Aden A, Eggeman T, et al. Biomass to hydrogen production detailed design and economics utilizing the battelle columbus laboratory indirectly-heated gasifier. Golden, CO (US): National Renewable Energy Lab; 2005. p. 161. [87] Khan MI, Shin JH, Kim JD. The promising future of microalgae: current status, challenges, and optimization of a sustainable and renewable industry for biofuels, feed, and other products. Microb Cell Fact 2018;17(1):36. [88] Zhu L. Microalgal culture strategies for biofuel production: a review. Biofuel Bioprod Biorefin 2015;9(6):801
14.

406

PART | IV Sustainability

[89] Handler RM, Canter CE, Kalnes TN, et al. Evaluation of environmental impacts from microalgae cultivation in open-air raceway ponds: analysis of the prior literature and investigation of wide variance in predicted impacts. Algal Res 2012;1(1):83
92. [90] Huesemann MH, Van Wagenen J, Miller T, et al. A screening model to predict microalgae biomass growth in photobioreactors and raceway ponds. Biotechnol Bioeng 2013;110(6):1583
94. [91] Kumar K, Mishra SK, Shrivastav A, et al. Recent trends in the mass cultivation of algae in raceway ponds. Renew Sustain Energy Rev 2015;51:875
85. [92] Demirbas A, Demirbas MF. Demirbas, algae energy: algae as a new source of biodiesel. Green energy and technology. London: Springer; 2010. p. 139. [93] Odjadjare EC, Mutanda T, Olaniran AO. Potential biotechnological application of microalgae: a critical review. Crit Rev Biotechnol 2017;37(1):37
52. [94] Show PL, Tang MSY, Nagarajan D, et al. A holistic approach to managing microalgae for biofuel applications. Int J Mol Sci 2017;18(1):215. [95] Pankratz S, Oyedun AO, Kumar A. Novel satellite based analytical model developed to predict microalgae yields in open pond raceway systems and applied to Canadian sites. Algal Res 2019;39:101431. [96] Pankratz S, Oyedun AO, Zhang X, et al. Algae production platforms for Canada’s northern climate. Renew Sustain Energy Rev 2017;80:109
20. [97] Pankratz S, Oyedun AO, Kumar A. Development of cost models of algae production in a cold climate using different production systems. Biofuel Bioprod Biorefin 2019;13(5):1246
60. [98] Pankratz S, Kumar M, Oyedun AO, et al. Environmental performances of diluents and hydrogen production pathways from microalgae in cold climates: open raceway ponds and photobioreactors coupled with thermochemical conversion. Algal Res 2019;47:101815. [99] Huijbregts MAJ. Application of uncertainty and variability in LCA. Int J Life Cycle Assess 1998;3(5):273. [100] Di Lullo G, Gemechu E, Oni AO, et al. Extending sensitivity analysis using regression to effectively disseminate life cycle assessment results. Int J Life Cycle Assess 2019;1
18. [101] Morris MD. Factorial sampling plans for preliminary computational experiments. Technometrics 1991;33(2):161
74. [102] Davis M, Moronkeji A, Ahiduzzaman Md, et al. Assessing renewable generation pathways for a coal-based electricity sector. Edmonton, AB: University of Alberta:; 2019. [103] Di Lullo G, Zhang H, Kumar A. Evaluation of uncertainty in the well-to-tank and combustion greenhouse gas emissions of various transportation fuels. Appl Energy 2016;184:413
26. [104] Di Lullo G, Zhang H, Kumar A. Uncertainty in well-to-tank with combustion greenhouse gas emissions of transportation fuels derived from North American crudes. Energy 2017;128:475
86. [105] Di Lullo GR. Uncertainty in life cycle assessments of well-to-wheel greenhouse gas emissions of transportation fuels derived from various crude oils. Mechanical engineering. Edmonton, AB: University of Alberta:; 2016. [106] Li G, Cui P, Wang Y, et al. Life cycle energy consumption and GHG emissions of biomass-to-hydrogen process in comparison with coal-to-hydrogen process. Energy 2019;116588. [107] Pootakham T, Kumar A. A comparison of pipeline versus truck transport of bio-oil. Bioresour Technol 2010;101(1):414
21. [108] Yaprak D, Spielberg ET, Baecker T, et al. A roadmap to uranium ionic liquids: anticrystal engineering. Chem Eur J 2014;20(21):6482
93.

Chapter 13

Integrated economic
environmental
social assessment of straw for bioenergy production Junnian Song, Kexin Li and Wei Yang College of New Energy and Environment, Jilin University, Changchun, P.R. China

Chapter Outline 13.1 Introduction 407 13.2 Methods 410 13.2.1 Estimation of straw available for energy production 410 13.2.2 Cost and profit of straw utilization for energy production 411 13.2.3 Environmental impacts of straw utilization for energy production 413 13.2.4 Selection of evaluation indicators 414 13.3 Case study 414 13.3.1 Estimation of the quantity of straw 416

13.3.2 Parameters of energy conversion technologies 419 13.4 Results and discussion 421 13.4.1 Energy potential of straw 421 13.4.2 Energy, environmental and socioeconomic benefits of straw utilization 421 13.4.3 Analysis of major factors affecting the results 425 13.5 Discussion 428 13.6 Conclusions 429 Subscripts and superscripts 430 References 431

13.1 Introduction The problems of energy shortage, climate change and environmental pollution caused by continuous economic growth are increasingly severe worldwide. In this context, bioenergy as the fourth largest energy source following coal, oil and natural gas is playing a considerable role on emerging renewable energy in the world [1]. Especially crop straw resources as typical bioresources are relatively abundant and low sulfur-containing [2]. In addition, Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00013-3 © 2021 Elsevier Inc. All rights reserved.

407

408

PART | IV Sustainability

the combustion of straw is considered as carbon neutral since the amount of carbon dioxide released is comparable to that absorbed from the atmosphere during crop growth [3]. PM2.5 emissions and the haze phenomenon aggravated by the open burning of discarded straw could be dramatically alleviated once it is utilized for energy use instead [4,5]. Straw can be converted through various energy conversion technologies (ECTs) into energy products, including electricity, heat and solid, gaseous and liquid biofuels, which are diversified and appeal to the market demand. Not only the environmental burdens could be ameliorated but also certain benefits for farmers could be created (employments, income from straw sales), if straw resources could be fully exploited for energy production to realize industrial development. Currently the theoretical researches on straw utilization for energy production are mainly focused on the following directions: (1) estimation of the available potential of straw resources; (2) optimization of the whole supply system of straw utilization and (3) environmental and economic assessments of ECTs. Quantification of the energy potential of straw is based on three levels: theoretical reserve, collectable quantity and quantity available for energy production [6]. The theoretical reserve is calculated based on the grain yield and straw
grain ratio, which is referred to The Food and Agriculture Organization’s calculation method. The collectable quantity is calculated based on the theoretical reserve excluding the loss during collection and transportation. The quantity available for energy production is calculated based on the collectable quantity through deducting that used in other utilization ways (returning to cropland, fertilizer, feed, papermaking, etc.). Jiang et al. [7] and Liu et al. [8] statically analyzed the temporal energy potential of different crop straw in different regions of China using the above method. Monforti et al. [9] and Weiser et al. [10] also estimated the energy potential of straw in the European Union, Germany and other regions in a similar way. On this basis, studies regarding dynamic prediction of the energy potential of straw were carried out. Ji [11] predicted the yield of crop residues in China with an artificial neural network mode. Che [12] adopted the geographical information system (GIS) technology to simulate the spatial distribution of straw resources in China and estimated the future resource potential by the grey prediction method. Based on regional energy potential and the energy density of straw resources, an optimal location for establishment of a straw-energy project needs to be identified considering spatial factors and a supply system (consisting of collection, pretreatment, storage and transportation) needs to be optimized to pursue a logistics process with lower cost and emissions [13
16]. Ho¨hn et al. [17] used a GIS based method to analyze the spatial distribution and amount of potential biomass feedstock for biogas production and optimal locations, sizes and number of biogas plants in Southern Finland. Aldana et al. [18] considered energy production maximization and total cost minimization by constructing a comprehensive mixed integer linear

Integrated economic Chapter | 13

409

programming model to analyze the supply chain of biofuel production with agricultural residues in Mexico. Zhao and Li [19] developed a bi-objective 0
1 integer programming model for designing optimal locations and corresponding feedstock supply chain based on relevant data of biomass power generation in China, to achieve a win
win situation between cost and greenhouse gas (GHG) emissions. Because of the divergence in the performances of energy production, cost, profit, GHG emissions, sustainability performance, ECTs of straw attracted more scholars to carry out assessments of these performances with life cycle assessment (LCA), emergy analysis, strengths, weakness, opportunities and threats analysis, multicriteria decision making analysis (MCDMA) and some others [20
23]. Hu et al. [24] conducted a preliminary LCA on a straw briquette fuel plant in China and covered only emissions of GHGs and air pollutants. Wang et al. [25] evaluated the energy consumption and GHG emissions of a direct-combustion power generation project with forestry residues as feedstock using the layered hybrid evaluation model. Zhao et al. [26] applied a five-force competitive model to assess the current situation and future development of China’s biomass power generation industry. Khishtandar et al. [27] used the MCDMA method based on the hesitant fuzzy linguistic data to deal with the prioritization of different bioenergy technologies in Iran. In addition, some researchers have conducted studies regarding industrial development of straw utilization from the perspectives of government, farmers and markets to put forward macro policy measures [28,29]. It could be found in the reviewed studies that the prediction of future quantity of straw resources is generally based on the historical data of straw’s quantity, without considering complex impacts on the yield of grain, the basis for the calculation of straw’s quantity. This would affect the accuracy of the prediction results of straw’s quantity, as well as the energy potential of straw. The reviewed studies have attached emphasis to the environmental and economic performances of specific ECTs, however, disregarding how a certain region could benefit from straw utilization through multiple ECTs quantitatively. At present, regional energy demand is diverse, stimulating development of various straw-energy industries (SEIs). This also necessitates evaluating the overall benefits contributed by industrialization of straw utilization with regards to energy security, environmental impacts, social contribution and economic benefits for a region. This study attempts to construct an integrated model for dynamically assessing the energy, environmental and socioeconomic benefits of straw utilization for energy production for a certain region. Taking Jilin Province, China as a typical study area, six factors that affect the grain yield are taken into consideration for estimating the quantity of straw resources spanning 15 years (2016
30) by principal component analysis (PCA) and autoregressive integrated moving average model. The straw available for energy

410

PART | IV Sustainability

production is considered to be allocated to typical ECTs to produce electricity, solid and liquid biofuels, hence allowing SEIs to be formed (each ECT corresponding to a specific SEI). Based on the LCA theory, the regional energy, environmental and socioeconomic benefits from development of three SEIs will be quantitatively evaluated by three groups of indicators (in total nine indicators) screened from Global Bioenergy Partnership’s (GBEP’s) sustainability indicators, in order to embody the comprehensiveness of the evaluation results [30]. It is an integration of dynamic analysis elaborating various stages of straw utilization for energy production as well as the industrial development process.

13.2 Methods 13.2.1 Estimation of straw available for energy production 13.2.1.1 Influential factors of grain yield The quantity of straw is determined by grain yield which is influenced by many factors, including technical level, natural condition, economic factor, etc. Given that u influential factors affect grain yield (Y), denoted as X1, X2,. . ., Xu. PCA is used to characterize the relationship between grain yield and these influential factors. It allows to eliminate the multiple colinear problem when dealing with the problem of multiple regression. First the primary data are standardized, followed by obtaining the eigenvalues of correlation coefficient matrix. Then the number of principal components can be determined based on the variance percentage (greater than 85%) for calculating corresponding eigenvectors. The relationship between the principal component and the variables can be expressed as: Zm 5 am X1 1 am X2 1 ?am Xu

ð13:1Þ

Finally a multivariate regression analysis of grain yield with Tm main components can be expressed as follows (c and k are the parameters of multivariate regression): Y5

Tm X

ðcm 1 km Zm Þ

ð13:2Þ

m51

13.2.1.2 Energy potential of straw The autoregressive integrated moving average (ARIMA) (p, d, q) model is selected for the prediction of grain yield using Eviews software. The model can be simply operated as it only requires endogenous variables instead of other exogenous variables and the accuracy of the prediction could be guaranteed. The specific steps are as follows:

Integrated economic Chapter | 13

411

1. Adjustment of the stability of the input data with the histograms of the autocorrelation and partial autocorrelation function (ACF and PACF) and ADF unit root test; 2. Differential treatment of nonstationary sequence and determination of the differential order d (d 5 0, if the primary sequence is stationary); 3. Estimation of other parameters in the ARIMA model based on the characters of ACF and PACF (determination of p and q); 4. Parameters estimation and hypothesis test and 5. Time series prediction with the verified model. Finally the grain yield can be predicted with Formula (13.2), with whose results the energy potential of straw can be calculated based on three levels: theoretical reserve (Qtl), collectable quantity (Qtc) and quantity available for energy production (Qt). Finally Qt could be expressed as grain yield (Yt) multiplying the straw
grain ratio (η), collection coefficient (λ) and energy utilization coefficient (μ) as: Qt 5 Qct 3 μ 5 Qlt 3 λ 3 μ 5 Yt 3 η 3 λ 3 μ

ð13:3Þ

The energy density of straw refers to the quantity of straw available for energy production in per unit sown area. It can be expressed as: Dt 5

Qt St 3 100

ð13:4Þ

13.2.2 Cost and profit of straw utilization for energy production It is difficult to ascertain the whole procurement process of straw because of its scattering distribution. At present, there are mainly three straw collection patterns. In the first one, the collected straw is transported directly to the energy project without any transfer or processing. In the second one, the collected straw is transported to a big stock center near the energy project for processing and then transported to the energy project. In the last one, the collected straw is transported to the center of some resource-islands for processing, which are distributed around one energy project. The resource-island distribution pattern for straw resources, which proves to be the most costeffective one has been widely adopted in theoretical studies of bioresource logistics and is employed in this study as depicted in Fig. 13.1 [19]. Several assumptions are given to simplify the complex conditions: 1. The cycle of straw harvest is 1 year; 2. Several resource-islands are distributed around one energy project which is the center of one circular collection area. Different types of straw are evenly distributed in one resource-island, with no differences in collection and transportation processes. The quantity of different types of straw is summable;

412

PART | IV Sustainability

FIGURE 13.1 Resource-island distribution mode for straw: transportation outside the island (left); transportation within the island (right).

3. The collection radius in each resource-island in the region is constant; 4. Straw is collected and transported to the center of the island, where binding and compression are carried out subsequently, a process called transportation within the island. Finally the straw harvested in all resource-islands is transported to the energy project, a process called transportation outside the island [31] and 5. The life cycle of one ECT is defined as from the collection of straw to the production of its energy product. The growing period of crops and the supply and consumption of the bioenergy products are not included. The life cycle cost of an ECT is composed of the cost for straw procurement (Cproc), the cost for materials (Cmate), energy (Cenergy), maintenance (Cmain) and labour (Clabour) for project operation, the cost for enterprise tax (Ctax) and (4) the cost for depreciation of fixed asset (Cdepr). Ci 5 Cproc;i 1 Cmate;i 1 Cmain;i 1 Cenergy;i 1 Clabour;i 1 Ctax;i 1 Cdepr;i

ð13:5Þ

Among them, total procurement cost includes purchase cost (Cpurc), process cost (Cprocess), transportation cost (Ctran) and other cost (Cothers) (e.g. loading and storage). The purchase cost, process cost and other cost are calculated by corresponding unit price (p) multiplying straw demand of one ECT (Mi), respectively. Cproc;i 5 Cpurc;i 1 Cprocess;i 1 Ctran;i 1 Cothers;i

ð13:6Þ

Cpurc;i 5 ppurc 3 Mi

ð13:7Þ

Cprocess;i 5 pprocess 3 Mi

ð13:8Þ

Cothers;i 5 pothers 3 Mi

ð13:9Þ

Total straw demand of one ECT is provided by a certain number of resource-islands. For one resource-island, the transportation cost (Ctran,k,i) consists of that within the island (Cin,k,i) and that outside the island (Cout,k,i).

Integrated economic Chapter | 13

413

They are calculated by the transportation distance multiplying the quantity of straw provided by an island (Qk,i) and the price rate of transportation (ϕ). Particularly the transportation distance inside the island is determined by the integral infinitesimal method as Formula (13.11). A tortuosity factor of road f is adopted considering that the road is not perfectly straight. Ctran;k;i 5 Cin;k;i 1 Cout;k;i ð Rk 2 Cin;k;i 5 2f πrk;i 2 Dk φdrk;i 5 f φπDk Rk;i 3 3 0

ð13:10Þ

Cout;k;i 5 f φQk;i Lk;i

ð13:12Þ

Ctran;i 5 Ctran;k;i 3

Mi Qk;i

ð13:11Þ

ð13:13Þ

The net profit generated by an ECT is determined by the production amount of energy products (Ni) and its unit price (ρi) and total cost (Ci), expressed as: NPi 5 Ii 2 Ci 5 ρi 3 Ni 2 Ci

ð13:14Þ

13.2.3 Environmental impacts of straw utilization for energy production For an ECT, the GHG and air pollutant emissions within the life cycle mainly originate from the processing and transportation stages of straw and the energy conversion stage. Specifically the emissions from the processing and transportation stages are considered as from diesel consumption of baling machine and truck. Consumption of coal-fired power during energy conversion stage is considered as the source of emissions. The emissions of processing stage (Eprocess,i) are calculated by straw demand of one ECT (Mi) multiplying the diesel consumption rate (ς) and corresponding emission coefficients (ed). Similar to the transportation cost, for one resource-island the emissions of transportation stage (Etran,k,i) include those from inside the island (Ein,k,i) and those from outside the island (Eout,k,i), calculated by the transportation distance multiplying the diesel consumption rate (ξ) and corresponding emission coefficients (ed). The emissions of the energy conversion stage (Econv) are calculated by the energy production (Ni) multiplying consumption of coal-fired power of unit energy product (τ i) and corresponding emission coefficients (ec). Ei 5 Eprocess;i 1 Etran;i 1 Econv;i

ð13:15Þ

Eprocess;i 5 ς 3 Mi 3 ed 3 1000

ð13:16Þ

414

PART | IV Sustainability

Etran;k;i 5 Ein;k;i 1 Eout;k;i Ein;k;i 5

ð Rk

2f πrk;i 2 Dk ed ξdrk;i 5

0

2 f πDk ed ξRk;i 3 3

ð13:17Þ ð13:18Þ

Eout;k;i 5 fed ξLk;i Qk;i

ð13:19Þ

Mi Qk;i

ð13:20Þ

Etran;i 5 Etran;k;i

Econv;i 5 τ i 3 Ni 3 ec

ð13:21Þ

13.2.4 Selection of evaluation indicators The GBEP published 24 sustainability indicators for bioenergy under three pillars. They are designed to evaluate the energy, environmental and socioeconomic impacts of bioenergy technologies and make prioritization of the technologies for a certain region/target/scenario [30]. This study aims to take advantage of some of these indicators to quantitatively evaluate the benefits from development of SEIs for a region. Therefore these indicators are screened and improved based on whether the indicators can be quantified or relevant data can be easily attained. The indicators that cannot be quantified or without enough data support are eliminated (such as ‘net energy balance’, ‘water quality’, ‘change in mortality’ and ‘burden of disease attributable to indoor smoke’). The nine selected indicators and their descriptions are presented in Table 13.1. The energy benefits are indicated by productivity (the amount of energy products that could be produced by SEIs, unified as calorific value) and fossil fuel substitution (the SEIs produce energy products which could substitute traditional fossil fuels such as coal and oil products). The socioeconomic benefits are indicated by farmers’ income from selling straw, employment newly created by SEIs and net profit contributed by SEIs. The environmental benefits are indicated by mitigation of GHG and air pollutant emissions. This entails to quantify ‘the amount of GHG and air pollutant emissions that could be reduced by consuming the energy products of SEIs instead of traditional fossil fuels’. Besides, the mitigated emissions due to avoided open burning of discarded straw are also covered. These indicators are expected to represent the overall benefits a region could obtain from development of SEIs.

13.3 Case study Jilin Province is one of the most important agricultural province of China, with large grain production. A continuously increasing trend for grain yield is presented currently, which also indicates abundant straw resources and a

Integrated economic Chapter | 13

415

TABLE 13.1 Selected assessment indicators and their descriptions. Evaluation objective

Indicators

Description

Environmental benefits

GHG mitigation

Compared with using fossil fuels (coal and oil), GHG and other air pollutants emissions can be reduced by using bioenergy.

SO2 mitigation NOx mitigation PM2.5 mitigation Socioeconomic benefits

Energy benefits

Farmers income

Direct income of farmers from sale of straw.

Employment

Job creation as a result of labour for SEIs.

Net profit

Total income deducting total cost.

Productivity

Energy output unified as calorific value.

Fossil fuel substitution

Substitution amount of fossil fuels with bioenergy.

GHG, Greenhouse gas; SEIs, straw-energy industries.

rising trend. In 2015, the grain yield of Jilin Province was 36.47 million t (Mt), producing 33.65 Mt of straw available for energy production. However, only a small proportion (5.92 Mt in 2015) was used for bioenergy production practically, limited by technological immaturity, insufficient investment and weak policy support, let alone the development of the industrialization of ECTs. Meanwhile, PM2.5 and other air pollutants released due to the yearly open burning of discarded straw is seriously deteriorating the air quality and human health. According to the 13th 5-year plan for energy development and the 13th 5-year plan for the development of new and renewable energy of Jilin Province, the installed capacity of biomass power generation should be 1166 MW and the production of briquette fuel and cellulose fuel bioethanol should be 5 Mt and 100,000 t (kt) by 2020; the installed capacity of biomass power generation should be 1700 MW and the production of briquette fuel and cellulose fuel bioethanol should be 8 Mt and 300 kt by 2030 [32,33]. In order to maximize the potential energy, environmental and socioeconomic benefits of straw utilization for energy production, this study considers straw utilization through three typical ECTs including direct-combustion power generation (ECT1), briquette fuel (ECT2) and cellulose fuel bioethanol (ECT3) with Jilin Province as an empirical

416

PART | IV Sustainability

study area. The gasification technology and biogas technology are not considered due to their technological immaturity and climatic requirement for the study area. According to the operation parameters of ECTs, corresponding benefits of industrialization of three SEIs can be assessed.

13.3.1 Estimation of the quantity of straw 13.3.1.1 Regional grain yield Combined with the analysis of the influential factors of grain yield performed by scholars engaged in agricultural research [34,35], six factors affecting grain yield are determined as: sown area (X1), amount of fertilizers used (X2), actual irrigation area (X3), total power of agricultural machines (X4), agricultural gross product (X5) and area affected by natural disasters (X6). According to the Jilin Statistical Yearbook and China Rural Statistical Yearbook, the panel data of grain yield and six influential factors spanning 30 years (1986
2015) are provided in Table 13.2. Based on the results of PCA, grain yield (Y) can be expressed by two principal components (Z1, Z2) and the regression equation is: Y 5 0:580 3 Z1 2 1:492 3 Z2 1 13:043

ð13:22Þ

The ARIMA model is used to predict grain yield with the above time sequences of principal components (Z1, Z2). Determination of model parameters (p, d, q) depends on the unit root test and residual analysis according to Akaike Information Criterion and Bayesian Information Criterion. It is found that Z1 is applicable to ARIMA (3, 1, 3) and Z2 is applicable to ARIMA (2, 2, 1). Finally prediction results of grain yield of Jilin Province are presented in Fig. 13.2.

13.3.1.2 Conversion coefficients of straw As it lacks the historical data regarding the yield of each kind of grain and the influential factors in the statistical materials, it is not performable to predict the yield of each kind of grain with the above models. Only the provincial total grain yield can be computed with currently available panel data. Considering that the yield of corn accounts for more than 80% of total provincial grain yield, the straw
grain ratio of corn will be used to represent that of all crops. The collection coefficient is derived from the results of crop straw investigation in Jilin Province [36]. The energy utilization coefficient is determined by the straw utilization ways, defined as the residual proportion except that used for returning to cropland, feed and industrial materials [37]. According to the study about current situation of straw utilization in Jilin Province, the relevant proportions are provided in Table 13.3 [38,39]. Finally the straw
grain ratio (η), collection coefficient (λ) and

TABLE 13.2 Data on grain yield and influential factors (1986
2015). Year

Y (106 t)

X1 (106 hm2)

X2 (106 t)

X3 (106 hm2)

X4 (106 kW)

X5 (109 USD)

X6 (106 hm2)

1986

13.98

3.47

0.61

0.72

5.29

4.43

2.21

1987

16.76

3.49

0.68

0.75

5.35

5.27

1.07

1988

16.93

3.42

0.67

0.77

5.53

5.47

1.86

1989

13.51

3.43

0.72

0.84

5.91

4.58

2.90

1990

20.47

3.53

0.85

0.88

6.29

6.13

0.47

1991

18.99

3.54

0.92

0.92

5.88

5.93

1.49

1992

18.40

3.54

0.91

0.91

5.91

6.01

2.15

1993

19.01

3.53

0.87

0.91

6.07

6.55

2.18

1994

20.16

3.57

0.92

0.91

5.99

6.79

2.23

1995

19.92

3.58

1.01

0.90

6.61

6.61

2.33

1996

23.27

3.62

1.07

0.94

7.19

7.78

0.99

1997

18.08

3.59

1.08

1.08

7.73

6.71

3.04

1998

25.06

3.57

1.13

1.25

8.28

8.44

1.74

1999

23.06

3.51

1.16

1.29

8.97

8.36

1.12

2000

16.38

3.36

1.12

1.32

10.15

7.21

3.25

2001

19.53

3.36

1.14

1.38

10.97

8.05

2.92

2002

22.15

4.04

1.17

1.50

11.51

9.12

1.43 (Continued )

TABLE 13.2 (Continued) Year

Y (106 t)

X1 (106 hm2)

X2 (106 t)

X3 (106 hm2)

X4 (106 kW)

X5 (109 USD)

X6 (106 hm2)

2003

22.60

4.01

1.22

1.55

12.31

9.57

1.91

2004

25.10

4.31

1.59

1.60

13.20

10.38

1.31

2005

25.81

4.29

1.59

1.61

14.71

11.02

1.77

2006

27.20

4.33

1.38

1.64

15.72

12.08

1.50

2007

24.54

4.33

1.54

1.64

16.78

11.59

2.96

2008

28.40

4.39

1.64

1.68

18.00

13.35

0.43

2009

24.60

4.43

1.74

1.68

20.01

13.09

2.66

2010

28.43

4.49

1.83

1.73

21.45

13.92

0.85

2011

31.71

4.55

1.95

1.83

23.55

14.94

0.62

2012

33.43

4.61

2.07

1.85

25.55

15.62

0.63

2013

35.51

4.79

2.17

1.85

27.27

16.56

0.66

2014

35.33

5.00

2.27

1.63

29.19

17.65

1.96

2015

36.47

5.08

2.31

1.79

31.53

18.50

0.85

Integrated economic Chapter | 13

419

FIGURE 13.2 Prediction of total grain yield of Jilin Province (2016
30).

TABLE 13.3 Current situation of crop straw utilization in Jilin Province. Utilization ways

Returning to cropland

Feed

Household burning

Edible fungus

Industrial materials

Discard

Proportion (%)

11.42

10.68

59.93

0.045

0.24

17.69

energy utilization coefficient (μ) in Formula (13.3) are determined as 1.32, 0.9 and 0.78, respectively.

13.3.2 Parameters of energy conversion technologies Specific parameters of three typical ECTs are presented in Table 13.4. According to the market price, the purchase price of straw is 32.11 USD/t [43]. The corporate tax is calculated as 10% of the gross income. The depreciation rate is 5% of the initial investment. Transportation cost is the most variable one in the procurement cost, varying due to different collection radius and transportation distance. The radius of a resource-island (Rk) is assumed as 3 km [see Formulas (13.11) and (13.18)]. Considering different yearly straw demand of ECTs, the transportation distance (Lk) outside the island are respectively assumed as 20, 5 and 30 km [see Formulas (13.12) and (13.19)]. The tortuosity factor of road (f) is 1.5 [see Formulas (13.11), (13.12), (13.18) and (13.19)] [26]. The price rate of transportation (ϕ) is 0.07 USD/m3 km [see Formulas (13.11) and

420

PART | IV Sustainability

TABLE 13.4 Economic parameters of ECTs [37,40
42]. Items

ECT1

ECT2

ECT3

25 MW

10,000 t

50,000 t

Straw demand (10 t/a)

210

12

300

Depreciation life (years)

20

10

15

Initial investmenta

36,928

482

77,249

9221

478

13,985

6743

385

9633

1057

60

1510

1252

23

2601

169

10

241

Chemical materials

1143

40

16,858

Energy consumptiona

286

88

5154

597

8

7980

462

46

1156

Scale 3

 

a

Procurement costs a

Straw purchase a

Process 

a

Transportation a

Others

a

a

Maintenance 

a

Labour



Corporate tax

1626

96

5839



a

1846

48

5150

15,181

805

56,122

16,256

963

58,390

 

a

Depreciation a

Total cost

a

Total income



Net profit

1075

158

2268



Output of product (103 kWh, t)

150,000

10

50

0.12

96.33

1168

a

Unit price of product (USD/kWh, USD/t)

Notes: The unit of ECT1’s energy product is kWh/a. The unit of ECT2 and ECT3’s energy products is t/a. “ ” denotes the calculated value of this study, others are parameters from references. ECTs, Energy conversion technologies. a Unit:103 USD/a.

(13.12)]. The density of straw-bundle after processing in the center of the resource-island is 0.4 t/m3. The diesel consumption rate of processing (ς) and transportation (ξ) are 4.25 kg/t [see Formula (13.16)] and 0.06 kg/t km [see Formulas (13.18) and (13.19)], respectively. In this study, the environment impacts are dealt with as GHG and air pollutant emissions. The environmental benefits can be embodied through the substitution of coal-fired power with straw direct-combustion power, the substitution of coal fuels with briquette fuel, and the substitution of oil fuels

421

Integrated economic Chapter | 13

TABLE 13.5 Environmental impact parameters of fossil fuels combustion and straw open burning [24,44,45]. Energy types (kg/t)

CO2

CO

CH4

Coal

2532.00

2.22

1.68

Oil

3530.00

0.20

NOx

PM2.5

SO2

9.95

11.54

11.50

3.41

7.24

1.95

12.00

Diesel

3161.62

20.24

0.18

12.14

1.76

0.68

Open burning of straw

1350.00

60.00

4.40

4.30

11.70

0.44

with bioethanol. Additionally the utilization of straw for energy production avoids the environment impacts of open burning of discarded straw. Several environmental impact parameters of fossil fuels are provided in Table 13.5.

13.4 Results and discussion 13.4.1 Energy potential of straw As illustrated in Fig. 13.3, in 2030, the theoretical reserve and the quantity available for energy production could amount to 67.39 and 47.10 Mt, respectively, with the latter 1.4 times that in 2015. The trends in industrial development of SEIs during 2016
30 are set as coinciding with the government planning as aforementioned. Suppose that all the energy products of straw utilization are used locally, total energy production of straw utilization could amount to 3.82 Mt standard coal equivalent (tce) by 2020, accounting for 4.13% of total regional primary energy consumption. By 2030, total quantity of straw utilized for energy production could be 25.68 Mt, with the utilization proportion (the actual quantity utilized for energy production divided by the total quantity available for energy production) rising from 17.61% in 2015 to 54.53% in 2030.

13.4.2 Energy, environmental and socioeconomic benefits of straw utilization As presented in Table 13.6, the data of nine indicators for ECTs are used to quantify the overall energy, environmental and socioeconomic benefits of SEIs. The amount of energy that could be converted through one ECT using 1 t straw is indicated by the energy indicator ‘productivity’. The amount of fossil energy that could be substituted through one ECT using 1 t straw (with standard coal equivalent as unit) is indicated by the energy indicator ‘fossil fuel substitution’. Energy product of each ECT is converted to calorific value. The socioeconomic indicators for each ECT are denoted by income/

422

PART | IV Sustainability

FIGURE 13.3 The quantity of straw resources (2015
30).

TABLE 13.6 Data of the assesment indicators for each ECT. Indicators

Environmental

Socioeconomic

Energy

Unit

Values ECT1

ECT2

ECT3

GHG mitigation

kg/GJ

992.12

219.71

252.64

SO2 mitigation

kg/GJ

10.29

0.14

0.26

NOx mitigation

kg/GJ

2.59

0.74

0.62

PM2.5 mitigation

kg/GJ

Farmers income

USD/GJ

5.07

1.39

1.35

13.87

2.29

3.60

Employment

Number/kJ

246.79

47.62

74.64

Net profit

USD/GJ

2.21

0.94

0.85

Productivity

GJ/t (straw)

2.32

14.00

8.93

Fossil fuel substitution

kgce/t (straw)

79.01

477.70

304.76

ECT, Energy conversion technology.

employment/net profit of unit energy output. The environmental indicators for each ECT are unified as mitigation amount of unit energy output. With the carbon-neutral character of straw as premise: (1) For ECT1, when consuming power generated by it, the induced life cycle emissions of ECT1 (as analysed in Section 2.3) are included without considering CO2 emissions from straw combustion process. The emissions of burning coal for power generation are considered as reference for comparison; (2) For ECT2, when consuming briquette fuel produced by it, the induced life cycle emissions of

Integrated economic Chapter | 13

423

ECT2 are included without considering CO2 emissions from burning the briquette fuel. The emissions from burning coal product are considered as reference for comparison; (3) For ECT3, when consuming bioethanol produced by it, the induced life cycle emissions of ECT3 are included without considering CO2 emissions from burning the bioethanol. The emissions from burning oil products are considered as reference for comparison. Besides, the ECTs consume the straw that would be originally discarded and openburned, bringing about additional mitigation of air pollutant emissions. The above analyses explicate how the mitigation coefficients in Table 13.6 are calculated. The results of the benefits during 2016
30 are depicted in Fig. 13.4. Seen overall, SEI1 has the best performance on SO2 mitigation. GHG mitigation, NOx mitigation and the socioeconomic benefits contributed by SEI1 and SEI2 are comparable, all more remarkable than those of SEI3. In terms of the energy benefits, SEI2 has greater advantage compared with SEI1 and SEI3. Some more detailed analyses are as follows. Firstly for the energy benefits, all the energy products of SEIs are equivalent to1.88 billion GJ and could replace 64.05 Mtce fossil fuels (see Fig. 13.4A,B). SEI2 contributes to 73% in fossil fuel substitution and productivity attributed to more energy production and higher calorific value of briquette fuel compared with electricity and bioethanol. The energy benefits of SEIs imply the potentials of SEIs for adjusting the energy structure to a cleaner and more sustainable level. Secondly for the socioeconomic benefits, 166,000.0 employments could be created accumulatively. Farmers could obtain 8.7 billion USD accumulatively by selling straw. The SEIs could obtain an accumulative net profit of 2.2 billion USD which could be termed as a driving force for the industrial development (see Fig. 13.4C
E). The socioeconomic benefits of each SEI are determined by the indicator of unit energy output (see Table 13.6) and total energy output collectively. SEI1 and SEI2 contribute to most of the socioeconomic benefits due to more energy output compared with SEI3. Finally for the environmental benefits, GHG mitigation is more significant compared with air pollutants mitigation (see Fig. 13.4F
I). The accumulative mitigation amount is 700.25 Mt within 15 years, with comparable contribution of SEI1 and SEI2 (both around 45%). The accumulative mitigation amount of SO2 is 3.99 Mt, with SEI1 contributing to the most (94%). The accumulative mitigation amount of NOx and PM2.5 is 2.05 and 3.94 Mt, respectively, with comparable contribution of SEI1 and SEI2 (both around 47%). Similar to the socioeconomic benefits, the environmental benefits of each SEI are determined by the mitigation amount of unit energy output (see Table 13.6) and total energy output collectively. For example GHG mitigation amount of unit energy output of SEI1 is larger than that of SEI2, however, its energy output is less than that SEI2. This ultimately leads to comparable contribution of GHG mitigation for SEI1 and SEI2.

FIGURE 13.4 The results for nine assessment indicators (2015
30). A
I represent the results for each individual indicator.

Integrated economic Chapter | 13

425

13.4.3 Analysis of major factors affecting the results The results of the energy, environmental and socioeconomic benefits of straw utilization are influenced by geographical conditions, energy density of straw, collection patterns, feedstock price, policy support, technical efficiency and other various factors. In a certain region where geographical conditions, energy density of straw and collection patterns are given, collection radius, purchase price and utilization proportion of straw are most fluctuant and significant, thus lead to more remarkable impacts. In the following, these three major factors are selected as focus to analyze the impacts of their changes on the energy, socioeconomic and environmental benefits of SEIs.

13.4.3.1 Changes in collection radius Because of scattered distribution and low density of straw resources, the transportation cost is significantly higher than that of traditional fossil fuels. The high transportation cost discourages the investors, leading to bottleneck of industrial development of straw utilization. This necessitates the clarification of how changes in collection radius affect the final net profit for SEIs. Given the resource-island collection pattern and the energy production target for each SEI (according to the government planning), the energy and social benefits are not affected by changes in collection radius. As the overall life cycle environmental benefits of SEIs are substantial, changes in collection radius only bring about tiny impacts on them. Only the economic benefits are greatly affected as depicted in Fig. 13.5. As transportation cost accounts for different proportions in total cost of each ECT, the sensitivity of each SEIs’ net profit to changes in collection radius is distinct. SEI3 is most insensitive to changes in collection radius, followed by SEI1 and SEI2. When the collection radius is extended longer than 39, 26 and 58 km, it becomes unprofitable for SEI1, SEI2 and SEI3, respectively.

FIGURE 13.5 The impact of changes in collection radius on the accumulative net profit.

426

PART | IV Sustainability

13.4.3.2 Changes in purchase price of straw The purchase price affects the total procurement cost of straw the most substantially. It also has significant impacts on the intention of farmers for selling straw and net profit for SEIs. According to literature review and field investigation, the purchase price of straw is around 24
56 USD/t. The analysis of the impacts of changes in purchase price of straw is conducted within this range. As depicted in Fig. 13.6, changes in purchase price have the greatest impact on SEI1, followed by SEI2 and SEI3. When the purchase price of straw rises to 37, 45 and 40 USD/t, the accumulative net profit for SEI1, SEI2 and SEI3 declines to zero, respectively. The purchase price of straw is 32 USD/t currently. The increasingly more prosperous development of SEIs may result in rising demand for straw, as well as intensified intension of farmers for income. Thereby the purchase price may appear a rising trend. 13.4.3.3 Changes in utilization proportion of straw The results of the energy, environmental and socioeconomic benefits of straw utilization presented in Section 4.2 are based on the bioenergy production targets proposed in the local government planning. The results indicate that the utilization proportion could rise from 17.61% in 2015 to 54.53% in 2030, presenting increment trend however without taking full advantage of straw available for energy production. It calls for an assessment of the maximized benefits of utilizing all straw available for energy production. Next, it assumes that all straw available for energy production are allocated to three SEIs annually, whose rates of industrial development are in accordance with those in the planning scenario. The results of nine assessment indicators of complete utilization of straw are illustrated in Fig. 13.7. The development scales of SEIs are directly determined by the utilization proportion. If utilized completely, the straw available for energy production could contribute

FIGURE 13.6 The impact of purchase price on accumulative net profit.

FIGURE 13.7 The effect of complete straw utilization on the accumulative benefits. A
I represent the results for each individual indicator.

428

PART | IV Sustainability

to the accumulative energy, environmental and socioeconomic benefits within 15 years more than twice those in the planning scenario. Especially the accumulative fossil fuel substitution and net profit could increase over 2.5 times. Compared with the planning scenario, the results further indicate that there is still large room for straw utilization for energy production and the industrial development.

13.5 Discussion The quantity of straw available for energy production as the basis of industrial development of straw utilization and ECTs is subjected to grain yield and the utilization ways of straw. The selected six factors in this study help to predict the changing trend of grain yield. Among them, the previous five factors have positive effects with comparable contribution to grain yield. Only the area affected by natural disasters plays a negative role, which is consistent with the objective fact. In the future, the positive factors tend to intensify along with the increment grain demand and expansion of arable land, with the area affected by natural disasters as an uncertain one. The comprehensive utilization structure of straw will also change along with various demands for straw, especially that of returning to cropland and industrial production. These factors could lead to uncertainties of the results of the quantity of straw available for energy production. In addition, this study assumes that the straw originally intended for household burning is used for energy production. On the one hand, the quantity of straw for energy production is determined to some extent by farmers’ willingness to sell straw practically. On the other hand, the income from selling straw stimulates the farmers to sell straw and use alternative energy instead of household straw burning with the possibility that there is extra money left. For instance, the income from selling 1 t straw can purchase 80 m3 natural gas generating equivalent amount of heat. Therefore there are also uncertainties of the results of the quantity of straw available for energy production with regard to this point. In this study, the development tendency of SEIs, as well as the utilization proportion of straw in the study area is estimated considering local government planning. The rate for production of three SEIs is similar, with an about 20% annual growth rate from 2015 to 2020. During 2020
30, the annual growth rate for SEI1 and SEI2 will decrease to around 4%, with that of SEI3 increasing to 12%. All three SEIs will be rapidly developing in the previous five years, the initial stage of their industrial development. In order to quantify the specific influence of straw utilization on regional sustainable development, the energy, environmental and socioeconomic benefits are elaborated by three groups of indicators (in total nine indicators). They are selected according to the target of this paper and can be changed for other regions/objectives.

Integrated economic Chapter | 13

429

In terms of the environmental benefits, this study focuses on the assessment of GHG, SO2, NOx and PM2.5 mitigations. With the environmental indicators, the benefits of mitigating GHGs and air pollutant emissions from two aspects replacing the fossil fuels and avoiding the open burning of discarded straw are quantified. The environmental benefits of each SEI are determined by the mitigation amount of unit energy output (see Table 13.6) and total energy output collectively. The former is determined by the life cycle emissions and the energy product of the ECT. The latter is determined by the target of local bioenergy production. Additional income (8.7 billion USD accumulatively) and chance for employment (0.17 million employments accumulatively) as social benefits of SEIs, present a direct benefit for farmers and encourage them with regard to positive attitude and acceptability for SEIs. Besides, the created employments could help to gather the idle labour and improve rural development. An accumulative net profit of 2.2 billion USD could be obtained by SEIs. The economic benefits are affected by the procurement cost most substantially, especially the purchase cost and transportation cost of straw. Changes in purchase price have the greatest influence on SEI1. Transportation cost is affected by collection radius which is different for SEIs according to their demand for straw. SEI2 is the most sensitive to changes in collection radius. Larger energy density of straw and more abundant straw resources contribute to shorter collection radius and thus lower procurement cost. Considering the purchase price related to farmers’ willingness to sell, there is hardly possibility for a declining trend. In order to save total procurement cost, it needs to make an arrangement of each link of straw procurement, such as mechanizing the collection process, optimizing the transportation route, improving technological efficiency, etc. The results of this paper reveal the overall benefits of straw utilization for energy production under current government planning targets, with an emphasis on the important role of decision makers on industrial development of SEIs. Practically more factors affect future trends of SEIs’ industrial development. Policies regarding taxes and subsidies and price of energy products formulated by government also affect net profit for SEIs. In addition, technological innovations should also be taken into account for increasing production efficiency and reducing equipment imports. The above analysis justifies the need for integrated assessment of straw utilization for energy production and the improvement of policy supporting system in a region, based on which industrial development of straw utilization for energy production could be optimized and promoted.

13.6 Conclusions An integrated model is developed to dynamically predict the energy potential of straw resources and assess the regional energy, environmental and

430

PART | IV Sustainability

socioeconomic benefits of straw utilization for energy production (with Jilin Province as a typical study area). The novelty and significance of this study lie in three aspects: the energy potential of straw is dynamically predicted considering six influential factors of grain yield; based on three groups of indicators (in total nine indicators) screened from GBEP’s sustainability indicators, the overall benefits of SEIs are evaluated annually; the assessment is an integration of dynamic analysis elaborating various stages of straw utilization for energy production as well as the industrial development process. For the study area, there is a continuous rising trend for the quantity of straw available for energy production, which in 2030 could amount to 47.10 Mt, allowing to produce 6.26 Mtce energy products under the government planning goal. Industrial development of SEIs contributes to mitigating 700.25 Mt GHGs, 3.99 Mt SO2, 2.05 Mt NOx and 3.94 Mt PM2.5 accumulatively. 2.2 billion USD net profit and 166,000 employments could be created accumulatively. Compared with the government planning scenario, if all straw available for energy production is utilized, all the benefit indicators could increase more than 2.5 times, showing large room for development of SEIs. With industrial development of straw utilization through ECTs, regional overall sustainable development could be substantially promoted. The results presented are expected to provide decision makers with guidance for regional development of new and renewable energy industries and reference with regard to the establishment of appropriate energy development strategies.

Subscripts and superscripts t i k Variables and coefficients Yt Xu Zm am Qt Qtc Qtl μ λ η Dt St

year (from 2016 to 2030); bioenergy conversion technologies (ECTs); resource-island (the total number is n);

grain yield (t); the u-th influential factor of grain yield; the m-th principal component (Tm is the total principal components, m 5 1, 2); corresponding eigenvector of principal component Zm; quantity of straw available for energy production (t); collectable quantity of straw (t); theoretical reserve of straw (t); energy utilization coefficient; collection coefficient; straw
grain ratio; energy density of straw (t/km2); sown area of grain (hm2);

Integrated economic Chapter | 13 Rk Lk,i Mi Ci p NPi Ni Ii ρi Ei f ϕ ξ ς τi ed ec

431

radius of resource-island k (km); transportation distance outside island k (km); straw demand of ECTi (t/a); total cost of ECTi (USD/a); unit price of purchasing (processing, loading and storing) straw (USD/t); net profit of ECTi (USD/a); production amount of energy product of ECTi (kWh, t); total income of ECTi (USD); price of energy product of ECTi (USD/kWh, USD/t); total emissions of ECTi (kg); tortuosity factor of road; price rate of transportation (USD/t km); diesel consumption rate of transportation (kg/t km); diesel consumption rate of processing (kg/kg); consumption of coal-fired power of unit energy product of ECTi (kWh/t); emission coefficient of diesel (kg/kg); emission coefficient of coal-fired power (kg/kWh).

References [1] Konur O. The scientometric evaluation of the research on the production of bioenergy from biomass. Biomass Bioenerg 2012;47:504
15. [2] Gawronska G, Gawronski K. The method of assessment of potential ecological effects of obtaining energy from the straw biomass. Acta Sci Pol-Form Circumiectus 2016;15:69
79. [3] Weldemichael Y, Assefa G. Assessing the energy production and GHG (greenhouse gas) emissions mitigation potential of biomass resources for Alberta. J Clean Prod 2016;112:4257
64. [4] Hong JL, Ren LJ, Hong JM, Xu CQ. Environmental impact assessment of corn straw utilization in China. J Clean Prod 2016;112:1700
8. [5] Ding AJ, Fu CB, Yang XQ, Sun JN, Petaja T, Kerminen VM, et al. Intense atmospheric pollution modifies weather: a case of mixed biomass burning with fossil fuel combustion pollution in eastern China. Atmos Chem Phys 2013;13:10545
54. [6] Said N, EI-Shatoury SA, Diaz LF, Zamorano M. Quantitative appraisal of biomass resources and their energy potential in Egypt. Renew Sust Energy Rev 2013;24:84
91. [7] Jiang D, Zhuang DF, Fu JY, Huang YH, Wen KG. Bioenergy potential from crop residues in China: availability and distribution. Renew Sust Energy Rev 2012;16:1377
82. [8] Liu J, Wu JG, Liu FQ, Han XG. Quantitative assessment of bioenergy from crop stalk resources in Inner Mongolia, China. Appl Energy 2012;93:305
18. [9] Monforti F, Lugato E, Motola V, Bodis K, Scarlat N, Dallemand JF. Optimal energy use of agricultural crop residues preserving soil organic carbon stocks in Europe. Renew Sust Energy Rev 2015;44:519
29. [10] Weiser C, Zeller V, Reinicke F, Wagner B, Majer S, Vetter A, et al. Integrated assessment of sustainable cereal straw potential and different straw-based energy applications in Germany. Appl Energy 2014;114:749
62. [11] Ji LQ. An assessment of agricultural residue resources for liquid biofuel production in China. Renew Sust Energy Rev 2015;44:561
75.

432

PART | IV Sustainability

[12] Che L. Research on resource estimation, distribution and utilization potential of crop residue [dissertation]. Dalian: Dalian University of Technology; 2014 [in Chinese]. [13] Shafie SM, Masjuki HH, Mahlia TMI. Rice straw supply chain for electricity generation in Malaysia: economical and environmental assessment. Appl Energy 2014;135:299
308. [14] Sun YF, Cai WC, Chen B, Guo XY, Hu JJ, Jiao YZ. Economic analysis of fuel collection, storage, and transportation in straw power generation in China. Energy 2017;132:194
203. [15] Venier F, Yabar H. Renewable energy recovery potential towards sustainable cattle manure management in Buenos Aires Province: site selection based on GIS spatial analysis and statistics. J Clean Prod 2017;162:1317
33. [16] Delivand MK, Cammerino ARB, Garofalo P, Monteleone M. Optimal locations of bioenergy facilities, biomass spatial availability, logistics costs and GHG (greenhouse gas) emissions: a case study on electricity productions in South Italy. J Clean Prod 2015;99:129
39. [17] Ho¨hn J, Lehtonen E, Rasi S, Rintala J. A geographical information system (GIS) based methodology for determination of potential biomasses and sites for biogas plants in sourthern Finland. Appl Energy 2013;113:1
10. [18] Aldana H, Lozano FJ, Acevedo J. Evaluating the potential for producing energy from agricultural residues in Mexico using MILP optimization. Biomass Bioenergy 2014;67:372
89. [19] Zhao XG, Li A. A multi-objective sustainable location model for biomass power plants: case of China. Energy 2016;112:1184
93. [20] Hiloidhari M, Baruah DC, Singh A, Kataki S, Medhi K, Kumari S, et al. Emerging role of geographical information system (GIS), life cycle assessment (LCA) and spatial LCA (GIS-LCA) in sustainable bioenergy planning. Bioresour technol 2017;242:218
26. [21] Martire S, Tuomasjukka D, Lindner M, Fitzgerald J, Castellani V. Sustainability impact assessment for local energy supplies’development
the case of the alpine area of Lake Como, Italy. Biomass Bioenergy 2015;83:60
76. [22] Portugal-Pereira J, Soria R, Rathmann R, Schaeffer R, Szklo A. Agricultural and agroindustrial residues-to-energy: techno-economic and environmental assessment in Brazil. Biomass Bioenergy 2015;81:521
33. [23] Vaidya A, Mayer AL. Use of a participatory approach to develop a regional assessment tool for bioenergy production. Biomass Bioenergy 2016;94:1
11. [24] Hu JJ, Lei TZ, Wang ZW, Yan XY, Shi XG, Li ZF, et al. Economic, environmental and social assessment of briquette fuel from agricultural residues in China
a study on flat die briquetting using corn stalk. Energy 2014;64:557
66. [25] Wang CB, Zhang LX, Chang Y, Pang MY. Biomass direct-fired power generation system in China: an integrated energy, GHG emissions, and economic evaluation for Salix. Energy Policy 2015;84:155
65. [26] Zhao ZY, Zuo J, Wu PH, Yan H, Zillante G. Competitiveness assessment of the biomass power generation industry in China: a five forces model study. Renew Energy 2016;89:144
53. [27] Khishtandar S, Zandieh M, Dorri B. A multi criteria decision making framework for sustainability assessment of bioenergy production technologies with hesitant fuzzy linguistic term sets: the case of Iran. Renew Sust Energy Rev 2017;77:1130
45. [28] Thompson JL, Tyner WE. Corn stover for bioenergy production: cost estimates and farmer supply response. Biomass Bioenergy 2014;62:166
73.

Integrated economic Chapter | 13

433

[29] Golecha R, Gan JB. Effects of corn stover year-to-year supply variability and market structure on biomass utilization and cost. Renew Sust Energy Rev 2016;57:34
44. [30] GBEP (Global Bioenergy Partnership). GBEP sustainability indicators for bioenergy [internet], ,http://www.globalbioenergy.org/programmeofwork/task-force-on-sustainability/gbep-report-on-sustainability-indicators-for-bioenergy/en/.; 2011. [31] Thakur A, Canter CE, Kumar A. Life-cycle energy and emission analysis of power generation from forest biomass. Appl Energy 2014;128:246
53. [32] JPBS (Jilin Province Bureau of Statistics). The 13th 5-year plan for energy development of Jilin Province. Beijing: China Statistics Press; 2017 [in Chinese]. [33] JPBS (Jilin Province Bureau of Statistics). The13th 5-year plan for the development of new and renewable energy of Jilin Province. Beijing: China Statistics Press; 2017 [in Chinese]. [34] Wang KY, Zhang PY. The research on impact factors and characteristic of cultivated land resources use efficiency-take Henan Province, China as a case study. IERI Procedia 2013;5:2
9. [35] Yang YF, Xu XR. Principal component analysis and GM (1,1) forecast of grain output influencing factors in Fujian Province. J South Agric 2014;697
703 [in Chinese]. [36] Bi YY. Study on straw resources evaluation and utilization [dissertation]. Chinese Academy of Agricultural Sciences, Beijing, 2010 [in Chinese]. [37] Song JN, Yang W, Higano Y, Wang XE. Dynamic integrated assessment of bioenergy technologies for energy production utilizing agricultural residues: an input
output approach. Appl Energy 2015;158:178
89. [38] Liu P, Na W, Wang XL, Wang XM, Zhang WD, Wang XF. Analysis on evaluation and energy utilization of main crop stalk resource in Jilin Province. J Jilin Agric Sci 2010;58
64 [in Chinese]. [39] JPG (Jilin Province Government). Agriculture committee of Jilin province actively promoted straw remanding [internet] [in Chinese]. Available from: ,https://www.agri.jl.gov. cn/xwfb/sxyw/201612/t20161230_4713102.html.; 2015. [40] Ran Y, Li Q, Peng DQ, Wang C, Tang XY, Luo XM. Economic benefits and technical characteristics analyses on the biomass solid fuel. J Anhui Agric 2015;322
5 [in Chinese]. [41] Song SZ, Liu P, Xu J, Chong CH, Huang XZ, Ma LW, et al. Life cycle assessment and economic evaluation of pellet fuel from corn straw in China: a case study in Jilin Province. Energy 2017;130:373
81. [42] Zheng SJ. Techno-economic analysis of cellulosic ethanol fuel [dissertation]. Beijing: Beijing University of Chemical Technology; 2011 [in Chinese]. [43] Yang ZT, Wu D, Liu WM, Shu KL, Zang XL, Shan KX. Maize straw resource utilization and countermeasures in Jilin province. J Maize Sci 2016;171
4 [in Chinese]. [44] Li L. A research on the development strategy of cellulosic ethanol in Tianjin [dissertation]. Tianjin: Tianjin University of Technology; 2015 [in Chinese]. [45] Gao R, Jiang W, Gao WD, Sun SD. Emission inventory of crop residue open burning and its high-resolution spatial distribution in 2014 for Shandong province. China Atmos Pollut Res 2017;8:545
54.

This page intentionally left blank

Chapter 14

Sustainability assessment of renewable energy-based hydrogen and ammonia pathways Yusuf Bicer and Farrukh Khalid Division of Sustainable Development (DSD), Hamad Bin Khalifa University (HBKU), Qatar Foundation (QF), Doha, Qatar

Chapter Outline 14.1 Introduction 435 14.1.1 Importance of energy storage 435 14.1.2 Chemical energy storage 436 14.2 Hydrogen and ammonia production pathways 439 14.2.1 Hydrogen production 439 14.2.2 Ammonia production 440 14.3 Methodology 448

14.3.1 Efficiency index 14.3.2 Cost 14.3.3 Environmental impact 14.3.4 Weighting scheme 14.4 Results and discussion 14.5 Conclusions Acknowledgements Nomenclature References

448 449 450 452 453 466 466 466 467

14.1 Introduction 14.1.1 Importance of energy storage While renewable energy resources are more commonly used in the near future, there arises a requirement of integrating alternative storage options. Storing energy in terms of chemicals presents a promising approach for several reasons such as transportability, minimal losses, and storage practices. In order to signify the importance of ammonia and hydrogen, the necessity of energy storage is discussed first. A massive interest has developed in the production of renewable energy as it has the ability to mitigate climate change. Due to the unpredicted behaviour of renewable sources such as Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00014-5 © 2021 Elsevier Inc. All rights reserved.

435

436

PART | IV Sustainability

sunlight intermittency, the benefits of energy storage systems become more significant, therefore, by implementing renewables and potential energy storage techniques, savings in emissions and conventional fuels can be achieved [1]. Consequently various energy storage techniques are essential to fulfill the global targets of climate change mitigation [2]. For the safety of the electricity grid, the quantity of electricity consumed by the end-users and produced by the producers is desired to be at a similar level to avoid grid damage and blackouts by balancing the supply and demand [3]. When there are cases of excess power generation, electricity is lost unless stored. This is where technologies of energy storage play an important role, as they are the solutions for balancing the demand and supply [3]. Energy can be stored in various forms such as thermal, electromagnetic, kinetic and chemical. In the following section, specifically, chemical energy storage is discussed.

14.1.2 Chemical energy storage Storage of chemicals is undoubtedly an appealing method especially for applications of renewables [4]. The highest quality form of energy is electricity, however, storing electricity directly is a challenging task. Battery technology has some limitations due to materials, degradation and cycle life. Chemicals such as hydrogen and ammonia can also act as a fuel for electricity generation through several routes such as fuel cells or combustion. In addition, combustion of these chemicals releases heat, which can be used for industrial processes where heat is vastly needed (e.g. for endothermic reactions) [4]. Moreover, hydrogen and ammonia contain zero-carbon atoms, which release zero CO2 emissions during utilization. There are specific hydrogen and ammonia fuel cells, which can receive them as fuel and convert into electricity with considerably higher efficiency than combustion. However, the production pathway of hydrogen and ammonia matters due to environmental impact. They are desired to be produced from renewable routes to make them environmentally benign. The following sections explain the specifications of hydrogen and ammonia along with renewable-based production pathways.

14.1.2.1 Renewable hydrogen (H2) Hydrogen bears one of the largest heating values (per mass) chemicals corresponding to about 120 MJ/kg as shown in Table 14.1. The specific energy of hydrogen is very high compared to other storage methods. Table 14.1 presents the thermophysical properties of hydrogen. 14.1.2.2 Renewable ammonia (NH3) Ammonia is produced by combining nitrogen and hydrogen with the help of any energy source. Hydrogen is obtained from water and nitrogen is

Sustainability assessment of renewable energy-based Chapter | 14

437

TABLE 14.1 Thermophysical properties of hydrogen. Property

Description 3

Density (kg/m )

0.1 (at 1 atm, 298 K for gas) 1.34 (at 1 atm, 20.3 K for vapor) 70.79 (at 1 atm, 20.3 K) 11.69 (at 350 atm, 298 K for gas)

Composition

75% ortho/25% para (at 298.15 K)More than 99.8% para (for cryogenic H2)

Lower heating value (MJ/kg)

119.9

Higher heating value (MJ/kg)

141.8

Chemical exergy (MJ/kg)

118.05

Autoignition temperature (K)

850

Flammability

4%
18% and 59%
75% mix with air

Critical temperature (K)

32.97

Critical pressure (bar)

12.9

Normal boiling point (K)

20.3 (at 1 atm) 99.8% para-hydrogen 1 0.02% ortho-hydrogen

Melting point (K)

14.01 (at 1 atm)

Thermal conductivity (W/mK)

0.177

Diffusivity in air (0.61 cm/s)

0.61

Detonation

13%
65% mix with air with 2 km/s

Flame speed (m/s)

2.75

Source: Taken from Dincer and Zamfirescu [6].

generally extracted from the air. The thermophysical properties of liquid ammonia are given in Table 14.2. Presently, Haber
Bosch process is the prime method that is used to produce ammonia globally [4,5]. In Haber
Bosch process, hydrogen and nitrogen react at very high pressure (200
250 atm) and moderate temperature (450 C
600 C) under the influence of catalyst namely iron oxide to produce ammonia. To enhance the efficiency of ammonia production and reduction in its cost, several new techniques (solid-state synthesis, thermochemical processes) are currently being studied. The feedstock for ammonia depends on the hydrogen source and it can be obtained from several petroleum products (naphtha, heavy fuel oil, etc.) and fossil fuels such as coal, and natural gas.

438

PART | IV Sustainability

TABLE 14.2 Thermophysical properties of liquid ammonia. Property

Description

Colour

Colourless 3

Density (kg/m )

638.6 (at 273.15 K, 1 atm) 682 kg/m3 (at 239.72 K, 1 atm)

Boiling point (K)

239.72 (at 1 atm)

Melting point (K)

195.44 (at 1 atm)

Critical temperature (K)

405.55

Critical pressure (MPa)

11.28

Critical viscosity (mPa s)

23.90 3 1023

Lower heating value (MJ/kg)

18.57

Higher heating value (MJ/kg)

22.54

Source: from Kaudy L, Rounsaville JF, Schulz G. Ullmann’s encyclopedia of industrial chemistry. 5th ed. Weinheim, Germany: VCH; n.d.

In ammonia production, the initial step is to generate hydrogen from these feedstocks and then combine with nitrogen (generally obtained from air separation unit). In gas turbines, combustion engines, burners and in many others, ammonia can be used as a sustainable fuel with only minor changes. However, in fuel cells ammonia can be directly used, which is a critical benefit as it is not in the case of other fuels [7]. Ammonia can be seen as a hydrogen carrier, which has better storage characteristics in terms of temperature and pressure. It is liquid at about 234 C at ambient pressure. During the utilization of ammonia, no greenhouse gases are emitted. Ammonia can be used as a cooling agent to cool the engine while it is being preheated to the temperature at which it is fed to the inlet of the combustion unit or fuel cell in the vehicle. By using ammonia (carbon-neutral fuel) as a fuel for marine transport, total greenhouse gas emissions can be reduced. Ammonia can be directly fed to engines of marine ships as a standalone or supplementary fuel. During operation, marine engine running on ammonia exhibits less greenhouse gas emissions resulting in lower environmental impact. Utilizing renewable energy sources such as solar, hydropower, and wind, ammonia can be produced. In this way, hydrogen is commonly produced from electrolysis. In this work, photoelectrochemical hydrogen production will also be considered.

Sustainability assessment of renewable energy-based Chapter | 14

439

14.2 Hydrogen and ammonia production pathways 14.2.1 Hydrogen production For sustainability assessment, the following hydrogen production methods are considered in this study: G G G G

Steam methane reforming (SMR) Wind power-based electrolysis Hydropower-based electrolysis Photoelectrochemical water splitting

14.2.1.1 Steam methane reforming Methane (or mainly natural gas) reforming by using steam at high temperatures (700 C
1000 C) is a well-known method. In this method, under the influence of a catalyst, methane reacts with high-temperature steam (at the pressure of 300
2500 kPa) to mainly produce carbon monoxide, hydrogen and a relatively minor fraction of carbon dioxide. Reforming of steam requires heat, which means that the process is endothermic. Mostly, the steam turbine exit from the power plant is used to provide the desired heat required for the steam reforming. 14.2.1.2 Wind power-based electrolysis Hydrogen production from wind energy is conducted through electrolysis of water, where the source of energy comes from a wind turbine. Hydrogen production via wind turbine requires mainly two devices that are a wind turbine (to produce electricity) and an electrolyzer (to dissociate water). Wind turbines capture wind energy to produce mechanical work, which is then, converted to alternating current (AC) electricity. The produced AC electricity is transmitted to the electric grid. Wind turbines’ efficiency are greatly influenced by the location (mainly wind velocity). Thus, they need to be installed at the location where wind velocities are high. Hydrogen production by wind energy is a promising route among other renewables due to high capacity factors of wind turbines, higher efficiency and stability throughout the year. 14.2.1.3 Hydropower-based electrolysis In this route hydropower plant electricity is the main source of the electrolysis unit for hydrogen production. One of the main advantages of this route is having very high energy conversion efficiencies. In addition, harmful emissions are minimum for this route.

440

PART | IV Sustainability

14.2.1.4 Photoelectrochemical water splitting Hydrogen production using solar energy can be executed through artificial photosynthesis, photoelectrolysis, water electrolysis, photocatalytic, thermochemical and photoelectrochemical water splitting methods. In this study, a photoelectrochemical cell (PEC) is used for generating photocurrent in the cell and electrolyze the water. PEC transforms solar energy into hydrogen by a light-enthused electrochemical process. In PEC, sunlight is absorbed by one or both photoelectrodes (one should at least be a semiconductor). Energy generated by the PECs is chemical or electrical in nature based on the need of its usage.

14.2.2 Ammonia production Here, two different ammonia synthesis methods initially described are as follows: (A) Haber
Bosch ammonia synthesis The prime and well-acquainted method to produce ammonia is the Haber
Bosch process [8]. Fig. 14.1 illustrates the schematic of this process. When hydrogen and nitrogen under the influence of a catalyst react with each other in a ratio of 3:1, ammonia is produced. The methods used to produce the nitrogen and hydrogen dictates the environmental impact of ammonia production in the Haber
Bosch process. Before entering the ammonia synthesis loop, a mixture of hydrogen and nitrogen gas is compressed to 12
22 MPa, based on the type of plant [10]. In ammonia synthesis reactor, due to thermodynamic equilibrium, only a fraction of mixture is converted to ammonia in a single pass. The unreacted mixture gas is once again passed over the converter. This makes a flow loop for the unreacted gas. Afterward, the ammonia in gaseous form and unreacted mixture gas reaches the ammonia recovery section of the synthesis loop. The gasses are cooled down to 210 C to 225 C by using refrigeration coolers and this allows condensing ammonia out of the mixture and leaving unreacted synthetic gas behind [7]. (B) Electrochemical ammonia synthesis Typically, electrochemical routes allows reactions to be carried out at much lower pressures compared to those used in the Haber
Bosch process. In electrochemical processes for liquid and polymer electrolytes, the operating temperature is near room temperature while for other electrolytic routes it ranges between 673 K and 1073 K. The operation at lower temperatures has certain advantages such as the requirement of lower cost material and less operational cost. Operation at a lower temperature also increases the lifetime of the reactor and can provide a higher rate of ammonia production with more faradaic efficiencies. Ammonia production by the selected

FIGURE 14.1 A simple layout of the Haber
Bosch ammonia synthesis process. Modified from Appl M. Complete ammonia production plants. In: Appl M, editor. Ammonia: principles and industrial practice. Weinheim, Germany: Wiley-VCH Verlag GmbH; 2007. p. 177
204. http://dx.doi.org/10.1002/ 9783527613885.ch05; Appl M. Ammonia, 3. Production plants, In: Appl M, editor. Ullmann’s Encyclopedia of Industrial Chemistry, Weinheim, Germany: Wiley-VCH Verlag; 2011. http://dx.doi.org/10.1002/14356007.o02_o12.

442

PART | IV Sustainability

electrolytic route (molten salts) at medium temperature offers several advantages for example; these systems can be coupled with surplus heat from nuclear or renewable energy sources. In this study, the following ammonia production methods are considered for sustainability assessment: G G

G

G

SMR and Haber
Bosch ammonia synthesis method Wind power-based electrolysis and Haber
Bosch ammonia synthesis method Hydropower-based electrolysis and Haber
Bosch ammonia synthesis method Photoelectrochemical water splitting and electrochemical ammonia synthesis method

14.2.2.1 Steam methane reforming and Haber
Bosch ammonia synthesis method Fig. 14.2 illustrates the production of ammonia from SMR. Currently the reforming of methane using steam to produce ammonia is the prime method. It is a quite mature technique where high-temperature steam is being used to produce hydrogen from a methane source (mainly natural gas). In this process under the influence of catalyst, steam at high temperature (process pressure ranges 3
25 bar) reacts with methane to mainly produce carbon monoxide, hydrogen and a relatively minor fraction of carbon dioxide. This process is endothermic in nature. In the Haber
Bosh process, combining the nitrogen with produced hydrogen results in the production of ammonia. 14.2.2.2 Wind power-based electrolysis and Haber
Bosch ammonia synthesis process During ammonia synthesis, greenhouse emissions can be reduced drastically if renewable energy powered (solar, wind, hydropower, etc.) water electrolysis is employed to produce hydrogen. In an electrochemical cell, water can be used directly as a hydrogen source. Ammonia production from wind energy can be accomplished as shown in Fig. 14.3. Initially there is hydrogen production by an electrolyzer where the electricity is sourced from wind turbine generators. There are numerous types of wind turbines varying in size, shape, capacity, etc. Therefore it is an advantage in terms of plant design aspects due to scalability. Since the power produced from the wind turbine relies on the speed of the wind available at the location and season, the variability of power generation becomes a significant issue to resolve. Hence chemical energy storage can be used for balancing production and demand. Once the hydrogen is generated from the electrolyzer, it is reacted with nitrogen from an air separation unit in a Haber
Bosch synthesis reactor for ammonia production. The electricity required for the compressors is taken from the wind turbines as well.

FIGURE 14.2 Steam methane reforming and the Haber
Bosch ammonia synthesis process.

FIGURE 14.3 Wind power-based electrolysis and the Haber
Bosch ammonia synthesis.

Sustainability assessment of renewable energy-based Chapter | 14

445

14.2.2.3 Hydropower-based electrolysis and the Haber
Bosch ammonia synthesis Hydropower plants are commonly two types: reservoir and run-of-river. The energy conversion efficiencies of hydropower plants are very high compared to other renewables due to mechanical energy conversions. The kinetic or potential energy of water is converted into electricity through water turbines such as Kaplan and Francis types. In this method, the electricity is initially generated by the hydropower plant and used in the electrolysis unit for hydrogen production, air separation unit for nitrogen production and Haber
Bosch ammonia synthesis unit for the compressors as illustrated in Fig. 14.4. 14.2.2.4 Photoelectrochemical water splitting and electrochemical ammonia synthesis In this route, solar energy along with electricity is used in a PEC for hydrogen generation. The additional electricity supplied to PEC unit is taken from photovoltaic (PV) modules. The produced hydrogen is then fed into an electrochemical cell for ammonia production at ambient pressure and medium temperature as the diagram is shown in Fig. 14.5. Electrochemical processing for ammonia production is carried out under ambient conditions and at different temperatures depending on the used electrolyte. Waste heat from industrial processes can become an alternative source of thermal energy required for medium/high-temperature electrochemical routes. Especially concentrated solar energy is a promising candidate for environmentally friendly electrochemical ammonia synthesis. Because the synthesis process of ammonia is an exothermic reaction, it is more favoured at high pressures and low temperatures. Thus a balance between the operating temperature, pressure and the ammonia yield needs to be considered for the electrochemical synthesis unit. The ammonia synthesis that uses water as a hydrogen source in the electrochemical process is attractive and eliminates the hydrogen production step from water, however, most of the time, there is a need for short- and medium-term storage in renewable power plants. In addition, hydrogen can be needed as an intermediate for further processing. Therefore in these cases, hydrogen is better to be produced separately, which can act as short-term storage and converted into electricity by hydrogen fuel cells. It can be combined with nitrogen for ammonia production and ammonia can be used in ammonia fuel cells for power generation as well. In this study, we have considered the electrochemical synthesis of ammonia in a molten salt electrolyte using hydrogen and nitrogen at atmospheric pressure at medium temperatures (B200 C) using a nanoscale catalyst. The reaction temperature is much lower than the usual Haber
Bosch process. The electrochemical cells have a nickel anode and cathode where they are dipped in a molten hydroxide

FIGURE 14.4 Hydropower-based electrolysis and the Haber
Bosch ammonia synthesis.

FIGURE 14.5 Photoelectrochemical hydrogen production and electrochemical ammonia synthesis.

448

PART | IV Sustainability

electrolyte (KOH and NaOH eutectic mixture). To reach the melting point of the eutectic mixture, the reactor is heated by heating elements surrounding the alumina crucible. Once the reactants, hydrogen and nitrogen, are sent to the reactor through the ceramic tubes, the yielded mixture of gases are collected from the other ceramic tube and then sent to ammonia trap. In this way, ammonia is captured, and the unreacted gases can be recirculated.

14.3 Methodology In this study, the sustainability index is calculated by considering the following aspects of hydrogen and ammonia production methods: G G G

Efficiency: This index includes both energy and exergy efficiencies. Cost: This index represents the cost of production. Environmental impact: This index considers (1) global warming potential (GWP), (2) human toxicity (HT) and (3) abiotic depletion potential (ADP).

14.3.1 Efficiency index The efficiency index accounts for energy and exergy efficiencies as per the following equation:

 FEfficiency 5 Nηen 3 Wηen 1 Nηex 3 Wηex ð14:1Þ where Nηen and Nηex are the normalized energy and exergy efficiencies (between 0 and 1), respectively and Wηen and Wηex are the weighting factors allocated to energy and exergy efficiency index, respectively. For normalization of the efficiency values, the following approach is employed since efficiency is aimed to be maximized: η Nη 5 actual ð14:2Þ ηitarget Here, the current efficiency values are given in Table 14.3; however, for the target efficiencies, the systems are considered individually since there are different levels of system maturities. The target efficiencies are defined based on the literature and reports as shown in Table 14.3. The energy and exergy efficiency values of the selected hydrogen and ammonia production routes are taken as shown in Table 14.3 based on the literature. Note that the SMR efficiency represents the case with carbon capture storage.

14.3.1.1 Energy efficiency Energy systems differ in their efficiencies and production performance. Energy efficiency is one of the most important subindexes for sustainability

449

Sustainability assessment of renewable energy-based Chapter | 14

TABLE 14.3 Energy and exergy efficiencies values of the selected hydrogen and ammonia production routes. ηen Energy efficiency (%)

Route

ηex Exergy efficiency (%)

Hydrogen (actual/target)

Ammonia (actual/target)

Hydrogen (actual/target)

Ammonia (actual/target)

SMR

54/65 [7,10,11]

56/70 [7,10,11]

52/62 [7,10,11]

50/63 [7,10,11]

Hydro

63/75 [12]

58/75 [7,10,11]

61/73 [12]

56/74 [7,10,11]

Wind

31.5/45 [12]

32/45 [7,10,11]

30.5/44 [12]

30/44 [7,10,11]

PEC

6.6/20 [13]

3.92/17 [13]

6.7/14 [13]

4.1/15 [13]

PEC, Photoelectrochemical cell; SMR, steam methane reforming.

assessment. It represents the performance (in percentage) of converting a feed into a useful product. In this context, the useful outputs are hydrogen and ammonia, where the required inputs include the electricity/heat consumed and other material feeds.

14.3.1.2 Exergy efficiency Exergy is interpreted as useful work defined by the second law of thermodynamics. It represents more insights into the energy systems. The most appropriate link between the environmental impact and the second law of thermodynamics has been namely exergy, mainly because exergy is a measurement of the departure of the state of a system from that of the environment. In this context, exergy efficiency accounts for the exergy losses and exergy destructions, which yield a more accurate performance indicator. 14.3.2 Cost The cost index in this assessment represents the production cost of these two chemicals, hydrogen and ammonia. Since the economy is considered as a part of sustainability, the financial aspect of the selected routes is also considered in the assessment as per the following equation: FCost 5 NCP 3 Wcost

ð14:3Þ

where NCP is the normalized cost of production of ammonia and hydrogen and Wcost is the weighting factor allocated to the cost index. For normalization of the cost values, the following approach is followed because the cost is targeted to be minimized: NCP 5

Cp min Cp actual

ð14:4Þ

450

PART | IV Sustainability

TABLE 14.4 Cost of production values of selected hydrogen and ammonia production routes. Route

Cp, Cost of production ($/kg) Hydrogen

Ammonia

SMR

2.2 [14]

0.44 [14]

Hydro

2.12 [14]

0.424 [14]

Wind

3 [14]

0.6 [14]

PEC

3.24 [13,14]

0.84 [13,14]

PEC, Photoelectrochemical cell; SMR, steam methane reforming.

here the minimum cost of production is taken for the hydropower-based route as 2.12 $/kg for hydrogen and 0.424 $/kg for ammonia since they are the lowest costs among other selected methods. The cost of production values of each selected route are written in Table 14.4 based on the literature.

14.3.3 Environmental impact The environmental impact index in this study consists of different environmental categories. The degree of influence of the energy system on its surrounding environment is a critical factor when considering its sustainability. This index exemplifies the level of suitability of the selected routes in terms of environmental acceptability. The environmental impact values of selected routes are given in Table 14.5. The following equation is used to calculate the environmental impact index: FEI 5 NEIGWP 3 WGWP 1 NEIADP 3 WADP 1 NEIHT 3 WHT

ð14:5Þ

where NEI is the normalized environmental impact of ammonia and hydrogen production and WGWP, WADP and WHT represent the weighting factors allocated to global warming, abiotic depletion and human toxicity categories, respectively. For normalization of the environmental impact values, the following equation is used because, the environmental impact is aimed to be minimized: NEI 5

EImin EIactual

ð14:6Þ

TABLE 14.5 Environmental impact values of selected hydrogen and ammonia production routes. Route

ADP

GWP

HT

Hydrogen

Ammonia

Hydrogen

Ammonia

Hydrogen

Ammonia

SMR

0.135 [15]

0.0187 [13]

9.05 [15]

1.83 [13]

0.479 [15]

0.707 [13]

Hydro

0.0016 [13]

0.0028 [13]

0.368 [12]

0.368 [13]

0.0141 [13,16]

0.129 [13]

Wind

0.002 [17]

0.0035 [13]

2.35 [12]

0.467 [13]

0.0782 [17]

0.713 [13]

PEC

0.00468 [13]

0.0082 [13]

5.485 [13]

1.09 [13]

0.0981 [13]

0.9 [13]

ADP, Abiotic depletion potential; GWP, global warming potential; HT, human toxicity; PEC, photoelectrochemical cell; SMR, Steam methane reforming.

452

PART | IV Sustainability

here the minimum environmental impact for the global warming category is taken for wind power route corresponding to approximately 2.35 kg CO2 eq. for hydrogen and 0.368 kg CO2 eq. for wind-powered ammonia production per unit mass. On the other hand, other impact categories are lower for the hydropower-based route corresponding to ADP of 0.0016 kg Sb eq. and HT of 0.0141 kg 1,4 DB eq. for hydrogen production and ADP of 0.0028 kg Sb eq. and HT of 0.129 kg 1,4 DB eq. for ammonia production from hydropower.

14.3.4 Weighting scheme To obtain the final sustainability index, weighting factors are applied. Weighted arithmetic mean aggregates the values, hence it is used in this study. The summation of weighting factors is 100%. The weighting scheme is initially determined equal weighting as listed in Table 14.6. Later based on the significance of the criteria, they are altered. In a balanced weighing scheme, a more sustainable approach is followed giving more importance to efficiency and environmental impact. In addition, the weighting scheme is changed depending on the dominancy of cost, environment or efficiency. The weighting factors used in this study are comparatively listed in Table 14.6. The overall sustainability index is calculated using the following equation: SI 5 Fefficiency 1 FEI 1 FCost

ð14:7Þ

TABLE 14.6 Various weighting schemes and values used during the assessment. Factor

Equal

Balanced

Environment dominant

Efficiency dominant

Cost dominant

ADP

0.111

0.15

0.25

0.1

0.1

Cost

0.333

0.3

0.1

0.1

0.6

Energy efficiency

0.166

0.1

0.1

0.3

0.05

Exergy efficiency

0.166

0.15

0.05

0.3

0.05

GWP

0.111

0.2

0.25

0.1

0.1

HT

0.111

0.1

0.25

0.1

0.1

ADP, Abiotic depletion potential; GWP, global warming potential; HT, human toxicity.

Sustainability assessment of renewable energy-based Chapter | 14

453

14.4 Results and discussion The summary of the sustainability index values of various hydrogen and ammonia production methods subjected to different weighing criteria is shown in Fig. 14.6. It is evident from the figure that for equal weighing criteria, hydrogen and ammonia produced from hydropower are the most sustainable while PEC hydrogen and PEC ammonia production methods are the least sustainable ones. For balanced, efficiency dominant and cost dominant weighting criteria, hydrogen and ammonia produced from PEC are the least sustainable ones while hydropower ammonia and hydrogen production methods are the most sustainable ones. The main reason for the PEC to be the least sustainable is the poor solar-to-fuel conversion efficiency (Table 14.3). However, this situation changes for the case of environment dominant criteria where hydrogen produced from SMR is the least sustainable because of more ADP and GWP values (Table 14.5). Furthermore it is evident from Table 14.5, that the HT value for the PEC ammonia is the highest. This higher value leads the PEC ammonia to be the least sustainable for the environmentally dominant criteria (Fig. 14.6). Fig. 14.7 shows the effect of energy efficiency on the sustainability of various ammonia and hydrogen production methods for environment dominant criteria. As the energy efficiency changes, the sustainability index of all the methods increases. This result shows that conversion efficiency plays an important role in the sustainability of the process. Thus, one needs to operate its process at higher efficiency for better sustainability. It is also to be noted that for the PEC process, there are no results after 20% energy efficiency. The reason for this missing trend is that 20% efficiency is the initial target efficiency for the PEC process. Thus having energy efficiency values more than target efficiency (20%) is not viable. For environment dominant criteria, SMR-based hydrogen production is the least sustainable and this situation remains unchanged despite the increase in the energy efficiency of the process. The main reason for this trend is that in the case of the SMR process for hydrogen production, the ADP, GWP and HT (Table 14.5) are the highest among all the studied hydrogen production methods in this study. For ammonia production methods, PEC is the least sustainable one. However, this trend changes when the energy efficiency of the PEC process becomes more than 14.0%. For PEC energy efficiency of more than 14.0%, the ammonia produced from this method no longer remains the least sustainable as it supersedes the SMR-based ammonia production method (Fig. 14.7). Furthermore it is evident from the figure that hydropower-based ammonia and hydrogen production methods are the most sustainable ones, irrespective of an increase in energy efficiency for all processes. This trend can be somewhat explained by the fact that among all the studied methods for ammonia and hydrogen production in this study, hydropower has the least environmental impact (ADP, GWP and HT) values (Table 14.5).

FIGURE 14.6 Summary of the sustainability index values of various hydrogen and ammonia production routes based on different weighting schemes (equal, balanced, cost, efficiency or environment dominant).

FIGURE 14.7 The effects of energy efficiency subindex on the final sustainability index of various hydrogen and ammonia production routes subjected to environment dominant criteria.

456

PART | IV Sustainability

The effect of exergy efficiency on the sustainability index of various hydrogen and ammonia production methods studied in this study subjected to environment dominant criteria is plotted in Fig. 14.8. As the exergy efficiency increases, the sustainability of all the processes rises. These results indicate that for better sustainability, processes need to be exergetically efficient. It is to be noticed from the figure that for PEC exergy efficiency more than 13.0% (approximate value), PEC-based ammonia production method no longer remains the least sustainable one as it supersedes the SMR-based ammonia production method. For PEC exergy efficiency values less than 13.0% (approximate value), PEC-based ammonia production method found to be the least sustainable one. For hydropower-based hydrogen and ammonia production methods, similar trend is found as observed in the case of energy efficiency, that is ammonia and hydrogen produced by the hydropower are the most sustainable ones. Variation of the sustainable index of hydrogen production methods used in this study with the cost of hydrogen production subject to criteria that is environment dominant is plotted in Fig. 14.9. As the cost of hydrogen production increases, the sustainability of all the processes decreases. These results suggest that for any process to be sustainable, it should be economically viable. Hydropower-based hydrogen production method is the most sustainable while the hydrogen produced from SMR is the least sustainable despite low cost of hydrogen production (SMR-based hydrogen production trend never supersedes the trends of other hydrogen production methods). Variation of the sustainable index of ammonia production methods used in this study with the cost of ammonia production subject to criteria that is environment dominant is plotted in Fig. 14.10. As the cost of ammonia production increases, the sustainability of all the processes decreases. These results suggest that for any process to be sustainable it should be economically viable. Hydropower-based ammonia production method is the most sustainable while the ammonia produced from PEC is the least sustainable. It is to be noted that in Fig. 14.10, the trend of PEC-based ammonia production never crosses the trend of SMR-based ammonia production. This can be somewhat explained by the fact that in this study energy and exergy efficiencies values of 3.92% and 4.1%, respectively, are used and these values are kept fixed while generating Fig. 14.10. However, the trend of PEC-based ammonia production can supersede the trend of SMR-based ammonia production if the conversion efficiency (energy and exergy efficiencies) of the process becomes more than 14.0%. The effects of global warming subindex on the final sustainability index of various hydrogen production routes subjected to environment dominant criteria are shown in Fig. 14.11. It is evident from the figure that the sustainability index declines as the GWP rises. The sustainability index of hydrogen produced from hydropower is the highest while the SMR-based hydrogen production method has the lowest sustainability index. The PEC-based

FIGURE 14.8 The effects of exergy efficiency subindex on the final sustainability index of various hydrogen and ammonia production routes subjected to environment dominant criteria.

FIGURE 14.9 The effects of the cost of hydrogen production on the final sustainability index of various hydrogen production routes subjected to environment dominant criteria.

FIGURE 14.10 The effects of the cost of ammonia production on the final sustainability index of various ammonia production routes subjected to environment dominant criteria.

FIGURE 14.11 The effects of global warming subindex on the final sustainability index of various hydrogen production routes subjected to environment dominant criteria.

FIGURE 14.12 The effects of global warming subindex on the final sustainability index of various ammonia production routes subjected to environment dominant criteria.

462

PART | IV Sustainability

hydrogen production method has a sustainability index very closed to SMRbased hydrogen because of the poor efficiency of the PEC method (Table 14.3). The effects of global warming subindex on the final sustainability index of various ammonia production routes subjected to environment dominant criteria are shown in Fig. 14.12. It is evident from the figure that the sustainability index declines as the GWP rises. The sustainability index of ammonia produced from hydropower is the highest while the PEC-based ammonia production method has the lowest sustainability index. The lowest sustainability index of PEC-based ammonia production method is somewhat explained by the fact that for PEC ammonia, HT is the highest among all the ammonia production method (Table 14.5). This is mainly due to photovoltaic panels employed to supply additional electricity to the PEC unit. Figs. 14.13 and 14.14 show the effects of ADP on the sustainability index of various hydrogen and ammonia production methods subjected to environment dominant criteria, respectively. It is evident from the figure that PECbased methods have the lowest sustainability whereas SMR-based methods have the highest sustainability. The reason for the PEC to have the lowest sustainability is explained earlier. With the increase in ADP, the sustainability of all the methods for hydrogen and ammonia production decreases. The effects of HT on the final sustainability index of various hydrogen production routes subjected to environment dominant criteria are shown in Fig. 14.15. It is evident from the figure that the sustainability index declines as the HT potential rises. The sustainability index of hydrogen produced

FIGURE 14.13 The effects of abiotic depletion potential subindex on the final sustainability index of various hydrogen production routes subjected to environment dominant criteria.

FIGURE 14.14 The effects of abiotic depletion potential subindex on the final sustainability index of various ammonia production routes subjected to environment dominant criteria.

FIGURE 14.15 The effects of human toxicity subindex on the final sustainability index of various hydrogen production routes subjected to environment dominant criteria.

FIGURE 14.16 The effects of human toxicity subindex on the final sustainability index of various ammonia production routes subjected to environment dominant criteria.

466

PART | IV Sustainability

from hydropower is the highest while the SMR-based hydrogen production method has the lowest sustainability index. The PEC-based hydrogen production method has a sustainability index close to SMR-based hydrogen because of the poor efficiency of the PEC method (Table 14.3). The effects of HT on the final sustainability index of various ammonia production routes subjected to environment dominant criteria are shown in Fig. 14.16. It is evident from the figure that the sustainability index declines as the GWP rises. The sustainability index of ammonia produced from hydropower is the highest while the PEC-based ammonia production method has the lowest sustainability index. The lowest sustainability index of PECbased ammonia production method is somewhat explained by the fact that for PEC ammonia, HT is the highest among all the ammonia production method (Table 14.5).

14.5 Conclusions Hydrogen and ammonia are carbon-free energy carriers suitable for various applications including energy storage. This study performs a sustainability assessment of various hydrogen and ammonia production pathways from environment, cost and efficiency perspectives. To assess sustainability, a new index is introduced. The newly introduced index is subject to various criteria such as equal, balanced, efficiency dominant, environment dominant and cost dominant. The results in the present work show that hydrogen produced from the SMR method has the lowest sustainability index when the environment-dominant weighing scheme is considered. However, PEC-based ammonia production has lower sustainability index values due to low energy conversion efficiency. Furthermore it is found that if the efficiency of the PEC process becomes more than 14%, the sustainability index surpasses some other methods, which emphasizes the importance of efficiency improvement in photoelectrochemical processes.

Acknowledgements The authors acknowledge Hamad Bin Khalifa University, Qatar Foundation, Doha, Qatar, for financial support.

Nomenclature Cp EI F N SI W

Cost of production Normalized environmental impact Factor Normalized Sustainability index Weighting

Sustainability assessment of renewable energy-based Chapter | 14

467

Abbreviations ADP DB GHG GWP HT PEC SMR

Abiotic depletion potential Dichlorobenzene Greenhouse gas Global warming potential Human toxicity potentials Photoelectrochemical Steam methane reforming

Greek letters η

Efficiency

Subscripts en ex

Energy Exergy

References [1] Petrillo A, De Felice F, Jannelli E, Minutillo M, Lubrano A, Lubrano A. Life cycle assessment (LCA) and life cycle cost (LCC) analysis model for a stand-alone hybrid renewable energy system. Renew Energy 2016;95:337
55. Available from: https://doi.org/10.1016/J. RENENE.2016.04.027. [2] Kapila S, Oni AO, Gemechu ED, Kumar A. Development of net energy ratios and life cycle greenhouse gas emissions of large-scale mechanical energy storage systems. Energy 2019;170:592
603. Available from: https://doi.org/10.1016/J.ENERGY.2018.12.183. [3] B. Zohuri, B. Zohuri. Energy storage technologies and their role in renewable Integration. In: Small modular reactors as renewable energy sources. 2019. doi:10.1007/978-3-31992594-3_8. [4] Dincer I, Bicer Y. Integrated energy systems for multigeneration. Amsterdam, Netherlands: Elsevier; 2020. Available from: http://dx.doi.org/10.1016/C2015-0-06233-5. [5] Dincer I, Zamfirescu C. Fossil fuels and alternatives [chapter 3]. In: Zamfirescu ID, editor. Advanced power generating system. Boston: Elsevier; 2014. p. 95
141. Available from: http://dx.doi.org/10.1016/B978-0-12-383860-5.00003-1. [6] L. Kaudy, J.F. Rounsaville, G. Schulz, Ullmann’s Encyclopedia of Industrial Chemistry, 5th, VCH, n.d. [7] L. Kaudy, J.F. Rounsaville, G. Schulz. Ullmann’s Encyclopedia of Industrial Chemistry. 5th. Weinheim, Germany: VCH; n.d. doi: 10.1016/j.ijhydene.2012.02.133. [8] Appl M. Complete ammonia production plants. In: Appl M, editor. Ammonia: principles and industrial practice. Wiley-VCH Verlag GmbH: Weinheim, Germany; 2007. p. 177
204. Available from: http://dx.doi.org/10.1002/9783527613885.ch05. [9] Appl M. Ammonia, 3. Production plants, In: Appl M, editor. Ullmann’s Encyclopedia of Industrial Chemistry, Weinheim, Germany: Wiley-VCH Verlag; 2011. doi:10.1002/ 14356007.o02_o12.

468

PART | IV Sustainability

[10] Acar C, Dincer I. Comparative assessment of hydrogen production methods from renewable and non-renewable sources. Int J Hydrog Energy 2014;39:1
12. Available from: https://doi.org/10.1016/j.ijhydene.2013.10.060. [11] DOE Hydrogen and Fuel Cells Program. DOE H2A Production Analysis, H2A Cent. Hydrog. Prod. Model. Version 3.1. ,https://www.hydrogen.energy.gov/h2a_production. html.; n.d. (accessed 07.01.2017). [12] Acar C, Dincer I. Hydrogen production. Comprehensive energy systems. Elsevier Inc; 2018. p. 1
40. Available from: http://dx.doi.org/10.1016/B978-0-12-809597-3.00304-7. [13] Bicer Y. Available from: https://ir.library.dc-uoit.ca/handle/10155/780Investigation of novel ammonia production options using photoelectrochemical hydrogen. University of Ontario Institute of Technology; 2017 (accessed 14.11.2018). [14] Dincer I, Bicer Y. Ammonia production. In: Dincer I, editor. Comprehensive Energy Systems. Elsevier; 2018. p. 41
94. Available from: http://dx.doi.org/10.1016/B978-0-12809597-3.00305-9. [15] Bicer Y, Dincer I. Life cycle assessment of nuclear-based hydrogen and ammonia production options: a comparative evaluation. Int J Hydrog Energy 2017;42:21559
70. Available from: https://doi.org/10.1016/j.ijhydene.2017.02.002. [16] Siddiqui O, Dincer I. Comparative assessment of the environmental impacts of nuclear, wind and hydro-electric power plants in Ontario: a life cycle assessment. J Clean Prod 2017;164:848
60. Available from: https://doi.org/10.1016/J.JCLEPRO.2017.06.237. [17] Suleman F, Dincer I, Agelin-Chaab M. Environmental impact assessment and comparison of some hydrogen production options. Int J Hydrog Energy 2015;40:6976
87. Available from: https://doi.org/10.1016/j.ijhydene.2015.03.123.

Chapter 15

An extended fuzzy divergence measure-based technique for order preference by similarity to ideal solution method for renewable energy investments Pratibha Rani1, Arunodaya Raj Mishra2, Abbas Mardani3, Fausto Cavallaro4, Raghunathan Krishankumar5 and Dalia Streimikiene6 1

Department of Mathematics, National Institute of Technology, Warangal, India, 2Department of Mathematics, Government College Jaitwara, Satna, India, 3Department of Marketing, College of Business Administration, University of South Florida, Tampa, FL United States, 4Department of Economics, University of Molise, Campobasso, Italy, 5School of Computing, SASTRA University, Thanjavur, India, 6Lithuanian Energy Institute, Kaunas, Lithuania

Chapter Outline 15.1 Introduction 469 15.2 Prerequisites 472 15.3 Divergence measures for fuzzy sets 475 15.3.1 An example for developed fuzzy divergence measures 479

15.4 Divergence measures-based fuzzy TOPSIS method 480 15.4.1 Case study of renewable energy investment 481 15.5 Conclusions 485 References 488

15.1 Introduction Recent decades have witnessed a meaningful shift of focuses from conventionally used forms of energy to renewable energy sources (RESs). Such a significant shift is due to destructive impacts of power plants that use conventional fossil fuels. Air and water pollution and climate change are damaging outcomes of the conventional approaches to the energy issue [1,2]. In addition, due to nonrenewability of fossil fuels, majority of governments are thinking of renewable energy (RE) investment and giving incentives to Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00015-7 © 2021 Elsevier Inc. All rights reserved.

469

470

PART | IV Sustainability

promote it effectively. The challenging issue is that fossil fuels are still the most widely used source of energy across the world. Unquestionably energy plays the main role in the achievement of the interrelated objectives of the current world; it satisfies the human beings’ requirement for cooling, heating, transportation, lighting, running a variety of appliances and providing energy for production systems. Since the energy crisis commenced, findings efficient ways to satisfy such needs has been a routine challenge and the solution has been primarily a matter of technology and money. As energy planning has direct and significant effects on the interests of numerous actors, any policy alternative cannot be suggested or implemented without considering the preferences of such stakeholders [3
5]. Various groups of experts are involved in these procedures, each of which has various criteria and perspective in accordance to its interests. This needs to be resolved in an understanding and mutual cooperation framework [6,7]. All actors on this stage, including environmental groups, local authorities/communities, academic institutions, potential investors, governments, etc. have their own interests, which have direct and indirect impacts on the processes of decision making [6,8
10]. In recent years, many researchers and practitioners have shown a great interest in exploiting RESs. In general, RE refers to the form of energy produced from natural sources such as wind, biomass, sunlight, sea tides, rain and geothermal heat. RES are expected to overcome the barriers that have appeared in environmental, economic, social, institutional and technical issues. The complicated issue of RES has brought about the challenge of selection from amongst a variety of exploitation proposals. The conception of fuzzy set (FS) pioneered by Zadeh [11] has received widespread concentration from various researchers, owing to its efficiency to deal with uncertain circumstances of real world. Later on, a variety of extensions of FSs have been proposed. Numerous scholars have widely taken into consideration the FSs doctrine and their applications. FSs have been found greatly efficient in the quantification of the similarities/dissimilarities between two items using a number of effective functions. Although, according to Montes et al. [12,13], the divergence definition is of a higher limitation compared with other comparison measures, which is completely essential to avoid counter intuitive situations. Encouraged by the divergence between two probability distributions, some researchers such as Montes et al. [14], Fan and Xie [15], Shang and Jiang [16] and Bhandari and Pal [17] have designed different divergence measures between FSs. Amongst them, Shang and Jiang [16] and Bhandari and Pal [17] took into account the logarithmic information gain functions in designing their divergence measures. Pal and Pal [18] conducted a critical analysis on the Shannon’s function and deliberated a number of limitations on that function. For the purpose of imbuing such limitations, Pal and Pal [18] defined a novel entropy measure on the basis of exponential function.

An extended fuzzy divergence measure-based technique Chapter | 15

471

Furthermore, Hwang and Yang [19], Verma and Sharma [20] and Mishra et al. [21,22] defined different version of entropy for FSs. A fuzzy divergence measure was introduced by Fan and Xie [15] on the basis of exponential function. In addition, they investigated its relationship with the divergence measure formerly proposed by Bhandari and Pal [17]. Chaira and Ray [23] and Mishra et al. [21,24] developed various divergence measures on the basis of exponential and logarithmic function. The applications of these measures in pattern recognition, medical diagnosis and crop disease diagnosis are also studied. Hooda et al. [25], Mishra et al. [26] and Hooda and Mishra [27] introduced entropy and divergence measures based on trigonometric function and applied it in strategic decision-making problems. There are many other efforts to define divergence between FSs along with applications in image segmentation [15,23,28
31]; Poletti et al. [32] studied its application in bioinformatics. A number of applications of fuzzy divergence measure was applied by Ghosh et al. [33] to the automated leukocyte recognition. In another project, Parkash et al. [34] designed a couple of divergence measures that were corresponding to Ferreri [35] probabilistic measure of directed divergence. Tomar and Ohlan [36] carried out a research on a sequence of fuzzy mean difference divergence measures through the establishment of inequalities amongst them and provided their application in the pattern recognition context and the consistency in the linguistic variables. Here, based on the Hwang and Yang [19] and Verma and Sharma [20], we introduce divergence measures for FSs and discuss some elegant characterization. An example of pattern recognition and multicriteria decision-making problem is used to show the high capacity of this extension. In general, decision-making processes are applied to the selection of the best option(s) from amongst a limited number of possible ones. This process commonly takes place during our daily life; it has an important effect on finance, management, business, political/social science, computer/engineering science, medicine, biology, etc. Multicriteria decision making (MCDM) deals with the selection of the optimal option(s) from amongst all viable options considering a variety of criteria. To do this, a decision maker takes into account the preference information for the option(s). In case of numerous actual decision-making situations, there is only some uncertain or inaccurate information to use. For an efficient simplification of the decision-making problems that involve such information, the FS doctrine has been recognized as a prevalent tool. Literature is consisted of different MCDM approaches proposed for efficiently solving the MCDM problems under different uncertain environments [31,37
42]. In recent times, TOPSIS method has been utilized by various authors to solve the MCDM problems under different fuzzy environments [43
46]. The TOPSIS approach is one of the easy and valuable techniques to deal with the decision-making issues. It works based on the minimum distance from the positive-ideal solution (PIS) and the maximum distance from the

472

PART | IV Sustainability

negative-ideal solution (NIS) for the purpose of choosing a most suitable alternative. The TOPSIS method has many benefits amongst which the principal ones are simplified steps for doing required computations, effective computation aptitudes, simplified numerical structure for signifying the relationships amongst the evaluation alternatives and applicability of various criteria to optimization processes. Because of its benefits, in this study, the authors extend the TOPSIS approach within the FSs context and examine its applicability through applying to RE investment problem. Since FSs have a vital effect on the investigation of uncertainty, the present study is focused on fuzzy environments. Here, an innovative approach is introduced for an objective determination of the attribute weights and for the aim of sorting existing alternatives in situations where the attribute weights are entirely unidentified or partially known and the attribute values are in the form of FSs. Now, the key outcomes of the present study are as follows: G G

G G

Divergence measures for FSs are reviewed and proposed. To tackle the MCDM problems, a classical TOPSIS method based on proposed divergence measure is extended within the context of FSs. To find the weights criteria, entropy method is utilized. A decision-making problem of RE investment is discussed to explore the applicability and validity of the developed fuzzy TOPSIS method. Comparison with existing methods shows the reliability of the proposed method.

The rest of this chapter is organized as follows. Some fundamental conceptions of FS, divergence measure, distance measure and entropy for FSs are discussed in Section 15.2. In Section 15.3, divergence measures based on exponential function for FSs are proposed. Divergence measures-based TOPSIS method for FSs is developed in Section 15.4. In Section 15.5, an illustrative RE investment problem is implemented to reveal the validity and applicability of the developed approach and then, compared the ranking obtained by the developed method with existing methods. Concluding remarks are presented in Section 15.6.

15.2 Prerequisites For any probability distribution Q 5 (q1, q2,. . ., qm)AΔm, the Shannon’s entropy [47] is described as follows: HðQÞ 5 2

m X

qi logqi :

ð15:1Þ

i51

Pal and Pal [18] proposed another measure called exponential entropy is given by

An extended fuzzy divergence measure-based technique Chapter | 15

HPal ðQÞ 5

m X

qi eð12qi Þ 2 1:

473

ð15:2Þ

i51

Pal and Pal [18] pointed out that the exponential entropy has an advantage
over Shannon’s entropy. For the uniform probability distribution  Q 5 m1 ; m1 ; . . .; m1 , the exponential entropy has a fixed upper bound
 limm-N H m1 ; m1 ; . . .; m1 5 e 2 1, which is not possible for Shannon’s entropy. Later on, Kullback and Leibler [48] introduced a divergence measure between two probability distributions Q and S, given as CKL ðQjjSÞ 5

m X i51

qi log

qi : si

ð15:3Þ

Its symmetric version, that is, Jeffreys invariant is given by J ðQjjSÞ 5 CKL ðQjjSÞ 1 CKL ðSjjQÞ:

ð15:4Þ

Definition 15.2.1 [11]: Let V 5 fv1 ; v2 ; . . .; vm g be the finite universal set. Then a FS F defined on V is given as 
  F 5 vi ; μF ðvi Þ :0 # μF ðvi Þ # 1; vi AV ;
 where the function μF ðvi Þ 0 # μF ðvi Þ # 1 is the membership degree of vi to F in V. All over the present research, R1 5 [0,N], and we consider FS(V) be the set of all FSs on the universe of discourse V and P(V) be the set of all crisp sets on the universe of discourse V. The term [a] is the FS of V for which μ½a ðvi Þ 5 a; ’vi AV ðaA½0; 1Þ. The term Fc denotes the complement of F; i:e:; μF c ðvi Þ 5 1 2 μF ðvi Þ; ’v  i AV: The union  F , G of the sets F and G is defined as μF , G ðvi Þ 5 max μF ðvi Þ; μG ðvi Þ ; ’vi AV and the intersection F -G of the  sets F and G is expressed as μF - G ðvi Þ 5 min μF ðvi Þ; μG ðvi Þ ; ’vi AV. Discrimination measure or cross entropy is utilized for the purpose of measuring the discrimination information. Based on the Shannon’s inequality, Bhandari and Pal [17] defined the fuzzy cross entropy. Definition 15.2.2: For F, GAFSs(V), I(F, G) is called discrimination or cross entropy between F and G if it satisfies the following postulates: (D1). I(F,G) $ 0, (D2). I(F,G) 5 0 iff F 5 G. A distance measure is used for showing the differences that exist amongst FSs. Liu [49] offered the axiom definition of distance measure on general universal set. Definition 15.2.3 [49]: A real function d:FS(V) 3 FS(V)-R1 is called a distance measure, if d have the following properties: (C1). d(F, G) 5 d(G, F), ’F,GAFSs(V); (C2). d(F, F) 5 0, ’FAFS(V);

474

PART | IV Sustainability

(C3). dðF; F c Þ 5

max

F;GAFSsðVÞ

dðF; GÞ;

(C4). If FCGCC, then d(F, C) $ d(F, G) and d(F, C) $ d(G, C). The entropy of a FS is a measure of fuzziness of the FS. First, Pal and Pal [18] introduced the exponential fuzzy entropy measure based on Eq. (15.2), which has m h i X
 1  μF ðvi Þeð12μF ðvi ÞÞ 1 1 2 μF ðvi Þ eμF ðvi Þ 2 1 : ð15:5Þ HP ðF Þ 5
1=2 m e 2 1 i51 Hwang and Yang [19] proposed the exponential entropy measure given by  m 
X 
 1 2μFc ðvi Þ  2μF ðvi Þ     I I 12e 1 12e HHY ðFÞ5
1 1 : μF ðvi Þ$ 2 μF ðvi Þ, 2 m 12e21=2 i51 ð15:6Þ Verma and Sharma [20] developed exponential entropy measure of order α; which has m h i X α 1 α  HVS ðFÞ5
ð120:5α Þ μF ðvi Þeð12μF ðvi ÞÞ 1ð12μF ðvi ÞÞeð12ð12μF ðvi ÞÞ Þ 21 ; 21 i51 m e α 6¼ 1ðα.0Þ: ð15:7Þ Mishra et al. [21] introduced the entropy for FSs is defined by H ðF Þ 5

m  X


 e 2 μF ðvi ÞeμF ðvi Þ 2 1 2 μF ðvi Þ eð12μF ðvi ÞÞ :

ð15:8Þ

i51

 pffiffiffi
pffiffiffi By Mishra et al. [21], we know that 0 # HðFÞ # m e e 2 1 . For the benefit, we can normalize it and define a entropy measure for FSs as m h i X 1  H ðF Þ 5 pffiffiffi
pffiffiffi e 2 μF ðvi ÞeμF ðvi Þ 2 ð1 2 μF ðvi ÞÞeð12μF ðvi ÞÞ : m e e 2 1 i51 ð15:9Þ For F, GAFSs(V), the simplest divergence measure for FSs as suggested by Bhandari and Pal [17] is given by  m  X  ð1 2 μF ðvi ÞÞ μ ðvi Þ
1 1 2 μF ðvi Þ log μF ðvi Þlog F CBP ðFOGÞ 5 : ð15:10Þ μG ðvi Þ ð1 2 μG ðvi ÞÞ i51 Fan and Xie [15] introduced an exponential fuzzy divergence measure which as m hn oi X
 CFX ðFOGÞ5 12 12μF ðvi Þ eðμF ðvi Þ2μG ðvi ÞÞ 1ð12μF ðvi ÞÞeðμG ðvi Þ2μF ðvi ÞÞ : i51

ð15:11Þ

An extended fuzzy divergence measure-based technique Chapter | 15

475

15.3 Divergence measures for fuzzy sets Corresponding to Eq. (15.7), the expected amount of information for discrimination in favour of F against G is given by  m  X α α  1
2 1 2 μF ðvi Þ eðð12μG ðvi ÞÞ 2ð12μF ðvi ÞÞ Þ : I 0 ðF; GÞ 5 ð15:12Þ 2 i51 Similarly amount of information for discrimination in favour of Fc against Gc as follows:  m  X 1 0 c c μαG ðvi Þ2μαF ðvi Þ 2 μF ðvi Þe I ðF ; G Þ 5 : ð15:13Þ 2 i51 Clearly I0 (F, G)6¼I0 (Fc, Gc), but discrimination between (F, G) and (Fc, G ) are same. Hence, the information for discrimination in favour of F against G is acquired as the addition of I0 (F, G) and I0 (Fc, Gc). Definition 15.3.1: Fuzzy information for discrimination of F against G is given by c

I ðF; GÞ 5 I 0 ðF; GÞ 1 I 0 ðF c ; Gc Þ; IðF; GÞ 5

ð15:14Þ

m h i X α α
 α α 1 2 μF ðvi ÞeμG ðvi Þ2μF ðvi Þ 2 1 2 μF ðvi Þ eðð12μG ðvi ÞÞ 2ð12μF ðvi ÞÞ Þ : i51

ð15:15Þ Theorem 15.3.1: If FAFS(V), then the relationship between Hα(F) and I (F, G) is presented by α


  eð120:5 Þ  I F; 12 : HVS ðF Þ512
ð120:5α Þ m e 21 Proof: 0 2 31 " m X ð120:5α Þ @ 415A ð120:5α Þ 5e e I F; 12μF ðvi Þe 2 i51 !α ! α 1 # 2ð12μF ðvi ÞÞ 2
 2 12μF ðvi Þ e 5

!α 1 2

!

2μαF ðvi Þ

m h i X α
 α α eð120:5 Þ 2μF ðvi Þeð12μF ðvi ÞÞ 2 12μF ðvi Þ eð12ð12μF ðvi ÞÞ Þ i51

52

m h X i51

52

m h X i51

i α
 α α μF ðvi Þeð12μF ðvi ÞÞ 1 12μF ðvi Þ eð12ð12μF ðvi ÞÞ Þ 21112eð120:5 Þ i α

 α α μF ðvi Þeð12μF ðvi ÞÞ 1 12μF ðvi Þ eð12ð12μF ðvi ÞÞ Þ 21 1m eð120:5 Þ 21



 α α 52m eð120:5 Þ 21 HVS ðFÞ1m eð120:5 Þ 21 :

476

PART | IV Sustainability

Thus α

HVS ðFÞ 5 1 2

eð120:5




α

Þ

m eð120:5 Þ 2 1


   I F; 12 :

Here, I(F, G) portrays fuzzy discrimination measure of F against G. Similarly m h i X α α
 α α I ðG; F Þ 5 1 2 μG ðvi ÞeμF ðvi Þ2μG ðvi Þ 2 1 2 μG ðvi Þ eðð12μF ðvi ÞÞ 2ð12μG ðvi ÞÞ Þ i51

is the average information for discrimination of G against F. As a result, divergence between F and G is given by Definition 15.3.2: For F, GAFSs(V), fuzzy divergence between F and G is given by

5

" m X i51

D1 ðF; GÞ 5 IðF; GÞ 1 IðG; FÞ α

α

α

α

2 2 μF ðvi ÞeμG ðvi Þ2μF ðvi Þ 2 μG ðvi ÞeμF ðvi Þ2μG ðvi Þ

α α
 2 1 2 μF ðvi Þ eðð12μG ðvi ÞÞ 2ð12μF ðvi ÞÞ Þ# α α
 2 1 2 μG ðvi Þ eðð12μF ðvi ÞÞ 2ð12μG ðvi ÞÞ Þ :

ð15:16Þ

For easement, we can normalize (15.16) and define a divergence measure for FSs as " m X 1 α α α α D1 ðF; GÞ 5 2 2 μF ðvi ÞeμG ðvi Þ2μF ðvi Þ 2 μG ðvi ÞeμF ðvi Þ2μG ðvi Þ 21 2mð1 2 e Þ i51 #
 ðð12μ ðv ÞÞα 2ð12μ ðv ÞÞα Þ
 ðð12μ ðv ÞÞα 2ð12μ ðv ÞÞα Þ G i F i F i G i 2 1 2 μ ðvi Þ e 2 1 2 μ ðvi Þ e : F

G

ð15:17Þ Taking into consideration that the elements in the universe of the members of discourse set may have distinct utilities or significance in the given scenario, thus needs to assign with diverse weights. Let w 5 (w1, w2,. . .,wm)T be a vector of elements viAV; i 5 1, 2,. . .,m. Then, the weighted divergence measure for FSs is expressed as " m X 1 α α α α wi 22μF ðvi ÞeμG ðvi Þ2μF ðvi Þ 2μG ðvi ÞeμF ðvi Þ2μG ðvi Þ Dw ðF;GÞ5 21 2ð12e Þ i51 #
 ðð12μ ðv ÞÞα2ð12μ ðv ÞÞα Þ
 ðð12μ ðv ÞÞα2ð12μ ðv ÞÞα Þ i i i i G F F G 2 12μ ðvi Þ e 2 12μ ðvi Þ e ; F

G

ð15:18Þ

An extended fuzzy divergence measure-based technique Chapter | 15

where wi $ 0 and

n P

wi 5 1. If w 5


1

 1 1 T m ; m ; . . .; m ;

i51

477

then measure Eq. (15.18)

reduces to the measure Eq. (15.17). Furthermore, from Eq. (15.6), we developed the cross entropy for FSs as follows:  m 



X 2ðμG ðvi Þ2μF ðvi ÞÞ 2ðμF ðvi Þ2μG ðvi ÞÞ     21 I μ ðv Þ$ 1 1 e 21 I μ ðv Þ, 1 : I2 ðF;GÞ5 e i

F

i51

F

2

i

2

ð15:19Þ Theorem 15.3.2: If FAFS(V), then the relation between HHY(F) and I2(F, G) is expressed as
  e21=2  I2 F; 12 : 21=2 m 12e


HHY ðF Þ 5 1 2 Proof: 21=2

e




I2 F;

1 2

21=2

5e

" m X

e2ð1=22μF ðvi ÞÞ 2 1 I

i51



2ðμF ðvi Þ21=2Þ

1 e

5

m  X

2ð12μF ðvi ÞÞ

e

21=2

2e

5

2ð12μF ðvi ÞÞ

e

i51

52

m  X

2 1 I

I

1 μF ðvi Þ , 2

21 I



 ;



2μF ðvi Þ  2 e21=2 I 1 1 e


2μF ðvi Þ 21I 1 1 e

μF ðvi Þ$ 2

12e2ð12μF ðvi ÞÞ I

  1 ;

μF ðvi Þ , 2



 21=2  : 1 1 12e

μF ðvi Þ, 2


2μ ðv Þ   F i 21 I 1 1 e

μF ðvi Þ$ 2

i51

1 μF ðvi Þ $ 2

μF ðvi Þ $ 2

i51 m  X

#



ð15:20Þ



21=2  1m 12e : 1

μF ðvi Þ, 2





5 2 m 1 2 e21=2 HHY ðFÞ 1 m 1 2 e21=2 : Hence, HHY ðF Þ 5 1 2


  e21=2  I2 F; 12 : m 1 2 e21=2


Again, I2 ðG;FÞ5

m 

X e2ðμF ðvi Þ2μG ðvi ÞÞ 21 Iμ i51

F



2ðμG ðvi Þ2μF ðvi ÞÞ  21 I μ 1 1 e ðv Þ$ i

2

 1



F ðvi Þ, 2

ð15:21Þ

478

PART | IV Sustainability

is an average information for discrimination of G against F. For that reason, divergence between F and G is given below: Definition 15.3.3: If F, GAFSs(V) then fuzzy divergence between F and G is expressed by "

D2 ðF; GÞ 5

D2 ðF; GÞ 5 I2 ðF; GÞ 1 I2 ðG; FÞ

m

X e2ðμG ðvi Þ2μF ðvi ÞÞ 1 e2ðμF ðvi Þ2μG ðvi ÞÞ 2 2 I



1 μF ðvi Þ $ 2

i51



1 e2ðμF ðvi Þ2μG ðvi ÞÞ 1 e2ðμG ðvi Þ2μF ðvi ÞÞ 2 2 I

#

ð15:22Þ

 :

1 μF ðvi Þ , 2

For easement, we can normalize Eq. (15.22) and define a divergence measure for FSs as " m

X 1   D2 ðF;GÞ5
21=2 e2ðμG ðvi Þ2μF ðvi ÞÞ 1e2ðμF ðvi Þ2μG ðvi ÞÞ 22 I 21 i51 1 2m e μ ðvi Þ$

1 e2ðμF ðvi Þ2μG ðvi ÞÞ 1e2ðμG ðvi Þ2μF ðvi ÞÞ 22 I

#

1 μF ðvi Þ, 2

F

2

 : ð15:23Þ

Theorem 15.3.3 (Properties of Proposed Divergence Measures for FSs): Let F, G, CAFSs(V) and α . 0(α6¼1) Then, the divergence measure Dk(F, G) (k 5 1, 2) given by Eqs. (15.17) and (15.23) satisfies the following postulates: (P1). Dk(F, G) 5 Dk(G, F) and 0 # Dk(F,G) # 1; (P2). Dk(F, G) 5 0 iff F 5 G; (P3). Dk(F, Fc) 5 1 iff FAP(V); (P4). Dk(F, G) 5 Dk(Fc, Gc); (P5). Dk(Fc, G) 5 Dk(F, Gc); (P6). Dk(F, G) # Dk(F, C) and Dk(G, C) # Dk(F,C) for FDGDC; (P7). Dk(F - G, F , G) 5 Dk(F, G); (P8). Dk(F , G, C) # Dk(F, C) 1 Dk(G, C);’CAFS(V); (P9). Dk(F - G, C) # Dk(F, C) 1 Dk(G, C);’CAFS(V). Proof: The proof of Theorem 15.3.3 is provided in Appendix. Theorem 15.3.4: The mapping Dk(F, G)(k 5 1, 2) is a distance measure on FS(V). Proof: From the above result, we know that Dk(F, G) satisfies (P1), (P2) and (P6). Therefore Dk(F, G), (k 5 1, 2) is a distance measure on FSs(V).

An extended fuzzy divergence measure-based technique Chapter | 15

479

15.3.1 An example for developed fuzzy divergence measures To tackle the uncertainties in the pattern recognition is a frequent problem. FS doctrine has provided various methods for pattern recognition problem. Here, we demonstrate the application of the developed divergence measures in the context of pattern recognition. To express the application of developed fuzzy divergence measures, we consider the problem of r known patterns M1, M2,. . .,Mr which have classifications P1, P2,. . .,Pr, respectively. These are characterized by FSs in V 5 {v1, v2,. . .,vs}, which as 
  Mi 5 vt ; μMt ðvt Þ :vt AV ; i 5 1ð1Þr; t 5 1ð1Þs: Given an unknown pattern N, represented by the fuzzy set as follows: 
  N 5 vt ; μN ðvt Þ :vt AV ; t 5 1ð1Þs: Next, this study aims at classifying N into one of the classes P1, P2,. . ., Pr. Based on the principle of minimum divergence measure between FSs, the process of assigning N to Pα is demonstrated by    ð15:24Þ α 5 argmin Dk ðMα ; N Þ ; where k 5 1; 2: Given algorithm is just a practical implementation of minimum divergence doctrine for FSs in pattern recognition. Example 3.1: Given that four known patterns M1, M2, M3 and M4 which have four classifications P1, P2, P3 and P4, respectively, are characterized by FSs in V.   M1 5 ðv1 ; 0:5Þ; ðv2 ; 0:6Þ; ðv3 ; 0:3Þ; ðv4 ; 0:2Þ ;   M2 5 ðv1 ; 0:7Þ; ðv2 ; 0:7Þ; ðv3 ; 0:7Þ; ðv4 ; 0:4Þ ;   M3 5 ðv1 ; 0:6Þ; ðv2 ; 0:5Þ; ðv3 ; 0:5Þ; ðv4 ; 0:6Þ ; and   M4 5 ðv1 ; 0:8Þ; ðv2 ; 0:6Þ; ðv3 ; 0:3Þ; ðv4 ; 0:2Þ : Given an unknown pattern N, represented by the fuzzy set as follows:   N 5 ðv1 ; 0:6Þ; ðv2 ; 0:4Þ; ðv3 ; 0:7Þ; ðv4 ; 0:5Þ : By using Eqs (15.23) and (15.24), we observe that N has correctly being classified to P3 is given by Table 15.1. Thus N has classification P3, since α 5 0.0191 is minimum.

TABLE 15.1 Divergence measure Dk(Mα, N) with α 5 {1, 2, 3, 4}.

N

M1

M2

M3

M4

0.0962

0.0352

0.0191

0.1058

480

PART | IV Sustainability

FIGURE 15.1 Implementation flowchart of fuzzy TOPSIS method.

15.4 Divergence measures-based fuzzy TOPSIS method In the following section, algorithm and implementation flowchart (Fig. 15.1) for fuzzy weighted divergence measures-based TOPSIS method is presented. Now, the procedure for proposed method is as follows: Step 1: Construction of decision matrix for FSs Let E 5 {E1, E2,. . ., Em} be a set of m alternatives and T 5 {T1, T2,. . ., Tn} be a set of n criteria. Now, fuzzy decision matrix Z 5 [zij]m 3 n of fuzzy value zij 5 (μij) is created, where zij presents the evaluation value of an alternative Ei over the criterion Qj. Step 2: Obtain information and normalized information matrix From Eq. (15.9), calculate the information of each fuzzy value in the fuzzy judgment
 matrix and get the information matrix as Z 5 (hij)m 3 n, where hi j 5 h1 r~i j . Then, normalize the information values in the mentioned decision matrix using h ij 5

hij ; maxhij

hij j 5 1ð1Þm; i 5 1ð1Þn

ð15:25Þ


 Thus the normalized information matrix is given as Z 5 h ij m 3 n : Step 3: Calculation of criteria’s weights T where ωi $ 0 and n Calculate the weight vector w 5 (ω1, ω2,. . .,ωn) P ωj 5 1; by applying the given formula j51

12 ωj 5

m2

n P

h ij

j51 m P n P i51 j51

; h ij

j 5 1ð1Þn:

ð15:26Þ

An extended fuzzy divergence measure-based technique Chapter | 15

481

Step 4: Determine the positive-ideal solution and negative-ideal solution The underlying logic of TOPSIS is defining the PIS and the NIS. PIS refers to a solution maximizing the benefit criteria and simultaneously, minimizing the cost criteria. On the other hand, NIS refers to a solution minimizing the benefit criteria whilst maximizing the cost criteria. The best alternative amongst all available ones is the alternative that shows the farthest distance from the NIS, whilst the shortest distance from the PIS. The estimation criterion can be characterized into two category, benefits and cost. Let B and F be the collection of benefits and cost criteria, respectively. According to fuzzy theory and the doctrine of the conventional TOPSIS technique, PIS E1 and NIS E2 can be expressed as below:  ( * + )   1 ð15:27Þ E 5 Qj ; maxμij jjAβ; min μij jAF :j 5 1ð1Þn ; i i  ( 2

E 5

 + )   Qj ; minμij jjAβ; max μij jAF :j 5 1ð1Þn ; i i  *

ð15:28Þ

for each i 5 1ð1Þm: Step 5: Determination of divergence measures from PIS and NIS With the use of Eq. (15.10), compute the divergence measure D(Ei, E1) amongst the options Ei(i 5 1(1)m) and the PIS E1 and the divergence measure D(Ei, E2) amongst the options Ei(i 5 1(1)m) and the NIS E2. Step 6: Calculate of relative closeness coefficient (CC) Eq. (15.29) is capable of calculating the relative CC of each option considering the fuzzy ideal solutions, as follows: Cc ðEi Þ 5

DðEi ; E2 Þ ; DðEi ; E2 Þ 1 DðEi ; E1 Þ

i 5 1ð1Þm:

ð15:29Þ

Step 7: Selection of optimal choice Select the highest value, which is signified by Cc(Ek) amongst the values Cc(Ei), i 5 1(1)m. As a result, Ek will be chosen as the best selection.

15.4.1 Case study of renewable energy investment The energy sector is of a great significance because of its important implications from economic, political and social perspective. The tremendous growth of world population and gross domestic product has resulted in a significant rise in the energy consumption rate across the world. In recent years, investment on RE has considerably increased. Although, traditional fossil fuels are still utilized as the main source for producing required energy. This chapter is mainly aimed to construct an effective framework

482

PART | IV Sustainability

FIGURE 15.2 Renewable energy investments criteria and alternatives.

TABLE 15.2 Fuzzy decision matrix for renewable energy investments alternative. Q1

Q2

Q3

Q4

Q5

E1

0.2846

0.75

0.234

0.23

0.24

E2

0.318

0.248

0.766

0.097

0.76

E3

0.759

0.75

0.315

0.32

0.278

E4

0.241

0.437

0.41

0.155

0.394

that gives helpful supports to decision-making processes specifically regarding the RE investments. To do this, literature was extensively reviewed and the energy sector practices were investigated in a way to identify the most important criteria that have significant effects on the decisions made by investors. The pertinent criteria are classified into five classes: environmental, technical, political, economic and social. The name of criteria and alternatives are presented in Fig. 15.2. Step 1: The fuzzy decision matrix is constructed and given in Table 15.2. Step 2: By using the entropy formula Eq. (15.4), we can compute the information measure of each fuzzy value in the above judgment matrix and get the following information matrix:

An extended fuzzy divergence measure-based technique Chapter | 15

2

0:8200 6 0:8717
 D 5 hij n 3 m 5 6 4 0:7389 0:7389

0:7569 0:7529 0:7569 0:9847

0:7244 0:7244 0:8674 0:9687

with formula Eq. (15.17), we transform the above normalized information matrix below: 2 1:0000 0:9230 0:8834 6 1:0000 0:8637 0:8310
 D 5 h ij n 3 m 5 6 4 0:8449 0:8655 0:9919 0:7504 1:0000 0:9838

0:7160 0:3588 0:8745 0:5331

483

3 0:7368 0:7368 7 7 0:8087 5 0:9566

information matrix to the 0:8732 0:4116 1:0000 0:5414

3 0:8985 0:8452 7 7: 0:9248 5 0:9715

Step 3: Utilize the normalized information matrix and by the formula Eq. (15.18), the weight vector of all the decision attributes are obtained as W 5 ð0:2092; 0:2138; 0:2169; 0:1473; 0:2128ÞT : Step 4: Fuzzy PIS and NIS are calculated by using Eqs. (15.19) and (15.20), given as follows: E1 5 f0:759; 0:75; 0:766; 0:32; 0:76g; E2 5 f0:241; 0:248; 0:234; 0:097; 0:24g: Step 5: Fuzzy divergence measure of each alternative from the PIS and NIS are calculated by using Eq. (15.10) and results are shown in Table 15.3 and Table 15.4. Step 6: Eq. (15.21) was used to calculate the relative CCs and the results are presented in Table 15.5. The TOPSIS and fuzzy TOPSIS methods are adopted to determine the ranking of the four above-mentioned alternatives and are demonstrated in Table 15.6. In the present research, assessing the RE investment is taken into account as an MCDM problem involving a number of conflicting criteria. Generally

TABLE 15.3 Divergence measure of each alternative from the PIS. α 5 0.2 1

α 5 0.7

α 5 1.2

D(E1, E ) 5 0.0811

D(E1, E ) 5 0.1366

D(E1, E1) 5 0.1416

D(E2, E1) 5 0.0054

D(E2, E1) 5 0.0078

D(E2, E1) 5 0.0056

1

1

D(E3, E ) 5 0.0450

D(E3, E ) 5 0.0759

D(E3, E1) 5 0.0783

D(E4, E1) 5 0.0333

D(E4, E1) 5 0.0553

D(E4, E1) 5 0.0541

PIS, Positive-ideal solution.

1

484

PART | IV Sustainability

TABLE 15.4 Divergence measure of each alternative from the NIS. α 5 0.2

α 5 0.7

2

α 5 1.2

2

D(E1, E ) 5 0.0237

D(E1, E ) 5 0.0393

D(E1, E2) 5 0.0395

D(E2, E2) 5 0.584

D(E2, E2) 5 0.0982

D(E2, E2) 5 0.1020

D(E3, E2) 5 0.0058

D(E3, E2) 5 0.0084

D(E3, E2) 5 0.0061

D(E4, E2) 5 0.0105

D(E4, E2) 5 0.0169

D(E4, E2) 5 0.0151

NIC, Negative-ideal solution.

TABLE 15.5 Relative closeness coefficients for α 5 0.2, 0.7 and 1.2. α 5 0.2

α 5 0.7

α 5 1.2

E1

0.2261

0.2234

0.2180

E2

0.9908

0.9264

0.9480

E3

0.1142

0.0996

0.0723

E4

0.2397

0.2341

0.2183

Rankings

E3 gE1 gE4 gE2

E3 gE1 gE4 gE2

E3 gE1 gE4 gE2

TABLE 15.6 Comparison of ranking order with different approaches. MA

Rankings

Optimal alternative

TOPSIS by Hooda et al. [25]

E3 gE1 gE4 gE2

E3

Fuzzy TOPSIS by Hooda and Mishra [27]

E3 gE1 gE4 gE2

E3

Proposed divergence measure-based TOPSIS method

E3 gE1 gE4 gE2

E3

the decision matrices contain vagueness and uncertainty, which are dealt with using FSs. Here, for the objective of comparison, we compute fuzzy values of fuzzy decision matrix and fuzzy PIS and fuzzy NIS. Afterwards, performances of RE investment with respect to each criterion are computed. Findings indicated that any conflict did not exist in the predilection

An extended fuzzy divergence measure-based technique Chapter | 15

485

authoritatively mandating of all the options by TOPSIS method, fuzzy TOPSIS method and developed method. From Table 15.6, the wind energy (E3) shows a comparatively better performance compared to the other threementioned methods concerning most of the criteria considered here, and this energy is closer to fuzzy PIS. Also, the ranking result acquired by the fuzzy TOPSIS approach is compared with some other TOPSIS methods within different uncertain contexts in Table 15.5 and thus we obtain that the option wind energy (E3) is the desirable RE investment alternative. Not only investors, but also public organizations need to be attentive about the pressing need of investing on RE and also the need for increasing their support on production and consumption of these forms of energy. Regulations and policies announced by governments are key criteria that are related to the decisions made by public organizations in the energy sector. As a result, public parties must reform their strategies and investment plans to meet the requirements of investors in a way to grow the share of RE in the total amount of energy generated in the country. If governmental institutes and authorities are capable of establishing effective motivating mechanisms and maintaining them different conditions like political/economic fluctuations and crises, then both foreign and local investors can grow their investment in RE. In future, we will focus our studies on regions with more cultural diversity with a new MCDM technique and compare with the results obtained by the present study as the experiences and preferences may be varying by tradition, country or socio-economic levels.

15.5 Conclusions The present study first made a comprehensive review of currently used divergence measures of FSs and proposed two new pairs of divergence measures for FSs with their elegant properties. Based on proposed divergence measure, the classical TOPSIS method was extended under fuzzy environment. In this method, the method based on entropy/cross entropy was used aiming for determining the criteria weights. The criterion with a small entropy and large cross entropy need to be allocated with a large weight. The approach proposed here is on the basis of the relative closeness of each alternative for determining the ranking order of all alternatives, which avoids the production of loss of too much information during the aggregation of information. The proposed method was examined in terms of applicability and validity with a decision-making problem of RE investment. Findings showed that the developed approach has an acceptable level of straightforwardness with less information loss, and it was found applicable to other decision-making problems under FSs. In future studies, we have the plan to extend the divergence measure proposed in this study in way to be effectively applied to Pythagorean FSs, hesitant FSs, interval-valued Pythagorean FSs and

486

PART | IV Sustainability

Pythagorean fuzzy linguistic term sets and we will attempt to develop algorithms for some new MCDM methods under these environments.

Appendix: Proof of the properties Properties (P1)
(P5) are proved from the definition. Therefore, we omit the proof. (P6). Let FDGDC, then μF(vi) # μG(vi) P # μC
(vi), ’viAV. Let ξi 5 μF(vi) 2 μG(vi) then Dk ðF; GÞ 5 i f ξi ; k 5 1; 2: Now, 0 $ ξi1 5 μF(vi) 2 μG(vi) $ ξi2 5 μF(vi) 2 μC(vi) $ 2 1 f(ξi1) # f 2 (ξi ), thus X
 X
 f ξ1i # f ξ 2i 5 Dk ðF; CÞ; k 5 1; 2: Dk ðF; GÞ 5 i

i

Similarly Dk(F, G) # Dk(F, C), k 5 1, 2. Divide V into two parts V1 and V2 such that V1 , V2 5 V, where V1 5 {viAV:F(v i)DG(vi)} and V2 5 {v i)}. It implies that   iAV:G(vi)DF(v   V1 5 vi :μF ðvi Þ $ μG ðvi Þ; ’vi AV and V2 5 vi :μF ðvi Þ , μG ðvi Þ; ’vi AV . Now, we elaborate notions as follows: In set V1,   F , G 5 Union of F and G3μF , G ðvi Þ 5 max μF ðvi Þ;μG ðvi Þ 5 μF ðvi Þ F - G 5 Intersection of F and G3μF - G ðvi Þ 5 min μF ðvi Þ; μG ðvi Þ 5 μG ðvi Þ. In set V2,   F , G 5 Union of F and G3μF , G ðvi Þ 5 max μF ðvi Þ;μG ðvi Þ 5 μG ðvi Þ; F - G 5 Intersection of F and G3μF - G ðvi Þ 5 min μF ðvi Þ; μG ðvi Þ 5 μF ðvi Þ. (P7). D1 ðF - G; F , GÞ " m X 1 α α α α 5 2 2 μF - G ðvi ÞeμF , G ðvi Þ2μF - G ðvi Þ 2 μF , G ðvi ÞeμF - G ðvi Þ2μF , G ðvi Þ 2mð1 2 e21 Þ i51 α α
 2 1 2 μF - G ðvi Þ eðð12μF , G ðvi ÞÞ 2ð12μF - G ðvi ÞÞ Þ #
 ðð12μ ðv ÞÞα 2ð12μ ðv ÞÞα Þ i i F G F , G 2 12μ ðvi Þ e : F,G

" X 1 α α α α 5 2 2 μG ðvi ÞeμF ðvi Þ2μG ðvi Þ 2 μF ðvi ÞeμG ðvi Þ2μF ðvi Þ 21 2mð1 2 e Þ vi A V1


α α α α 
 2 1 2 μG ðvi Þ eðð12μF ðvi ÞÞ 2ð12μG ðvi ÞÞ Þ 2 1 2 μF ðvi Þ eðð12μG ðvi ÞÞ 2ð12μF ðvi ÞÞ Þ

#

An extended fuzzy divergence measure-based technique Chapter | 15

487

" X 1 α α α α 1 2 2 μF ðvi ÞeμG ðvi Þ2μF ðvi Þ 2 μG ðvi ÞeμF ðvi Þ2μG ðvi Þ 21 2mð1 2 e Þ vi A V2

#  ðð12μ ðv ÞÞα 2ð12μ ðv ÞÞα Þ
 ðð12μ ðv ÞÞα 2ð12μ ðv ÞÞα Þ G i F i F i G i 2 1 2 μG ðvi Þ e : 2 1 2 μF ðvi Þ e


" m X 1 α α α α 5 2 2 μF ðvi ÞeμG ðvi Þ2μF ðvi Þ 2 μG ðvi ÞeμF ðvi Þ2μG ðvi Þ 2mð1 2 e21 Þ i51

#  ðð12μ ðv ÞÞα 2ð12μ ðv ÞÞα Þ
 ðð12μ ðv ÞÞα 2ð12μ ðv ÞÞα Þ i i i i G F F G 2 1 2 μG ðvi Þ e 2 1 2 μF ðvi Þ e :


5 D1 ðF; GÞ Thus D1(F - G, F , G) 5 D1(F, G). Similarly D2(F - G, F , G) 5 D2(F, G). Hence, Dk(F - G, F , G) 5 Dk(F, G),k 5 1, 2 (P8). D1 ðF; CÞ 1 D1 ðG; CÞ 2 D1 ðF , G; CÞ 5

m  X 1 α α α α 2 2 μF ðvi ÞeμC ðvi Þ2μF ðvi Þ 2 μC ðvi ÞeμF ðvi Þ2μC ðvi Þ 2mð1 2 e21 Þ i51

α α α α

 2 1 2 μF ðvi Þ eðð12μC ðvi ÞÞ 2ð12μF ðvi ÞÞ Þ 2 1 2 μC ðvi Þ eðð12μF ðvi ÞÞ 2ð12μC ðvi ÞÞ Þ 

1

m  X 1 α α α α 2 2 μG ðvi ÞeμC ðvi Þ2μG ðvi Þ 2 μC ðvi ÞeμG ðvi Þ2μC ðvi Þ 2mð1 2 e21 Þ i51

α α α α

 2 1 2 μG ðvi Þ eðð12μC ðvi ÞÞ 2ð12μG ðvi ÞÞ Þ 2 1 2 μC ðvi Þ eðð12μG ðvi ÞÞ 2ð12μC ðvi ÞÞ Þ 

2

m  X 1 α α α α 2 2 μF , G ðvi ÞeμC ðvi Þ2μF , G ðvi Þ 2 μC ðvi ÞeμF , G ðvi Þ2μC ðvi Þ 21 2mð1 2 e Þ i51

α α α α

 2 12μF ,G ðvi Þ eðð12μC ðvi ÞÞ 2ð12μF ,G ðvi ÞÞ Þ 2 12μC ðvi Þ eðð12μF ,G ðvi ÞÞ 2ð12μC ðvi ÞÞ Þ :

5

X 1 α α α α 22μF ðvi ÞeμC ðvi Þ2μF ðvi Þ 2μC ðvi ÞeμF ðvi Þ2μC ðvi Þ 2mð12e21 Þ vi A V2

α α α α

 2 12μF ðvi Þ eðð12μC ðvi ÞÞ 2ð12μF ðvi ÞÞ Þ 2 12μC ðvi Þ eðð12μF ðvi ÞÞ 2ð12μC ðvi ÞÞ Þ 

1

X 1 α α α α 22μG ðvi ÞeμC ðvi Þ2μG ðvi Þ 2μC ðvi ÞeμG ðvi Þ2μC ðvi Þ 2mð12e21 Þ vi A V1

488

PART | IV Sustainability

α α α α

 2 12μG ðvi Þ eðð12μC ðvi ÞÞ 2ð12μG ðvi ÞÞ Þ 2 12μC ðvi Þ eðð12μG ðvi ÞÞ 2ð12μC ðvi ÞÞ Þ 

$0: Thus D1(F, C) 1 D1(G, C) $ D1(F , G, C) Similarly D2(F, C) 1 D2(G, C) $ D2(F , G, C). Hence, Dk(F, C) 1 Dk(G, C) $ Dk(F , G, C),k 5 1, 2. (P9). The proof is on similar lines as in (P8).

References [1] Wang JJ, Jing YY, Zhang CF, Zhao JH. Review on multi-criteria decision analysis aid in sustainable energy decision-making. Renew Sustain Energy Rev 2009;13(9):2263
78. [2] Ozorhon B, Batmaz A, Caglayan S. Generating a framework to facilitate decision making in renewable energy investments. Renew Sustain Energy Rev 2018;95:217
26. [3] Strantzali E, Aravossis K. Decision making in renewable energy investments: A review. Renew Sustain Energy Rev 2016;55:885
98. [4] Cavallaro F, Zavadskas EK, Streimikiene D, Mardani A. Assessment of concentrated solar power (CSP) technologies based on a modified intuitionistic fuzzy topsis and trigonometric entropy weights. Technol Forecast Soc Change 2019;140:258
70. [5] Cavallaro F, editor. Assessment and simulation tools for sustainable energy systems. Theory and applications. Springer Part of the Green Energy and Technology book series; 2013. [6] Tsoutsos T, Drandaki M, Frantzeskaki N, Iosifidis E, Kiosses I. Sustainable energy planning by using multi-criteria analysis application in the island of Crete. Energy Policy 2009;37(5):1587
600. [7] Haralambopoulos DA, Polatidis H. Renewable energy projects: structuring a multi-criteria group decision-making framework. Renew Energy 2003;28(6):961
73. [8] Mardani A, Jusoh A, Zavadskas EK, Cavallaro F, Khalifah Z. Sustainable and renewable energy: an overview of the application of multiple criteria decision making techniques and approaches. Sustainability 2015;7(10):13947
84. [9] Cavallaro F. Multi-criteria decision aid to assess concentrated solar thermal technologies. Renew Energy 2009;34(7):1678
85. [10] Mateo JRSC. The renewable energy industry and the need for a multi-criteria analysis. Multi criteria analysis in the renewable energy industry. London, UK: Springer; 2012. p. 1
5. [11] Zadeh LA. Fuzzy sets. Inf Control 1965;08:338
53. [12] Montes I, Janis V, Montes S. An axiomatic definition of divergence for intuitionistic fuzzy sets . ISBN 978-90- 78677-00-0 Advances in intelligent systems research, EUSFLAT 2011. Aix-Les Bains: Atlantis Press; 2011. p. 547
53. [13] Montes I, Pal NR, Janis V, Montes S. Divergence measures for intuitionistic fuzzy sets. IEEE Trans Fuzzy Syst 2015;23:444
56. [14] Montes S, Couso I, Gil P, Bertoluzza C. Divergence measure between fuzzy sets. Int J Approx Reason 2002;30(2):91
105. [15] Fan J, Xie W. Distance measure and induced fuzzy entropy. Fuzzy Sets Syst 1999;104:305
14.

An extended fuzzy divergence measure-based technique Chapter | 15

489

[16] Shang XG, Jiang WS. A note on fuzzy information measures. Pattern Recognit Lett 1997;18(5):425
32. [17] Bhandari D, Pal NR. Some new information measures of fuzzy sets. Inf Sci 1993;67:204
28. [18] Pal NR, Pal SK. Object-background segmentation using new definitions of entropy. IEE Proc 1989;136(4):284
95. [19] Hwang CH, Yang MS. On entropy of fuzzy sets. Int J Uncertain Fuzz 2008;16 (4):519
27. [20] Verma RK, Sharma BD. On generalized exponential fuzzy entropy, World Academy of Science. Eng Technol 2011;5:886
9. [21] Mishra AR, Hooda DS, Jain D. On exponential fuzzy measures of information and discrimination. Int J Comput Appl 2015;119:01
7. [22] Mishra AR, Jain D, Hooda DS. On fuzzy distance and induced fuzzy information measures. J Inf Optim Sci 2016;37:193
211. [23] Chaira T, Ray AK. Segmentation using fuzzy divergence. Pattern Recognit Lett 2003;24:1837
44. [24] Mishra AR, Jain D, Hooda DS. On logarithmic fuzzy measures of information and discrimination. J Inf Optim Sci 2016;37:213
31. [25] Hooda DS, Mishra AR, Jain D. On generalized fuzzy mean code word lengths. Am J Appl Math 2014;02:127
34. [26] Mishra AR, Hooda DS, Jain D. Weighted trigonometric and hyperbolic fuzzy information measures and their applications in optimization principles. Int J Comput Math Sci 2014;03:62
8. [27] Hooda DS, Mishra AR. On trigonometric fuzzy information measures. ARPN J Sci Technol 2015;05:145
52. [28] Ansari MD, Ghrera SP, Mishra AR. Texture feature extraction using intuitionistic fuzzy local binary pattern. J Intell Syst 2016. Available from: https://doi.org/10.1515/jisys-20160155. [29] Fan S, Yang S, He P, Nie H. Infrared electric image thresholding using two dimensional fuzzy entropy. Energy Procedia 2011;12:411
19. [30] Bhatia PK, Singh S. A new measure of fuzzy directed divergence and its application in image segmentation. J Intell Syst Appl 2013;04:81
9. [31] Mishra AR, Rani P. Shapley divergence measures with VIKOR method for multi-attribute decision-making problems, Neural Computing and Applications 2019;31:1299
1316. [32] Poletti E, Zappelli F, Ruggeri A, Grisan E. A review of thresholding strategies applied to human chromosome segmentation. Comput Methods Prog Biomed 2012;108:679
88. [33] Ghosh M, Das D, Chakraborty C, Ray AK. Automated leukocyte recognition using fuzzy divergence. Micron 2010;41:840
6. [34] Parkash O, Sharma PK, Kumar S. Two new measures of fuzzy divergence and their properties. SQU J Sci 2006;11:69
77. [35] Ferreri C. Hyperentropy and related heterogeneity divergence and information measures. Statistica 1980;40(2):155
68. [36] Tomar VP, Ohlan A. Sequence of fuzzy divergence measures and inequalities. AMO Adv Model Optim 2014;16(2):439
52. [37] Mishra AR. Intuitionistic fuzzy information with application in rating of township development. Iran J Fuzzy Syst 2016;13:49
70.

490

PART | IV Sustainability

[38] Mishra AR, Rani P, Jain D. Information measures based TOPSIS method for multicriteria decision making problem in intuitionistic fuzzy environment. Iran J Fuzzy Syst 2017;14:41
63. [39] Mishra AR, Jain D, Hooda DS. Exponential intuitionistic fuzzy information measure with assessment of service quality. Int J Fuzzy Syst 2017;19:788
98. [40] Mishra AR, Jain D, Hooda DS. Intuitionistic fuzzy similarity and information measures with physical education teaching quality assessment. In: Proceedings of the Second IC3T, Advances in Intelligent Systems and Computing; 2016c; 379. pp. 387
399. [41] Mishra AR, Kumari R, Sharma DK. Intuitionistic fuzzy divergence measure-based multicriteria decision-making method. Neural Computing and Applications 2019;31:2279
94. [42] Das S, Guha D. A centroid-based ranking method of trapezoidal intuitionistic fuzzy numbers and its application to MCDM problems. Fuzzy Inf Eng 2016;08:41
74. [43] Rani P, Mishra AR, Rezaei G, Liao H, Mardani A. Extended Pythagorean fuzzy TOPSIS method based on similarity measure for sustainable recycling partner selection. Int J Fuzzy Syst 2020;22(2):735
47. Available from: https://doi.org/10.1007/s40815-01900689-9. [44] Shen F, Ma X, Li Z, Xu Z, Cai D. An extended intuitionistic fuzzy TOPSIS method based on a new distance measure with an application to credit risk evaluation. Inf Sci 2018;428:105
19. [45] Yu C, Shao Y, Wang K, Zhang L. A group decision making 828 sustainable supplier selection approach using extended TOPSIS under interval-valued Pythagorean fuzzy environment. Expert Syst Appl 2019;121:1
17. Available from: https://doi.org/10.1016/j. eswa.2018.12.010. [46] Rani P, Mishra AR, Mardani A, Cavallaro F, Alrasheedi M, Alrashidi A. A novel approach to extended fuzzy TOPSIS based on new divergence measures for renewable energy sources selection. Journal of Cleaner Production, In Press, 2020, https://doi.org/ 10.1016/j.jclepro.2020.120352. [47] Shannon CE. A mathematical theory of communication. Bell Syst Tech J 1948;27 (3):379
423. [48] Kullback S, Leibler RA. On information and suffciency. Ann Math Stat 1951;22:79
86. [49] Liu XC. Entropy, distance measure and similarity measure of fuzzy sets and their relations. Fuzzy Sets Syst 1992;52:305
18.

Chapter 16

Multicriteria decision making for the selection of the best renewable energy scenario based on fuzzy inference system Jingzheng Ren, Yi Man, Ruojue Lin and Yue Liu Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, P.R. China

Chapter Outline 16.1 Introduction 16.2 Method 16.3 Application

491 493 496

16.4 Conclusions References

504 505

16.1 Introduction The depletion of fossil fuels and the current serious environmental contaminations drive the researchers and industries to explore more sustainable energy sources/systems to meet the needs for promoting the development of economy and reducing the environmental loads simultaneously. Renewable energy with a wide range of sources has gradually attracted more and more attention to the governments of many countries [1]. There are various types of renewable energy, such as wind power, solar energy, wave energy and biomass [2]. These renewable energy sources/systems usually have different merits and shortcomings in different aspects. For instance, solar energy is easily accessible, but it can only be collected during the daytime. Wind power can be collected the whole day, but it is usually limited by the local climate conditions. In other words, since a renewable energy source/system has both advantages and weaknesses, it is difficult for the decision-makers to determine which one is the best, the most suitable or the most sustainable when there are several different alternative energy sources/systems. Therefore, it is necessary and essential to evaluate different renewable Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00016-9 © 2021 Elsevier Inc. All rights reserved.

491

492

PART | IV Sustainability

energy alternatives and conduct optimal planning of renewable energy systems to simultaneously achieve the harmonious development of economy, environment and society. Decision-making requires useful guidelines, and these guidelines can evaluate the alternatives and steer the choice of the decision-makers towards the most suitable or sustainable option [3]. There is plenty of work focusing on the methods for the assessment and decision-making on renewable energy system selection. And it is also apparent that decision-making on renewable energy selection is a multicriteria decision analysis [also called ‘multicriteria decision making’ (MCDM)] problem
multiple alternatives with various characteristics need to be evaluated [4
9]. A life cycle environmental assessment and life cycle cost analysis was conducted to evaluate new renewable energy systems and select the optimum one [10]. Akella et al. [5] analyzed the social, economic and environmental performances of renewable energy systems. Therefore, multidimensional indicators (i.e. economic, environmental, social, and technical) should be incorporated in the decisionmaking process for selecting the best renewable energy system among multiple alternatives. Accordingly, MCDM methods have been widely used for renewable energy selection. For instance, Lee et al. [6] established a new MCDM method based on analytic hierarchy process (AHP) associated with benefits, opportunities, costs and risks for wind farm selection. Similarly, AHP method was applied to choose the most convenient location for a wind observation station [8]. Life cycle assessment combining with AHP method was employed to evaluate the sustainability performances of the distributed renewable energy alternatives and find the optimum one among multiple alternatives [11]. An AHP
based MCDM method was applied to rank four renewable energy resources in Malaysia by considering the indicators in technical, economic, social and environmental aspects [12]. The VIKOR method combining with AHP method was employed to select the most suitable renewable energy project based on the governmental planning [7]. A decision supporting framework based on ELimination Et Choix Traduisant la REalite´ III was proposed and applied for the selection of the best alternative among the multisource energy systems [9]. Preference Ranking Organization METHod for Enrichment of Evaluations II outranking approach was used to establish a group decision-making framework for the selection of the best renewable energy project in the context of a Greece city [4]. According to the abovementioned literature review, AHP and its improved methods by combining with some other methods are the most frequently used in the decision-making on the selection of renewable energy system. More related studies can be found in the selection of renewable energy investment choices [13] and the selection of the most sustainable energy system [14]. In the decision-making practice, uncertainties caused by incomplete and vague information generated from the subjective expressions and linguistic

Multicriteria decision making Chapter | 16

493

descriptions are very common. Some of the traditional MCDM methods do not have the ability to deal with the uncertain situations. Therefore, the methods for the evaluation of renewable energy systems and achieving decisionmaking under uncertain conditions among a set of alternatives are urgently needed. Currently some studies employed fuzzy set theory in the decisionmaking process. Kahraman et al. [15] combined fuzzy set theory with axiomatic design and AHP method to handle the vague information in the selection of the best renewable energy option [15]. A fuzzy MCDM method was established with the considerations of life cycle impacts and uncertainties to prioritize the renewable energy technologies [16]. Some researchers also combined stochastic models with MCDM models for sustainable energy system assessment and selection under uncertainties [17]. A modified fuzzy TOPSIS and fuzzy AHP were combined to establish an MCDM framework for energy planning which allowed the users to use linguistic terms to express their opinions [18]. Therefore, many studies employed the fuzzy set theory
based MCDM to handle the uncertainties in the decision-making process for renewable energy system selection. Besides the classical fuzzy set theory
based MCDM methods, fuzzy inference system which is based on the fuzzy logic principles can also be used for MCDM problems [19,20]. It has been widely used in various fields, such as water quality evaluation [21], risk assessment [22,23], supplier selection [24] and ecodesign [25]. Fuzzy inference system was also combined with neural networks for estimation and forecasting [26,27]. Sustainability assessment can also be conducted by applying fuzzy inference system which can simplify the procedures and improve the precision [28]. Although fuzzy inference system can be applied in sustainability evaluation towards green products, there are a limited number of studies using fuzzy inference system for the evaluation of renewable energy systems and the decision-making on renewable energy system selection. To the best of our knowledge, there are quite a few studies that use fuzzy inference system for the evaluation of renewable energy. To fill this research gap, fuzzy inference system was presented and introduced to evaluate renewable energy alternatives in this study. Besides this section, the fuzzy inference system was introduced in Section16.2. In Section 16.3, an illustrative case with 13 renewable energy alternatives was studied by using fuzzy inference system. In the final section, this chapter was finally concluded.

16.2 Method Fuzzy logic inference has been used to solve the linguistic uncertainties in decision-making and has been widely used in energy evaluation and control [29
31]. In this section, fuzzy inference system was presented and introduced to evaluate renewable energy alternatives.

494

PART | IV Sustainability

FIGURE 16.1 Fuzzy inference system.

The fuzzy inference system is a process of mapping from a given input set to an output set using the fuzzy logics. The fuzzy inference system contains three major steps, including fuzzification, processing operation rules and defuzzification [23] as shown in Fig. 16.1. These three steps of fuzzy inference system are presented as follows [22,23,32,33]: Step 1. Fuzzification In this step, fuzzy sets are used as the media for mapping quantitative scope and linguistic judgement. Fuzzy set theory is a mathematical tool to address the imprecision and uncertainties existing in human judgments through the use of linguistic terms and degrees of membership [24]. Membership function can be used to calculate the membership. Triangular and trapezoidal functions are usually used in fuzzy logic, because they are simple and can reduce the execution time [34]. The triangular and trapezoidal membership functions are shown in Eqs (16.1) and (16.2), respectively [35]. 8 0 x#a > > x2a > > a , x#b > >

b,x#c > > >b2c > > : 0 x.c 8 0 x # a or x $ d > > x2a > > > a,x#b >

> > x 2 d > > x.c > :c2d

Multicriteria decision making Chapter | 16

495

where a, b, c and d are membership function parameters;μA~ ðxÞ represents the ~ membership of x belongs to fuzzy setA. Step 2. Determination of operation rules In this step, IF-THEN rules are determined to execute the logic inference. IF-THEN rule consists of if part, which is called the antecedent, and then part, which is called the consequent [36]. Fuzzy logic operations are used in the fuzzy inference system, for instance: IF x satisfies A AND y satisfies B, THEN C. IF x satisfies A OR y satisfies B, THEN C. To quantify the process, the standard fuzzy set operations are usually used. The standard fuzzy set operations consist of union (OR), intersection (AND) and additive complement (NOT), and these three fuzzy set operations have been shown in Eqs (16.3)
(16.5), respectively. AND: μA~ - B~ ðxÞ 5 min ðμA~ ðxÞ; μB~ ðxÞÞ

ð16:3Þ

OR: μA~ , B~ ðxÞ 5 max ðμA~ ðxÞ; μB~ ðxÞÞ

ð16:4Þ

NOT: μA ðxÞ 5 1 2 μA~ ðxÞ

ð16:5Þ

~ where μA~ ðxÞ represents the membership of x belongs to fuzzy set A. Step 3. Defuzzification Defuzzification is to translate the fuzzy inference solution to a crisp number or a real world solution. There are many defuzzification methods including centre of area, centre of sums, centre of largest area, first of maxima, middle of maxima, max criterion and height defuzzification [34]. Centre of area method of defuzzification [37] has been used widely for determining the output, and it was used for defuzzification in this chapter, as shown in Eq. (16.6). Ð μ ~ ðxÞxdx x 5 Ð A ð16:6Þ μA~ ðxÞdx where x represents the result of fuzzy inference system and μA~ ðxÞ represents ~ the membership of x belongs to fuzzy set A. By conducting the abovementioned three steps, the output of a fuzzy inference system can be generated. In this study, a fuzzy inference model containing three fuzzy inference systems is developed, as shown in Fig. 16.2. The attributes for the selection of renewable energy systems can be divided into benefit-type criteria and cost-type criteria. The benefit-type criterion has positive influences on the performance of renewable energy systems when the value concerning the criterion increased. Similarly, if a criterion is the cost-type, the performance of the alternative becomes worse when the value concerning the criterion increases. Therefore, the fuzzy logic for benefit-type criteria and that for the cost-type criteria are different. In this case, the benefit-type criteria and the cost-type criteria are used to design

496

PART | IV Sustainability

FIGURE 16.2 Fuzzy inference system for the evaluation of the renewable energy alternatives.

two separate inference systems, and they are the benefit-type criteria inference system and cost-type criteria inference system, respectively (Fig. 16.2). The results determined by benefit-type criteria inference system and that determined by the cost-type criteria inference system can be integrated by using the integrated performance inference system. The alternative renewable energy systems will then be ranked according to the outputs of integrated performance inference system. In this study, the Mamdani inference system was selected and the Fuzzy Logic Toolbox of Matlab was used to evaluate the alternatives [38].

16.3 Application Fuzzy inference system has been used for the evaluation of 13 alternatives for electricity generation based on the work of San Cristo´bal [39], and the details of each alternative are presented in Table 16.1. To analyze the renewable energy options comprehensively, multiple attributes were selected for the assessment, including power (P), operating hours (OH), useful life (UL), tons of CO2 avoided (tCO2A) investment ration (IR), implementation period (IP) and operating and maintenance costs (O&MC) [39]. In this study, the benefit-type criteria consist of P, OH, UL and tCO2A, and the cost-type criteria consist of IR, implementation period (IP) and operating and

Multicriteria decision making Chapter | 16

497

TABLE 16.1 Alternatives for electric generation [39]. Alternative A1

Wind power P # 5 MW

A2

Wind power 5 MW # P # 10 MW

A3

Wind power 10 MW # P # 50 MW

A4

Hydroelectric P # 10 MW

A5

Hydroelectric 10 MW # P # 25 MW

A6

Hydroelectric 25 MW # P # 50 MW

A7

Solar thermoelectric P $ 10 MW

A8

Biomass (energetic cultivations) P # 5 MW

A9

Biomass (forest and agricultural waste) P # 5 MW

A10

Biomass (farming industrial wastes) P # 5 MW

A11

Biomass (forest industrial wastes) P # 5 MW

A12

Biomass (con-combustion in conventional central) P $ 50 MW

A13

Biofuels P # 2 MW

maintenance costs (O&MC). The alternatives for electricity generation are evaluated based on the attributes shown in Table 16.2. The evaluation of the input variables including P, PH, UL, tCO2A, IR, IP and O&MC and the output variables including benefit-type criteria, cost-type criteria and integrated performance has been divided into three types: ‘low’, ‘medium’ and ‘high’. To indicate the fuzzy boundary of the criteria values, the trapezoidal function has been used to define the membership functions of the variables. For each criterion, the parameters for membership function are determined by experts to map the linguistic judgements to the fuzzy numbers, and the parameters of membership functions used in this study are shown in Table 16.3. Taking the input variable (criterion) P as an example, the membership function of input variable P is shown in Eqs (16.7)
(16.9) and Fig. 16.3.  1 0 # x # 1:2 ð16:7Þ μLow ðxÞ 5 2 2 0:83x 1:2 , x # 2:4 8 0 x # 1:8 or x $ 4:2 > > < 1:67x 2 3 1:8 , x # 2:4 ð16:8Þ μMedium ðxÞ 5 1 2:4 , x # 3:6 > > : 7 2 1:67x x . 3:6

TABLE 16.2 Attributes of the alternatives for electric generation [39]. P (MW 3 103)

OH (h 3 103)

UL (years)

tCO2A (tons 3 103)

IR (Euro 3 103)

IP (years)

O&MC (Euro 3 103)

A1

0.5

2.35

20

1.93

0.937

1

1.47

A2

1

2.35

20

3.22

0.937

1

1.47

A3

2.5

2.35

20

9.65

1.5

1

1.51

A4

0.5

3.1

25

0.47

0.7

1.5

1.45

A5

2

2

25

0.26

0.601

2

0.7

A6

3.5

2

25

0.26

5

2.5

0.6

A7

5

2.59

25

0.48

1.803

2

4.2

A8

0.5

7.5

15

2.52

1.803

1

7.11

A9

0.5

7.5

15

2.52

1.803

1

5.42

A10

0.5

7.5

15

2.52

1.803

1

5.42

A11

0.5

7.5

15

2.52

1.803

1

2.81

A12

5.6

7.5

20

4.84

0.856

1

4.56

A13

0.2

7

20

5.91

1.503

1.5

2.51

IR, investment ration; IP, implementation period; OH, Operating Hours; O&MC, Operating and Maintenance Costs; P, Power; tCO2A, Tons of CO2 avoided; UL, useful life.

499

Multicriteria decision making Chapter | 16

TABLE 16.3 Parameters for the membership function of the variables. Trapezoidal parameters

Low: (a, b, c, d)

Medium: (a, b, c, d)

High: (a, b, c, d)

P

(0, 0, 1.2, 2.4)

(1.8, 2.4, 3.6, 4.2)

(3.6, 4.8, 6, 6)

OH

(0, 0, 1.6, 3.2)

(2.4, 3.2, 4.8, 5.6)

(4.8, 6.4, 8, 8)

UL

(0, 0, 5, 10)

(7.5, 10, 17, 17.5)

(15, 20, 25, 25)

tCO2A

(0, 0, 2, 4)

(3, 4, 6, 7)

(6, 8, 10, 10)

IR

(0, 0, 1, 2)

(1.5, 2, 3, 3.5)

(3, 4, 5, 5)

IP

(0, 0, 0.6, 1.2)

(0.9, 1.2, 1.8, 2.1)

(1.8, 2.4, 3, 3)

O&MC

(0, 0, 1.6, 3.2)

(2.4, 3.2, 4.8, 5.6)

(4.8, 6.4, 8, 8)

Benefit-type criteria

(0, 0, 0.2, 0.4)

(0.3, 0.4, 0.6, 0.7)

(0.6, 0.8, 1, 1)

Cost-type criteria

(0, 0, 0.2, 0.4)

(0.3, 0.4, 0.6, 0.7)

(0.6, 0.8, 1, 1)

Integrated performance

(0, 0, 0.2, 0.4)

(0.3, 0.4, 0.6, 0.7)

(0.6, 0.8, 1, 1)

IR, investment ration; IP, implementation period; OH, Operating Hours; O&MC, Operating and Maintenance Costs; P, Power; tCO2A, Tons of CO2 avoided; UL, useful life.

FIGURE 16.3 Membership function of input variable (criterion) P.

 μHigh ðxÞ 5

0:83x 2 3 1

3:6 , x # 4:8 4:8 , x # 6

ð16:9Þ

After determining the membership functions of fuzzy number with respect to each criterion, the IF-THEN rules should be set for inference system in the next step. A total of 81, 27 and 8 IF-THEN rules have been defined and implemented in ‘benefit-type criteria inference system’, ‘costtype criteria inference system’ and ‘integrated performance inference system’, respectively. Some of the rules in benefit-type criteria inference system

500

PART | IV Sustainability

TABLE 16.4 Examples of IF-THEN rules in ‘benefit-type criteria inference system’. IF

THEN

(P is low) AND (OH is low) AND (UL is low) AND (tCO2A is low)

Benefit-type criterion is low

(P is low) AND (OH is low) AND (UL is low) AND (tCO2A is medium)

Benefit-type criteria is low

(P is medium) AND (OH is low) AND (UL is low) AND (tCO2A is low)

Benefit-type criteria is low

(P is high) AND (OH is high) AND (UL is high) AND (tCO2A is high)

Benefit-type criteria is high

(P is high) AND (OH is high) AND (UL is high) AND (tCO2A is medium)

Benefit-type criteria is high

IR, investment ration; IP, implementation period; OH, Operating Hours; O&MC, Operating and Maintenance Costs; P, Power; tCO2A, Tons of CO2 avoided; UL, useful life.

TABLE 16.5 Examples of IF-THEN rules in ‘cost-type criteria inference system’. IF

THEN

(IR is low) AND (IP is low) AND (O&MC is low)

Cost-type criteria is high

(IR is low) AND (IP is low) AND (O&MC is medium)

Cost-type criteria is high

(IR is low) AND (IP is medium) AND (O&MC is high)

Cost-type criteria is medium

(IR is high) AND (IP is high) AND (O&MC is medium)

Cost-type criteria is low

(IR is high) AND (IP is high) AND (O&MC is high)

Cost-type criteria is low

IR, investment ration; IP, implementation period; OH, Operating Hours; O&MC, Operating and Maintenance Costs; P, Power; tCO2A, Tons of CO2 avoided; UL, useful life.

and cost-type criteria inference system are shown in Tables 16.4 and 16.5, respectively. The rule viewers that show the values of the inputs and the corresponding computed output in ‘benefit-type criteria inference system’, ‘cost-type criteria inference system’ and ‘integrated performance inference system’ are presented in Figs 16.4
16.6, respectively. According to Eq. (16.6), the results of benefit-type criteria and cost-type criteria with respect to each alternative can be determined by using the data

Multicriteria decision making Chapter | 16

501

FIGURE 16.4 Rule reviewers of ‘benefit-type criteria inference system’.

presented in Table 16.2 as inputs to run the two fuzzy inference systems. The results are shown in Table 16.6. The results of cost-type criteria and benefit-type criteria determined by two fuzzy inference systems can be integrated by running the integrated performance inference system. The integrated results are also shown in Table 16.6. The larger the score generated for integrated performance, the better the renewable energy alternative will be. Therefore, the priority sequence of these 13 alternatives can be determined (see Table 16.6). Observed from the results, A3 (wind power with 10 MW # P # 50 MW) has been recognized as the best renewable energy system. A13 (biofuels with P # 2 MW) and A12 (biomass from con-combustion in conventional central with P $ 50 MW) also perform very well according to the results determined by the fuzzy inference system. The alternatives including A6

FIGURE 16.5 Rule reviewers of ‘cost-type criteria inference system’.

FIGURE 16.6 Rule reviewers of ‘integrated performance inference system’.

Multicriteria decision making Chapter | 16

503

TABLE 16.6 The results of the alternatives. Alternative

Cost-type criteria

Benefit-type criteria

Integrated performance

Rank

A1

0.181

0.172

0.5

4

A2

0.181

0.179

0.5

4

A3

0.181

0.5

0.847

1

A4

0.153

0.158

0.5

4

A5

0.339

0.181

0.319

8

A6

0.5

0.163

0.153

11

A7

0.661

0.5

0.359

7

A8

0.661

0.163

0.179

10

A9

0.553

0.163

0.153

11

A10

0.553

0.163

0.153

11

A11

0.366

0.163

0.237

9

A12

0.339

0.847

0.821

3

A13

0.254

0.5

0.836

2

(hydroelectric with 25 MW # P # 50 MW), A9 (biomass from forest and agricultural waste with P # 5 MW) and A10 (biomass from farming industrial wastes with P # 5 MW) were ranked at the bottom among all renewable energy systems. To evaluate the feasibility of fuzzy inference system, we summarized the rankings of alternatives with respect to each criterion, as shown in Fig. 16.7. According to Fig. 16.7, the alternatives A1, A2, A3, A4, A12 and A13 should have better ranking compared with other alternatives, as the number of highranking criteria with respect to the alternatives mentioned above is more than that of the remaining alternatives. Similarly, the number of low-ranking criteria of A6 is more than that of other alternatives, so it should be one of the worst options in this study. Therefore, the result is feasible and validated, since the result determined by inference system is consistent with the analysis of original data. In addition, the results determined by the fuzzy inference system are consistent with the original research that has been carried out by using the data envelopment analysis
A3 (wind power with 10 MW # P # 50 MW) was also ranked as the best [39]. Therefore, it could be concluded that fuzzy inference system is feasible for the selection of the best renewable energy alternative.

504

PART | IV Sustainability

FIGURE 16.7 Summary of the original data.

16.4 Conclusions The selection of the best renewable energy alternative is vital for the planning of the development of renewable energy in a region. In the selection process, the uncertainties and subjectivity existing in the decision-making process need to be addressed. A fuzzy inference system is efficient in formulating the mapping from a given input to an output by using fuzzy logics. Therefore, the method for the evaluation of renewable energy alternatives based on fuzzy inference system has been presented in this study. The knowledge of the experts can be imported into the decision-making process when designing the membership functions of the variables and defining the IF-THEN rules in the fuzzy inference systems. An illustrated example has been studied with the proposed fuzzy inference system, and the result determined by the proposed method is consistent with original data and that determined by the data envelopment analysis method. It reveals that fuzzy inference system is feasible for the evaluation of renewable energy alternatives. The fuzzy inference system can be popularized to other cases for the evaluation of the renewable energy systems and the selection of the best one among multiple renewable energy systems. However, all the indicators for the evaluation of renewable energy alternatives are assumed to be equally important, and the fuzzy inference system used in this study cannot incorporate the relative importance (weights) of the indicators. Therefore, the future work of the authors is to combine weighting methods (i.e. fuzzy best-worst method and fuzzy AHP) and fuzzy logics to develop a new fuzzy inference system which can consider relative

Multicriteria decision making Chapter | 16

505

importance (weights) of the indicators in the process for selecting the best energy source/system among multiple alternatives.

References [1] Wang Y, Sun T. Life cycle assessment of CO2 emissions from wind power plants: methodology and case studies. Renew Energy 2012;43:30
6. Available from: https://doi.org/ 10.1016/j.renene.2011.12.017. [2] Nikolaidis P, Poullikkas A. A comparative overview of hydrogen production processes. Renew Sustain Energy RevAvailable from: https://doi.org/10.1016/j.rser.2016.09.044. [3] Cavallaro F. Multi-criteria decision aid to assess concentrated solar thermal technologies. Renew Energy 2009;34:1678
85. Available from: https://doi.org/10.1016/j.renene.2008.12.034. [4] Haralambopoulos DA, Polatidis H. Renewable energy projects: structuring a multi-criteria group decision-making framework. Renew Energy 2003;28:961
73. Available from: https://doi.org/10.1016/S0960-1481(02)00072-1. [5] Akella AK, Saini RP, Sharma MP. Social, economical and environmental impacts of renewable energy systems. Renew Energy 2009;34:390
6. Available from: https://doi. org/10.1016/j.renene.2008.05.002. [6] Lee AHI, Chen HH, Kang HY. Multi-criteria decision making on strategic selection of wind farms. Renew Energy 2009;. Available from: https://doi.org/10.1016/j. renene.2008.04.013. [7] San Cristo´bal JR. Multi-criteria decision-making in the selection of a renewable energy project in spain: the VIKOR method. Renew Energy 2011;. Available from: https://doi. org/10.1016/j.renene.2010.07.031. [8] Aras H, Erdo˘gmu¸s S, ¸ Koc¸ E. Multi-criteria selection for a wind observation station location using analytic hierarchy process. Renew Energy 2004;29:1383
92. Available from: https://doi.org/10.1016/j.renene.2003.12.020. [9] Catalina T, Virgone J, Blanco E. Multi-source energy systems analysis using a multicriteria decision aid methodology. Renew Energy 2011;36:2245
52. Available from: https://doi.org/10.1016/j.renene.2011.01.011. [10] Hong T, Koo C, Kwak T, Park HS. An economic and environmental assessment for selecting the optimum new renewable energy system for educational facility. Renew Sustain Energy Rev 2014;. Available from: https://doi.org/10.1016/j.rser.2013.08.061. [11] Va¨isa¨nen S, Mikkila¨ M, Havukainen J, Sokka L, Luoranen M, Horttanainen M. Using a multi-method approach for decision-making about a sustainable local distributed energy system: a case study from Finland. J Clean Prod 2016;137:1330
8. Available from: https://doi.org/10.1016/j.jclepro.2016.07.173. [12] Ahmad S, Tahar RM. Selection of renewable energy sources for sustainable development of electricity generation system using analytic hierarchy process: a case of Malaysia. Renew Energy 2014;63:458
66. Available from: https://doi.org/10.1016/j. renene.2013.10.001. [13] Strantzali E, Aravossis K. Decision making in renewable energy investments: a review. Renew Sustain Energy Rev 2016;. Available from: https://doi.org/10.1016/j. rser.2015.11.021. [14] Wang JJ, Jing YY, Zhang CF, Zhao JH. Review on multi-criteria decision analysis aid in sustainable energy decision-making. Renew Sustain Energy Rev 2009;. Available from: https://doi.org/10.1016/j.rser.2009.06.021.

506

PART | IV Sustainability

[15] Kahraman C, Kaya I, Cebi S. A comparative analysis for multiattribute selection among renewable energy alternatives using fuzzy axiomatic design and fuzzy analytic hierarchy process. Energy 2009;. Available from: https://doi.org/10.1016/j.energy.2009.07.008. [16] Karunathilake H, Hewage K, Me´rida W, Sadiq R. Renewable energy selection for netzero energy communities: life cycle based decision making under uncertainty. Renew Energy 2019;130:558
73. Available from: https://doi.org/10.1016/j.renene.2018.06.086. [17] Begi´c F, Afgan NH. Sustainability assessment tool for the decision making in selection of energy system-Bosnian case. Energy 2007;. Available from: https://doi.org/10.1016/j. energy.2007.02.006. [18] Kaya T, Kahraman C. Multicriteria decision making in energy planning using a modified fuzzy TOPSIS methodology. Expert Syst Appl 2011;38:6577
85. Available from: https:// doi.org/10.1016/j.eswa.2010.11.081. [19] Jang J.S.R. Fuzzy modeling using generalized neural networks and Kalman filter algorithm AAAI 1991;91:762
767. [20] Jang JSR. ANFIS: adaptive-network-based fuzzy inference system. IEEE Trans Syst Man Cybern 1993;. Available from: https://doi.org/10.1109/21.256541. [21] Ocampo-Duque W, Ferre´-Huguet N, Domingo JL, Schuhmacher M. Assessing water quality in rivers with fuzzy inference systems: a case study. Env Int 2006;32:733
42. Available from: https://doi.org/10.1016/j.envint.2006.03.009. [22] Guimara˜es ACF, Lapa CMF. Fuzzy inference to risk assessment on nuclear engineering systems. Appl Soft Comput J 2007;7:17
28. Available from: https://doi.org/10.1016/j. asoc.2005.06.002. [23] Elsayed T. Fuzzy inference system for the risk assessment of liquefied natural gas carriers during loading/offloading at terminals. Appl Ocean Res 2009;. Available from: https://doi. org/10.1016/j.apor.2009.08.004. [24] Amindoust A, Ahmed S, Saghafinia A, Bahreininejad A. Sustainable supplier selection: a ranking model based on fuzzy inference system. Appl Soft Comput J 2012;12:1668
77. Available from: https://doi.org/10.1016/j.asoc.2012.01.023. [25] Herva M, Franco-Ur´ıa A, Carrasco EF, Roca E. Application of fuzzy logic for the integration of environmental criteria in ecodesign. Expert Syst Appl 2012;. Available from: https://doi.org/10.1016/j.eswa.2011.09.148. [26] Mamlook R, Badran O, Abdulhadi E. A fuzzy inference model for short-term load forecasting. Energy Policy 2009;37:1239
48. Available from: https://doi.org/10.1016/j. enpol.2008.10.051. [27] Moghaddamnia A, Ghafari Gousheh M, Piri J, Amin S, Han D. Evaporation estimation using artificial neural networks and adaptive neuro-fuzzy inference system techniques. Adv Water Resour 2009;32:88
97. Available from: https://doi.org/10.1016/j. advwatres.2008.10.005. [28] Sabaghi M, Mascle C, Baptiste P, Rostamzadeh R. Sustainability assessment using fuzzyinference technique (SAFT): a methodology toward green products. Expert Syst Appl 2016;. Available from: https://doi.org/10.1016/j.eswa.2016.02.038. [29] Kaushal J, Basak P. A novel approach for determination of power quality monitoring index of an AC microgrid using fuzzy inference system, Iran. J Sci Technol Trans Electr Eng 2018;. Available from: https://doi.org/10.1007/s40998-018-0087-z. [30] Rustum R, Kurichiyanil AMJ, Forrest S, Sommariva C, Adeloye AJ, Zounemat-Kermani M, et al. Sustainability ranking of desalination plants using mamdani fuzzy logic inference systems. Sustain 2020;. Available from: https://doi.org/10.3390/su12020631.

Multicriteria decision making Chapter | 16

507

[31] Cavallaro F. A Takagi-Sugeno fuzzy inference system for developing a sustainability index of biomass. Sustainability 2015;. Available from: https://doi.org/10.3390/ su70912359. [32] Guimara˜es ACF, Lapa CMF. Fuzzy inference system for evaluating and improving nuclear power plant operating performance. Ann Nucl Energy 2004;31(3):311
22. [33] Hemdi AR, Saman MZM, Sharif S. Sustainability evaluation using fuzzy inference methods. Int J Sustain Energy 2013;32(3):169
85. [34] Mazloumzadeh SM, Shamsi M, Nezamabadi-pour H. Evaluation of general-purpose lifters for the date harvest industry based on a fuzzy inference system. Comput Electron Agric 2008;. Available from: https://doi.org/10.1016/j.compag.2007.06.005. [35] Wang YJ. Ranking triangle and trapezoidal fuzzy numbers based on the relative preference relation. Appl Math Model 2015;39(2):586
99. Available from: https://doi.org/ 10.1016/j.apm.2014.06.011. [36] Gharibi H, Sowlat MH, Mahvi AH, Mahmoudzadeh H, Arabalibeik H, Keshavarz M, et al. Development of a dairy cattle drinking water quality index (DCWQI) based on fuzzy inference systems. Ecol Indic 2012;20:228
37. Available from: https://doi.org/10.1016/j. ecolind.2012.02.015. [37] Nathan AJ, Scobell A. How China sees America. Foreign Aff 2012;91. Available from: https://doi.org/10.1017/CBO9781107415324.004. [38] Gulley N, Jang J-SR. MATLAB fuzzy logic toolbox user’s guide. USA: The MathWorks. Inc; 1997. [39] San Cristo´bal JR. A multi criteria data envelopment analysis model to evaluate the efficiency of the renewable energy technologies. Renew Energy 2011;36:2742
6. Available from: https://doi.org/10.1016/j.renene.2011.03.008.

This page intentionally left blank

Part V

Policy

This page intentionally left blank

Chapter 17

How much is possible? An integrative study of intermittent and renewables sources deployment. A case study in Brazil Fernando Amaral de Almeida Prado, Jr Sinerconsult Consultoria Treinamento e Participac¸o˜es Limitada, Sa˜o Paulo, Brazil

Chapter Outline 17.1 Introduction
understanding of the question 511 17.2 Irresistible expansion 515 17.2.1 Wind 516 17.2.2 Solar 516 17.3 Undesirable effects of the intermittent renewable resources expansion 516 17.3.1 Complexity 517 17.3.2 The operation problem with the increasing insertion of

intermittent renewable resources 521 17.3.3 Economic effects 526 17.3.4 Externalities and the merit order effect 529 17.4 Rebound effect
social acceptance of intermittent renewable sources
the opponents 533 17.5 Conclusions 535 17.6 Acknowledgments 535 References 535

17.1 Introduction
understanding of the question This article discusses the existence of limits to the expansion of intermittent renewable resources (IRRs) to meet electricity demand and the risks that societies are willing to face when considering this expansion. As we know, energy risks comprise four dimensions: supply capacity (adequacy), energy quality (power quality), risks associated with high costs (which society may not be able to cope with), and environmental risks. This article does not directly analyse the dimension of energy quality, mainly

Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00017-0 © 2021 Elsevier Inc. All rights reserved.

511

512

PART | V Policy

FIGURE 17.1 Generation capacity in Brazil (%). Source: Elaborated from EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia-2029. ,http://www.epe.gov.br.; 2020 [accessed 20.03.20] [2].

associated with the availability of transmission and distribution networks, although this dimension is also implicitly addressed in the text. Power generation and the risks associated with the scale of availability are absolutely linked to the existence of natural resources (renewable or otherwise) such as water availability, oil and gas reserves or wind and sunlight. It happens that, within an economic logic, the closest resources are always prioritized and therefore the most economical. As resources become scarce, either because they are depleted or more distant from the main centres of consumption, environmental impacts tend to be more intense and costs higher. The example of Hydroelectric Power Plants (HPP) in Brazil, a country that has guided its electric power industry from water resources and still has ample potential to be explored,1 is emblematic. Due to growing opposition from environmental activists, Brazilian hydroelectric plants have increasingly been designed to format run-of-river plants, avoiding large reservoirs. The ‘stock’ of energy that can be stored in these hydroelectric plants has been reducing in the timeline [1]. Fig. 17.1 shows the reduction of the electricity production capacity of the Brazilian hydroelectric plants (installed capacity) at the same time that it signals a rather robust expansion of intermittent sources in this decade. The first dimension of risk deals with the capacity to serve (adequacity) from an infrastructure of plants and systems capable of using energy

1. Despite the fact that these are located at great distances from the load centres (usually more than 3000 km away) and with a greater complexity due to the fact that they are mostly found in the area of the plain that surrounds the great Brazilian rainforest
the Amazon.

How much is possible? Chapter | 17

513

´ FIGURE 17.2 Excess capacity on demand %. Source: Elaborated from PSR. Analise estrutural do suprimento. Energy Rep. 2020;157:15
6.

resources. It happens that with the exhaustion of the most favourable resources, the search for new projects, less favourable from the economic point of view or more distant or even more sensitive to environmental impacts is the solution. Despite these restrictions, as the Brazilian economy has faced a severe economic recession in recent years, the supply dimension cannot be considered problematic, with a consistent structural surplus in the coming years. The mechanisms of compulsory contracting by distribution companies, defined by the Brazilian regulator, also contribute to assure the security. Fig. 17.2 shows this stable structural surplus over the next 5 years.2,3 The environmental security dimension, however, is permanently in conflict because of the existence of natural resources for the construction of new HPPs in the northern region of the Amazon’s outskirts. This fact has intensified the debate on the possibilities about future generation could be based only on intermittent renewable ones, a thesis that is ardently defended by environmentalists. At the end of 2005, Brazil had 69.6 GW of installed capacity in hydroelectric plants and planned to install a further 31.1 GW until 2015 from hydraulic sources. However, elapsed 5 years (2010) the installed capacity of hydroelectric plants had grown by only 4.7 GW and the official planner of the Brazilian electrical system estimated that by 2020 new 40.8 GW of hydroelectric projects would be implemented. The reality showed that in the last decade (2010
19) 23.3 GW were implemented, that is, little more than the planned half. Again, in a new planning cycle, which is now beginning, there is an expectation of growth in hydroelectric works, but now with much 2. It should always be noted that it does not make sense to assess supply risks for horizons longer than 5 years, because in the event of an imbalance between supply and demand, there would always be time to implement new projects. 3. The figures are analysed before the COVID-19 crises.

514

PART | V Policy

TABLE 17.1 HPP expansion planning 3 reality. Year

Horizon (years)

HPP capacity (GW)

Projected addition of HPP in the planning horizon (GW)

Implemented in the period (%)

2005

10

69.6

31.1

42.4

2010

10

74.3

40.8

57.1

2020

10

97.6

1.7




Source: Elaborated from EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia-2029. ,http://www.epe.gov.br.; 2020 [accessed 20.03.20]; EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia-2015. ,http://www.epe.gov.br.; 2005 [accessed 20.03.20]; EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia2027. ,http://www.epe.gov.br.; 2017 [accessed 20.03.20].

less emphasis, since only 1.7 GW of large hydro resources are being planned for implementation by 2030, which shows a potential withdrawal of Brazil to make use of the still possible uses. All of this information are available in the planning reports from EPE [1,3,4], Table 17.1 presents a summary of this context. These figures exemplify the importance of environmental risk with the ‘victory’ of the HPP opponents, either by the inability to meet the planned expansion or by the very rapid expansion of intermittent sources already implemented or planned. Table 17.2 presents the evolution of these renewable sources and the associated planning scenarios. As it is easy to see, as in different ways to the hydroelectric power plants, intermittent renewable sources have been surpassing the expectations of the Energy Research Company
EPE,4 the Brazilian entity responsible for energy planning in Brazil. In the dimension of the tariffs, considering that Brazil is a country with a long tradition of inflation and indexed prices, there is a culture of monitoring prices in terms of present value. Fig. 17.3 presents the evolution of residential tariffs of a large electricity utility on a 100 basis, corrected for inflation for 2016 present values. Although this figure refers to a specific concessionaire, the profile of this evolution is homothetic with the other 64 companies that operate in this segment around the country. Again, it is easy to understand that real prices have been reduced in the timeline, which would reflect the conclusion that also in the drive of prices the risk is not relevant.5

4. EPE is the acronym in Portuguese language. 5. We are not analysing whether a possible worsening of the economy contributes to the increase of the economic risk of energy prices based on the per capita income of the population.

How much is possible? Chapter | 17

515

TABLE 17.2 Wind and solar expansion planning 3 reality. Year

Capacity (GW)

Planned expansion of capacity in 10 years (GW)

Wind

Photovoltaic

Wind

Photovoltaic

2010

0.9










2015

8.7

0.04

10.0




2020

15.4

2.7

25.5

8.5

Source: Elaborated from EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia-2029. ,http://www.epe.gov.br.; 2020 [accessed 20.03.20]; EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia-2015. ,http://www.epe.gov.br.; 2005 [accessed 20.03.20]; EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia2027. ,http://www.epe.gov.br.; 2017 [accessed 20.03.20].

FIGURE 17.3 Evolution of tariffs in Residential Segment-Base 100 (present value 2016). Source: Elaborated from ANEEL-Agencia Nacional de Energia Ele´trica. Resoluc¸o˜es tarifariasdiversos anos, 2001
2016. ,http://aneel.gov.br/biblioteca/resolucoeshomologatorias.; 2020 [accessed 20.12.17] [5].

Therefore, keeping the other conditions constant,6 the article deals with the risk conditioning factors that arise from the significant expansion of intermittent renewable sources in order to characterize an eventual limit to this expansion.

17.2 Irresistible expansion The publication A New World
The Geopolitics of the Energy Transformation of the International Renewable Energy Agency [6] acknowledges the great energy transformation the world is experiencing with the unprecedented growth rate of renewable energy sources, in particular solar 6. Ceteris paribus.

516

PART | V Policy

and wind, as well as the International Energy Agency’s World Energy Outlook [7] evidence where every year expectations are exceeded by reality. The mentioned reports attribute this explosion in growth, which they prefer to call energy revolution, to some forces of change, including the declining costs, pollution and climate change, the emergence of new business models, and policies to encourage renewables stand out.

17.2.1 Wind The expansion of installed capacity of wind farms in the world has grown at an average rate of 20.7% per year since 2001, reaching about 591 GW at the end of 2018 according to the information contained in the Global Wind Statistics report published by the Global Wind Energy Council [8]. In Brazil, the growth profile of wind power plants is no different. In the period between 2005 and 2018, the average annual growth was 62.3% per year according to the Brazilian Wind Energy Association [9]. The same entity informs that in the forecast up 2023 this growth will still remain extremely high with an average annual performance of 44.2% per year [9].

17.2.2 Solar The expansion of solar sources shows the same very strong trend. In 2018, for the first time 100 GW were installed in a single year. In the year 2000, the world installed capacity of photovoltaic plants was only 1.2 GW, when Germany started its feed-in tariff incentives. In 2007, there were 9.2 GW of solar photovoltaic plants and by 2018 this capacity had already reached 505 GW, that is, a growth of more than 42,000% since the beginning of the century. Fig. 17.4, elaborated from the Renewables 2019 Global Status Report, shows the annual growth of installed and accumulated capacity over the last 10 years [10]. For the future, the Solar Power Europe organization signals a continuity of this exacerbated growth, predicting in its most conservative scenario that in 2023 there will be 610 GW of installed capacity and in the most daring scenario 1044 GW, resulting in a growth of 15.4% per year (pessimistic scenario) or 25.9% in the most optimistic scenario [11]. In Brazil, this movement is still in its beginning, but the trend will be similar to the international picture as evidences from Table 17.2.

17.3 Undesirable effects of the intermittent renewable resources expansion The previous section presents data from an extraordinary expansion of the IRRs; however, there may be effects not as desirable as it might seem at first analysis.

How much is possible? Chapter | 17

517

FIGURE 17.4 Photovoltaic capacity (GW). Source: Elaborated from UNEP-United nation Environment Programme. REN
Renewables Global Status Report. ,https://wedocs.unep.org/ 2019 bitstream/handle/20.500.11822/28496/REN2019.pdf?sequence 5 1&isAllowed 5 y.; [accessed 14.01.20].

17.3.1 Complexity The first dimension of these effects is the increasing technical complexity in the operation of interconnected electrical systems. As is well known, the interconnected operation of large electrical energy systems has never been trivial. Even in the early years of the industry, adjustments to maintain the stability of the pioneering interconnections was a very difficult task and required frequent adjustments between the operators of different plants. Until the mid of the 1930s, adjustments were manual and required frequent telephone contacts between operators, with low quality and insecurity of supply being the marks of this process [12]. The economic crisis that pervaded the 1930s, still in the last century, was curiously a drive to motivate advances in the search for greater reliability. President Roosevelt, involved with the serious economic and social problems in the United States, tried in every way to rescue the economy and minimize unemployment. The creation of the Federal Power Commission (FPC), the Tennessee Valley Authority and the implementation of a comprehensive rural electrification plan increased the demand for greater controls [13]. In 1935, the FPC declared that the reliability of electrical systems was not just an engineering problem, but an economic and social issue and as such required all possible efforts in its equation [13]. The Second World War increased the need for reliable systems as a result of the high demand for energy for the industrial purposes required in the war effort. In the 1950s, the technological race was to automate load and generation controls in the search for savings for utilities. Only in the late 1950s did the North American Power System Interconnection Committee consider the possibility of interconnections being extended from coast to coast [13].

518

PART | V Policy

In 1967, the American interconnected systems (involving some regions of Canada) accounted for 94% of all connections. Part of this interconnection effort was greatly encouraged by the 11/9, 1965 blackout, which involved some 30 million people in an area of 200,000 km2 for 13 hours [14]. It was the first electrical accident of this magnitude and became a benchmark for damages to the economy and the safety of people and institutions. It is worth noting that accidents of this magnitude had not yet been mapped as a possibility. According to Cohn, the specialists in electrical systems until then had not anticipated that electrical grids could be subject to such a relevant cascading effect. Many other countries and even North American interconnections later experienced other episodes of major shutdowns. A relevant example of the impacts that these accidents can have on people’s lives and the economy is the Great Blackout of India in 2012, which lasted over 48 hours and involved 700 million people [13]. Brazil also had experiences with massive blackouts, in 1999 with 97 million people affected, in 2009 with 67 million affected, and in 2014 involving 12 million people [14,15]. The need for coordinated operation of electrical systems, as already evidenced, arises as a result of nontrivial technical demands, which require the operation and planning functions of interconnected networks to enable the most economical and safe (reliable) drives of plants and the grid. They also need to provide nondiscriminatory access to the interconnected system, facilitating the highest possible level of competitiveness. The second dimension is that of ensuring the competitiveness promoted by competitive access to networks. Free access to the Transmission and Distribution networks not only requires coordination in the short term, but also medium- and long-term arrangements. This is a dimension of the problem that had not arisen when, in the mid-1960s, the interconnection of systems was becoming a necessity. At the time, the concessionaires were building the lines for their own use and were not dedicated to evaluating a pricing system for their availability. With this reality, fundamental questions arise. Who will focus on planning and building lines for the competition? What are the fair prices? What are the rights of who builds and who can become an accessor? If networks are in absolute control of a few agents, how would this affect competitiveness and how can discrimination be avoided? Alternatively, if the networks are owned by multiple agents, who will exercise coordination in a neutral way? All these questions can be found in Sally Hunt’s classic book and serve as an economic justification for the existence of independent operators of the electric grid [16]. These reasons transcend electrical and energy reasons. Hunt’s questions were present in the experience in California in 2001, when Enron scandalously manipulated prices by exercising market power for which the regulator was not properly prepared to avoid [17,18]. Interestingly, the competitiveness bottlenecks initially pointed out by Hunt [16] may be changing. Asset ownership, which has always represented

How much is possible? Chapter | 17

519

market power, is changing thanks to online accounting and measurement apps capabilities. The situation is reversed and the ‘Small Distributed Generation Aggregator,’ with no fixed asset investment costs, now has advantages. The market power now resides in free access to networks, operating costs and tariffs that apply to the services provided by these assets. In this scenario, regulatory agencies are increasingly dependent on the expertise of independent operators for the balance between competitiveness and security associated with the challenges of expansion of the Small Distributed Generation units and its aggregators. Another dimension to consider is the planning of the ‘electric grid,’ which still has to live with the obsolescence of the infrastructure. This finding is made explicit by Bakke [13], exemplifying that in the United States transmission lines are on average more than 25 years old and generation plants, also in average terms, more than 35 years old. This challenge ends up being the responsibility of the Independent System Operators (ISOs) and Regional Transmission Operators (RTOs) if not for institutional determinations, for practical reasons for safety. In the same reference [13], Bakke concludes that these demands are not only of technology and investments, but of rules and laws that determine the incentives to promote the upgrading of infrastructure to the required level. The priority of IRRs as public policy, when it happens, needs to be accompanied by infrastructure to meet the new operational demands that will result from these choices. These policies can facilitate or create entry barriers for investors in renewable projects and even traditional enterprises such as thermal power plants. The latter are fundamental for supplying reserves, acting on the basis of the system or in a complementary way. The intense growth of these IRRs, it is not too much to repeat, has led to the emergence of nondispatchable plants.7 In spite of these intermittent characteristics, the IRRs can be used to meet energy needs, but they can also help in strategies to modulate the load and to supply ancillary services, but obviously they depend on competence not yet fully installed in the ISOs/RTOs. Among other demands, it is possible that these independent entities are called upon to manage the so-called ancillary services, which briefly represent the electrical needs of the system (electric and not energetic). They may also require the existence of systems dedicated to supervision and communication, allowing the achievement of operational stability and security in subregions affected by specific contingencies from the regional point of view, but recurring in the timeline. The border measurements between subsystems, the exchange of energy and power between adjacent systems, and short-,

7. Plants whose natural resources happen without any link with the needs of demand, and generation may occur at times where there are no requirements for its production.

520

PART | V Policy

medium- or long-term planning, as already mentioned, also enter the list of attributions of these operative entities. The same can be applied in relation to decisions related to safety, as we can exemplify the Brazilian case with the existence of the Electric Sector Monitoring Committee, which can, in a discretionary manner, establish operating orders outside economic merit order in a manner superior to the decisions of the Brazilian National Operator. Thus there may be ancillary services of an electrical level and of a security. Its establishment and funding are at the core of independent operators’ activities that affect the tariffs. The complexity of the problems faced by these entities still tends to take shape due to the growing importance that electric energy has gained in all facets of modern life, where any interruption produces physical and economic risks that are no longer acceptable to contemporary societies. This is also due to the increasing and accelerated insertion of IRRs, which bring with their intermittence a great volatility of production. The same happens with the uncertainties about energy production that have always existed in technological, climatic and demand behaviour aspects. The atomization of production, with thousands (in some time there will be millions) of units of Small Distributed Generation (SDG) with consumers who alternate the role of producers, consumers and storage, makes this picture even more complex. It is also worth mentioning that the RTOs/ISOs have barely begun to confront this SDG expansion and the future influence of electric mobility, which is still in its initial phase. The literature on these consequences is in its childhood. To cope with these uncertainties, there is a need for increasingly rapid responses from operational decisions, and without such independent entities, this ability to respond would be unfeasible. Apart from the advantages to be found in this proliferation of IRRs, there is no doubt that systems are becoming increasingly complex. The main challenges are to standardize technical criteria and give consistency between adjacent ISOs/RTOs and even between all operators in the same country, a fact that is far from being achieved. The conceptual problem of RTOs/ISOs has not only become increasingly complex (it is not too much we repeat again), but it is no longer, simply a problem of attending to imbalance energy and power, but it is also an issue that affects the governance of the electricity industry and the very shaping of public policies. To better illustrate this dimension of public policies, one can exemplify with decisions regarding the pursuit of competition (free access, transparency of rules, isonomy of treatment), quality, safety and the existence of structured subsidies specifically for intermittent sources. In Brazil, this situation resulting from the volatility of production capacity (and prices) does not really represent a picture of novelty, since the hydraulic generation with HPPs with no reservoirs has already been a reality for the last 20 years. The increase of this complexity is due to the

How much is possible? Chapter | 17

521

impossibility of new hydroelectric plants with storage capacity in multiyear storage reservoirs, especially for environmental licensing reasons.

17.3.2 The operation problem with the increasing insertion of intermittent renewable resources It seems to be clear that system operators have always faced difficulties in adjusting production to demand; however, this has always occurred more intensely (except in electric accidents) due to fluctuations on the demand side (economy, climate and other specificities). At present, with exacerbated effects, the difficulties of adjustment between supply and demand have become more intense due to the intermittence of renewable sources and whose resources cannot be stored such as coal, oil, gas and hydroelectricity. Another highlight is that this intermittence and seasonality are greatly aggravated by the difficulty of forecasting systems. Just to give a single example we can cite Ackerman and others [19] quote that on 29 March 2013, the forecasting systems indicated a potential supply of 11 GW of wind power plant capacity in the Spanish system, but what was available in practice was only 2 GW. It is known that the production predictability techniques of these IRRs are being implemented with significantly progress, but are still subject to large deviations. The predictability of the load has a good assertiveness of its forecast, with errors in the order of 1.5% for the following day and 5% for the following week as pointed in the previous section. The biggest source of errors is caused by temperature fluctuations. Contingencies are in these cases adjusted by base generation, intermediate response plants and finally by rapid response sources. It should be noted that reserves have always been required for adjustments; however, the intermittence of renewable sources (not dispatchable) has made these needs more drastic. Not only are these adjustment requirements becoming more and more frequent, but the speed of adjustment required has been more and more intense. The ‘duck curve’ is an excellent example of this intensity. The famous curve, which identifies the variance of the load in California in the face of the increasing insertion of photovoltaic generation in the timeline, shows that in 2020 projection there is a need for a 13 GW ramp in just 3 hours [20], as we can see in Fig. 17.5. In specific situations, these ramps can be even more drastic. The EPE official Brazilian energy planning entity reports the occurrence in August 2018 when it was necessary to climb 9079 MW in a single hour [1]. These challenges have changed the dynamics of operators’ time. In the book by Madrigal and Porter [21], the timescale is portrayed as being established in three scales of order of magnitude, namely, seconds to minutes, hours and the following day.

522

PART | V Policy

FIGURE 17.5 The duck curve. Source: Reproduced from CAISO-California ISO. What the duck curve tell us about managing a green grid. ,http://www.caiso.com/Documents/ FlexibleResourcesHelpRenewables_FastFacts.pdf. [accessed 09.02.20].

The dimension of the Electric Regulation is that where the Operator continuously requests adjustments for increase or decrease of the generation, generally in a scale that goes until a few minutes (generally 10
15 minutes). Sometimes, these adjustments are even made automatically by Automatic Generation Control (AGC) systems. Settings that transcend this short-term interval require the input (or output) of units that are available as preprepared reserves or those that can be quickly triggered. Finally, they require those units that can be triggered with a response time that can transcend a few hours and even the next day. Table 17.3 presents a summary of these timescales in selected countries. Reserves are usually defined as the additional capacity that the System Operator needs to have (‘on line’ or ‘off line’) to meet energy demand and ensure reliability whenever the load or generation differs from previously planned. The insertion of IRRs has made this task more difficult. Although this issue is preponderant in the impact of additional costs, it has such a fundamental aspect in the operation of the interconnected system that it must be treated separately. The basic problem concerns the predictability of the IRRs’ production and how much assertiveness can be obtained. It is always possible to minimize generation problems at very short intervals (seconds to hours), but predicting the next day can be much more difficult. The existence of different prices at different times of the day encourages the production of energy in the right quantity, at the right time as a result of the greater economic pressure of the settlement of the difference between the contractor and the actual consumption. Therefore the economic dimension represents an incentive to improve the forecasting assertiveness of the agents.

How much is possible? Chapter | 17

523

TABLE 17.3 Reserve time in selected countries. Country

Fast reserves

Medium time reserves

Reserves in long time

Germany

Primary reserves available in 30 s, Dispatched by the RTO

Secondary reserves
5 min dispatched by RTOs

Secondary reservations required up to 15 min

N.A.

Ireland

Primary Reserves
required in 15 s (inertial or quick response)

Secondary Reserves
15
90 s

Tertiary Reserves
Above 90 s

N.A.

United States

1
5 s 1 min
1 h

1-h horizon, but in increments of 5 in 5 or 10 in 10 min

Response time. 1 h to 1 week with hourly increments

Source: Elaborated from Madrigal M, Porter K. Operating and planning electricity grids with variable renewable generation. World Bank; 2013. p. 125.

The literature also indicates that regulatory status can make a difference in these definitions, and the ‘obligation to serve’ on the part of distributors ends up inducing higher costs than those where there is an environment where the competitive market is compulsory. The problem, so important as it is, has led many ISOs to adapt their operating codes to accommodate the growing insertion of IRRs. Despite this perception, in many cases, the adaptations focus much more on ‘ancillary’ services than on the operation itself. This is due to the geographical dispersion of the IRRs and the dependence on the portfolio of pre-existing plants before the massive insertion of renewable energies. The quality of the assertiveness of this task (prediction of reserves) depends on the size of the control areas (submarkets) and the capacity of exchange between different regions (i.e. how robust the transmission system is) and finally the balance between supply and demand. The problem of defining the required reserves is still dependent on the speed of entry into operation of the available plants ‘on line,’ the level of available automation (AGC) and the number of plants considered inflexible in that interconnected system (technical or commercial inflexibility8).

8. A nuclear power plant could be a good example of technical inflexibility and Natural Gas power plants with contracts take or pay represent a commercial inflexibility.

524

PART | V Policy

This theme is still developing and at the frontier of knowledge. The operators have found many alternatives, but none of them has yet been consecrated as a definitive experience. The California ISO, incidentally the first to express concern about the massive expansion of IRRs, uses as a basic rule that 1% of installed capacity should be added to the available reserves of dispatchable sources for every 100 MW of intermittent sources added to the system [21]. The Independent Transmission Operator of Germany considers the need for reserves with a primary capacity of 3 GW capable of going into operation in 30 seconds. Another 4.9 GW are available to be required in 5 minutes, basically backed by hydroelectric plants some of them reversible. A third block of plants made available for a reserve can even be composed of IRRs plants with an amount of 2.4 GW for which a start-up is expected within 15 minutes [21]. The interconnected system of Texas analyses the capacity of reserve plants actually used in the last 30 days and the same month in the previous year. It also makes the same analysis of the installed capacity of IRRs in the last 30 days and compare with the same month of the previous year. For each additional 1000 MW of installed IRRs capacity, a multiplier ( . 1) is used on the reserves actually used in the most relevant case (last month or the previous year) [21]. In Brazil, the discussion about operational reserves gains other strategic contours, since there are still unexplored hydroelectric plants with large storage capacity or, alternatively, for environmental reasons, the preferential choice for plants without reservoirs (run-of-river). It is clear that these requirements ultimately demand greater technical and operational competence from the system operators, so more resources are required to cope with these demands. Consequently, there is also an increase in operating costs. These result from the greater demand for available reserves, greater computational resources, automation of the operation and more human resources, including greater intellectual competence to cope with the challenges required in this new context. In addition, the required safety levels tend to be amplified by the growing importance that electric energy is gaining in society and economy. Recent studies indicate additional costs of up to 10 USD/MWh for up to 20% of installed capacity to be performed by IRRs [21
23] The problem of additional costs in operating systems with high IRRs penetration is of extreme complexity, as there are criteria that need to be agreed on whether or not to consider some items. For example, there is a discussion if we should to whether or not considered the investment for more robust transmission systems, those are required by more frequent exchange between subsystems. The same applies to consideration of opportunity costs of generators that are required to remain in reserve by ISOs/RTOs requisition. They should be considered or not under this heading of additional costs?

How much is possible? Chapter | 17

525

Although these costs are real, usually only the operating costs of more qualified teams, complementary studies and different forms of operating interventions are computed as additional costs, such as the availability of short run reserves, entry into operation with more intense ramps and entry into operation without the proper optimization of the climbing process. Thus the cost can increase if the reservation requirements are also increased with a higher demand side escalation. Note that not all costs are in principle from the IRR generator but they are the main cause of them, although eventually they may be borne by the back-up source generators. Obviously, at some point these costs will need criteria about their partition. There are basically two alternatives for calculating the additional cost, as reported by the World Bank [21]. The first alternative contemplates modelling with different IRRs penetration scenarios and with different generation scenarios (stochastic generation models) comparing the expected operating cost for each of these scenarios compared with the same installed capacity without IRRs. The difference between IRRs disregarded just for cost assessment can be made from a hypothetical capacity considering the same composition actually installed as traditional sources or considering a set of traditional sources with the best available technology. The second alternative defines a period of time (e.g. 3 years) and models an analysis considering that the production of the IRRs of the last 3 years would have been deterministic and compared with the actual cost incurred. To minimize the hypothesis of total assertiveness, it can be adopted that a relevant percentage of intermittent production would have been assertively anticipated by the modelling. The additional cost is determined by the difference between the cost that actually occurred and that which would have occurred if the IRRs had been predicted with high assertiveness [21]. These alternatives face difficulties in their implementation, as data are not always available at short intervals (e.g. data every half hour), and the construction of coherent hypotheses of hypothetical complementary sources of IRRs substitution is also complex. The methods described here are generally intensive in data and team availability (often requiring several years to complete). This kind of experience is often postponed because of the inherent difficulties and costs. Rarely are developed with the insertion of IRRs less than 10%. However, the problem is not only how to calculate the add-ons from IRRs, but also how to divide them among the agents. It is important to remember that some costs for thermal agents exist only because there are IRR agents. So what would be the fairest logic of cost apportionment: G G

Individually? As a proportion of the impact of the plant on the subsystem where it is inserted?

526 G G G G

PART | V Policy

Isonomically among all the generators? Isonomically among all consumers? Any hybrid model of these alternatives? The same concept can be faced to share systemic benefits, such as the availability of avoided emissions certificates.

Finally, considering that there is an inevitable increase in costs, what would be the alternatives to reduce its impact? Although the answers may seem obvious, they are very difficult to implement, requiring increased predictability of intermittent production and conditions for energy storage. The opportunities for more intense storage are still far from becoming viable, but even so, the theme is already on the radar of aggregating companies. The main theme will be the possibility of arbitrage of energy stored at very low or even negative prices. However, many questions remain to be addressed in the 5
10 years, such as the still high costs of storage in batteries, a better portability and a real understanding of whether its intense use could cause more exacerbated environmental impacts. What we can anticipate is that as the batteries advance in the market, probably due to an expansion of electric mobility, the limits discussed here may change drastically [24,25].

17.3.3 Economic effects As it was already possible to foresee in the development of the previous section, the complexities pointed out, result inexorably in economic complexities, many of which are characterized by various controversies. Wouldn’t accelerated expansion be an indication that renewable sources have become competitive and that they would not need more incentive regulations? How much expansion would be possible given that intermittence will require a more complex operation, requiring operational reserves and complex ancillary services. A literature review usually indicates limits ranging from 20% to 30% of total installed capacity as a physical and economic limit to IRR penetration. There are many references in the literature, to nominate a few we can list Trainer [26], Mills [27] and Hirth [28]. This last one inclusive believes that the limits are under 20%. On the other hand, there are even the understanding of some authors that technological advance could minimize the difficulties faced of this accelerated expansion and promote overpassing the limit of 30% and around [29]. For example, (1) the new digital controls of wind turbines allow them to contribute inertia, reactive support and even reserve to offset variability; (2) the geographical diversification of sources (portfolio effect) allows a ‘firm’ generation comparable to that of a thermal plant at the base; (3) the ‘binding dispatch’ in the price offer for the ‘day ahead’ led to an extraordinary

How much is possible? Chapter | 17

527

improvement in the forecasting capacity of wind production in Europe, greatly reducing the need for reserve. In the specific case of Brazil, there is still an additional advantage to reinforce this conclusion. The Belo Monte hydroelectric power plant, the third largest hydroelectric plant in the world, has high seasonality and ends up allowing its production to be synergic with wind energy produced in its absolute majority in the Northeast region of Brazil once the cycles are very complementary. This phenomenon does reduce the risks of adequacity and also allows large transmission systems over 3000 km not to remain idle for much of the year. Although it is an obvious advantage, it is usually denied by environmentalists who needed in this case recognize a strategic advantage of a large hydroelectric plant located in the Amazon region, what they do not want. One can also add to these possibilities of storage in HPPs, the development of more efficient and economical storage system-batteries. Apparently, this advance would be linked to an expansion of electric mobility that would promote the scale for the reduction of battery prices. It happens that in the measure that the IRRs expand, either by reduced cost of equipment, for environmental reasons or the existence of incentives, and considering that these sources are not dispatchable, at many times the marginal costs of operation fall sharply and even negative prices can happen. The main factor influencing short-term prices (spot prices) is still determined by demand (it should be noted that demand is an exogenous factor to the generation industry). The second main factor is already represented by the IRRs generation itself and its intermittence. This phenomenon results from the excess supply of electricity with nondispatchable plants associated with low demand situations. When this phenomenon occurs, it can even be further aggravated by the low flexibility of a series of conventional plants such as thermal plants (especially those of coal) and nuclear plants that can demand several hours for their adjustment. For this reason, in the occurrence of negative prices it is usual that they last for a few hours, at least 3
4 hours. A period that has been conventionally called ‘Downward Adequacity,’ that is, the lack ability of a system to adapt to the low demand in relation to the production of not dispatchable sources. Although negative prices are very rare in the economy, they can be associated with the costs arising from the need to stock a product that has no demand in a given period and whose production cannot be interrupted. Economic history still records paradigms that can be associated with these negative costs, such as the destruction of products in order to avoid their depreciation in the market. During the crisis of the 1930s in the last century, the Vargas Government in Brazil burned coffee stocks in order to balance the price suitable for the market. Occurrences of negative prices have already been registered in Germany, Denmark, Austria, Belgium, France, United Kingdom, Switzerland, Canada,

528

PART | V Policy

Australia, United States, and in regional markets such as NordPool (Barbour et al. [30], Davis [31] and Ambec and Crampes [32]). Ambec and Crampes also reported occurrences of 200 h/MWh in June 2016 in France a very high figure for this phenomenon [32]. An effect that once was cited as mere curiosity has been increasingly frequent. According to Davis [31], this situation has increasingly surpassing 100
150 hours/year while Go¨tz et al. [33] estimate that these kinds of occurrences could surpass 1000 hours/year in 2022 in Germany. One has to ask why plants having to pay to keep their operation active prefer to operate than to speed up their shutdown. The reasons can be multiple: (1) inability to change the plant’s status quo in the period in which negative prices are expected to be maintained,9 (2) because of obligations assumed in the supply of ancillary services,10 (3) obligations assumed in contracts for heat supply in district heating systems and/or contracts linked to cogeneration services, (4) plants with long-term contracts with a preestablished price and that do not face extra costs of operating on a flat basis (because of the specific conditions of these contracts) may not have the necessary incentives to reduce their generation, (5) in the existence of tariffs ‘feed in,’ the loss is supported by some government fund or by regulated tariffs, (6) why there may be contract conditions with ‘self-dealing’ conditions where a possible loss of one part of the production chain is compensated by the profits of another part of the same shareholder and (7) for hydroelectric plants that can not to stop for safety criteria. Although this is possible far from the point of view of flexibility, it may arise from the nonexistence of spillways (small run-of-rivers power plants) or from environmental restrictions that limit the use of spillways by excessive oxygenation of the water spilled (in both cases what remains is to turbine the water from the reservoirs). It should also be noted that the effects are not restricted to microeconomic effects, but may have macroeconomic consequences. To exemplify this effect, we can refer to the effect that negative prices can have on energy exports (a kind of submarket risk11 that overpasses from one country to different others). The effect is not restricted to commercial transactions among agents of the electric industry when there is the phenomenon of negative prices. Countries with the relevance of IRRs like Germany influence the prices in the countries where the surplus could be destined [34] It is common 9. Some plants would take a long period of time to shut down. When the shutdown could be reached, probably prices would return to normal levels and demand a reverse and costly operation, not only due to the additional fuel spent on ramp-up (or ramp-down) but also due to wear and tear additional equipment. 10. This challenge will require greater ability of these conventional plants to price their ancillary service markets. 11. The submarket risk is usually the risk of congestion of transmission lines. But in its case is the flow of energy with unrealistic prices.

How much is possible? Chapter | 17

529

for negative prices in Germany to affect spot prices in France, aggravated by the fact that France being a country with a predominance of nuclear power plants, these do not have enough flexibility to reduce its production and take advantage of the possibility of importing from Germany This reduction in spot prices, induced by the penetration of IRRs, produces a rebound effect, as it creates a strong incentive for conventional plants (especially thermal plants that are not so flexible) to lose interest in developing new projects (which are essential for operation as safety reserves). Otherwise, if these are developed, investment metrics will require higher returns. As a consequence, the expansion of IRRs would find viability limits, being found in several publications, as pointed before, such as Milligan and others [35] that there would be a limit of the order of 30% as the upper limit of the percentage of insertion of these intermittent sources in relation to total installed capacity. This effect has become known in the literature on the subject as merit order effect (MOE). Formally, the definition of MOE can be found in McConnel et al. [36]. These authors explain that the generation offers are ranked by the price of available options in the auction that will define the economic dispatch. This is a procedure that the ISOs use daily aiming at meeting the projected demand in the most economical way. This set of plants thus selected meet in principle the demand projected electricity for a specified period in future, usually the next day. The last source select in the auction defines the marginal operational cost that will be pay as a rent for all generators selected by this process. Since the operational cost of the IRRs is very low, this model ends up producing a reduction in spot prices. Figs. 17.6 and 17.7 in a qualitative way evidence that when we have a IRR with a very small cost of operation the price that meet the equilibrium will reduce. In this single example, the sources of natural gas will be out of the portfolio defined by the auction and the spot price change for a less expensive value.

17.3.4 Externalities and the merit order effect Although the presence of energy from IRRs being part of the auctions and have the ability to affect spot prices, it cannot be overlooked that the important presence of these alternatives resulted from previous incentive policies (PROINFA12 or Feed-in tariffs
for example) that represent a cost to society. If, on the one hand, consumers and traders operating in competitive markets end up having advantages with the MOE, consumers with regulated tariffs in many countries have higher tariffs because of cross-subsidies. 12. Alternative Sources Incentive Program developed in Brazil.

530

PART | V Policy

FIGURE 17.6 Marginal operational cost
no IRRs. Source: Elaborated by author, adapted from Benhmad F, Percebois J. An econometric analysis of merit order effect in electricity spot price: the German case. Cahier de Recherche No. 1801120. Centre de Recherche en L´ economie ´ et droit de Lenergie; 2018. p. 1
20 [37].

This effect exemplifies well the concept of externality when a cost or benefit13 affects the welfare of society without, however, affecting the direct costs of industry. In the study developed by the National Academies of Sciences, Engineering and Medicine [38], there is a large series of references on studies developed on externalities, from the pioneering work of Pigou [39] to the contributions of Hohmeyer [40] one of the first authors to exemplify environmental problems as externalities to be considered in the electric power industry. In the same work, Cohon et al. [38] draw attention to the fact that public policies affect IRRs in an intense way in the incidence of externalities. However, these externalities are not restricted to the expansion of generation capacity. For example, distribution utilities in Brazil, located in poorer regions, tend to have higher tariffs because of higher quotas of subsidies for discounts destined to the ‘low-income’ class and higher costs of works for the universalization of electricity in rural areas. Energy markets are affected by externalities. For example, the existence of Feed in Tariffs leads to an increased supply of IRRs and consequently increases the MOE. In a recent work by Hildmann et al. [41], they identified that in many countries, the beginning of the encouraged insertion of IRRs occurred simultaneously with market release reforms and thus the price formation effects are often difficult to identify from the duality of causes and effects. These complexities must be conceptualized from different perspectives, according to the role of each agent and it is clear that all these difficulties act as barriers to the expansion of IRRs.

13. Externalities can be positive or negative.

How much is possible? Chapter | 17

531

FIGURE 17.7 Marginal operational cost
with IRRs. Source: Elaborated by author, adapted from Benhmad F, Percebois J. An econometric analysis of merit order effect in electricity spot price: the German case. Cahier de Recherche No. 1801120. Centre de Recherche en L´ economie ´ et droit de Lenergie; 2018. p. 1
20.

The question affecting producers focuses on how to maintain the viability of conventional projects in markets with high IRRs participation and consequently affected by MOE. One of the alternatives to address this problem is the development of a capacity market, separating the ‘ballast’ from the traded energy. The issue that mainly affects consumer protection class entities focuses on how to avoid that incentive policies produce a very intense increase in regulated tariffs.14 Distribution dealerships, on the other hand, need to live with the risk arising from the so-called ‘Death Spiral’.15 It is worth noting, as Foster et al. have pointed out [42], that the externalities associated with MOEs can have a transshipment effect from the energy industry to other related industries, such as the fuel industry. As these authors explain [42], as IRRs once achieve cost parities with conventional sources, so in theory they should no longer have subsidies. On this occasion, it is likely that there will be a competitive reaction of the prices of conventional plants (mostly thermal plants) due to the reduction of the spot price. This trend may not be absolute in all cases, since an eventual price reduction in the search for competition would require the reduction of fossil fuels that represent the highest operational cost of traditional plants, but these 14. It is clear that regulated consumers actually finance the difference in volume traded with incentives (Feed in tariffs, for example) and that which would occur if these incentives did not exist. 15. Expansion effect of IRRs that lead to excess capacity in distribution systems (lines, substations and transformers and human resources and infrastructure for maintenance), which subsequently lead to the need for real tariff increases and consequently become an incentive for greater IRR penetration, especially for small-scale ones, causing a self-fed effect.

532

PART | V Policy

prices are normally linked to other market logics and eventually to other regulations that influence their price formation. In the same context, Foster et al. remind that IRRs are priced down by technological advance and the learning curve of the industry, in particular wind and sunlight predictability techniques, while traditional plants have their cost very dependent on fuels. It should also be noted that fuels do not always have their prices formed through competitive processes. The International Gas Union report explains that in research conducted with producers in 2013 found that only 43% of prices were formatted by competitive processes, 19% by contracts linked to the price of oil, 33% by government decisions and 5% by other criteria [43]. Particular cases are affected by regional contingencies, for example, the influence in the MOE in several US states because of shale gas. It should also be considered that the thermo plants will continue to be necessary to function as operational safeguards16 so that at a certain level of IRRs penetration these traditional plants must have their prices adjusted or otherwise no new investments will emerge. There are a few studies of the influence of MOE in the long term, an exception is Li’s Master’s thesis which concludes that this effect tends to be minimized, probably by growth and costs associated with greater operational complexity [44]. Li also addresses the fact that the expansion of IRRs and the lower prices produced by the MOE also affect the price of carbon certificates, which intuitively should represent a percentage value of the total energy cost. Some authors believe that long-term contracts signed by traditional plants need to be respected and that at the end of these new conditions should be renegotiated reflecting the MOE. However, the existence of relevant MOE may also create conditions for claims to be made while the contracts are still in force, as a kind of compensation for a new type of ‘stranded costs.’ The MOE can make short-term gains for consumers, but, according to Pham and Lemoine [45], these gains are not enough to finance the incentives developed to leverage the IRRs. In other words, the benefit produced for some would be less than what was invested by society as a whole. The complexity of all these analyses makes the decision of what is the optimal expansion of IRRs for a given country increasingly difficult to achieve. Prof. Hirth teaches [28] that while these constraints are changing over time, the optimal penetration of IRRs will depend on the existing infrastructure, the regulatory framework in place, and the relative costs of investments in each generation alternative. He also notes that many of these costs are centralized (e.g. Treasury transfers for incentives to environmentally friendly sources) and the benefits are decentralized among market agents. 16. At least until demand response and storage alternatives are sufficiently structured and cost competitive.

How much is possible? Chapter | 17

533

These studies of the economic impacts of IRRs allow us to conclude that this expansion produces many effects, for example: (1) reduced spot market prices; (2) increased volatility in spot market prices; (3) generator income transfers to the market; (4) the existence of the MOE creates unfavourable arguments for IRR incentive policies; (5) there is no consensual evidence that the income lost by generators always reaches final consumers; (6) the MOE represents an excellent example of the risk of very long contracts based on increasingly volatile and unstable assumptions; (7) an expansion of deregulated markets will allow greater access for small consumers at prices affected by the MOE; (8) there are not many studies on the MOE in the long term,17 but the few that have been published consider the possibility of the effect being reduced and even cancelled in longer terms. This could be explained by the perspective that long-term prices reflect more the expectations of market agents about the fundamentals of the energy industry in this long-term perspective; (9) the ideal penetration of IRRs depends on endogenous conditions of the industry (installed infrastructure, equipment prices, consumer market and regulation), but also depends on exogenous factors such as availability of natural resources, substitute products and climate conditions.

17.4 Rebound effect
social acceptance of intermittent renewable sources
the opponents The author understands that IRRs will not be accepted automatically by society. This idea may be surprising to the extent that they apparently have only favourable and virtuous attributes, such as being cheap and having strong environmental appeal. This section analyses that this automatic acceptance could be not true. The acceptance must occur in three different plans: (1) socio-political acceptance where it is discussed how these policies and technologies are seen by the general public, (2) acceptance of the impacted community, which may become a relevant barrier, (3) acceptance of the market where investors and governments need to create favourable conditions economically and also it is necessary to occur the technical feasibility arising from the impacts on operation, energy security and reliability of the electricity system. In the foreground, we can mention the devaluation of land around the plants, the decrease in local tourism, the increase in tariffs (due to the existence of subsidies for the IRRs), the eventual inability of the public authorities in their licensing, media opinion, noise, health risks, visual aspects related to the ecology of the landscape and memory links, and ecological debates that counteract clean energy with the impacts of fauna and flora (green 3 green) among other aspects. 17. No studies involving MOE were found in the Brazilian market.

534

PART | V Policy

In the dimension of the impacted community, the discussion points address experiences in their relationship with the project developers, the importance of community leaders in forming opinion about their convenience and the role of the media, advocacy and logistics arrangements, such as distance from the plant to the impacted agents. The definition of metrics for these impacts (e.g. sound impacts), the size and geometric arrangement of the generation plant and the political perception of the impacts in relation to major themes are also important, for example: how compensations and mitigations were implemented, and the ideological positioning in the face of climate change. Among the findings of this research, it is worth noting that the neighbourhood effect (NIMBY18) has been identified as of little importance in the international literature [46,47]. The third dimension discusses how the conditions of governance provided by regulation and market players can foster full acceptance or resistance to these projects. One can exemplify the dimensions of regulatory safety and the impacts that the intensive penetration of these alternatives will provoke in the operation, direct costs, marginal costs of operation19 and of third parties (e.g. greater dispatch of thermal plants to cover intermittence), financing terms, impacts on the profitability of distributors,20 existence of lobbying for specific technologies such as coal or nuclear that would be displaced by the growing insertion of IRRs, among others. Specifically, denialist agents of the existence of Climate Change may want to defend more orthodox generation alternatives such as natural gas or coal on the grounds that climate issues are misused to subsidize the IRRs. All of these different factors are pointed in several papers [48
51]. The research carried out shows that while the very intense expansion of IRRs can be seen as a parameter of success, both by entrepreneurs and by institutional agents and public policy formatters, this success should not be seen as a tacit recognition that this expansion will continue without there being an emerging opposition, similar to what has already occurred in France and in other countries. This may cause delays and increases in the costs of new projects, because of the increasing difficulty of licensing and because of increased commitments to mitigate and compensate for the social and environmental impacts caused. This opposition could still arise if the expansion of the IRRs was lead to increased tariffs for electricity consumers, either because of incentive policies or because of increased operational complexity and the growing need for operational reserve plants to cope with intermittency as discussed in previous sections. 18. Not in my backyard. 19. Merit Order effect, for example. 20. Notably the so-called death spiral effect, where the reduction in network uses as a result of the growing use of distributed small generation ends up requiring an increase in tariffs and consequently increases the incentive for new users to join these systems.

How much is possible? Chapter | 17

535

17.5 Conclusions The expansion of IRRs is a reality and should remain at a sustainable rate of expansion for its favourable attributes, however, the absolutist theses of a 100% IRR-based electricity industry should not happen. The technical, political, regulatory and economic complexities are very impactful and diverse affecting multiple players. This analysis indicates that the limits taught by contemporary literature are coherent when establishing limits (qualitative ones) of the order of 30% of all installed generation capacity in the next decade. In fact, this is the estimated penetration (order of magnitude) to happen in Brazil until 2030. Some countries may go a little further, but without sufficient storage systems in place or with viable costs there will still be limits and the industry will persist for a long time to come depending on fossil fuels.

17.6 Acknowledgments This article was developed as a stage of a Research and Development Project in the Brazilian Electric industry conducted under the Programme coordinated by ANEEL
National Electric Energy Agency, entity that exercises regulatory power in Brazil. The aforementioned project is called IRIS Integration of Intermittent Renewables: a model for simulating the operation of the Brazilian electrical system, to support planning, operation, commercialization and regulation (ANEEL Code PD-0610-1004/2015). The author manifests his acknowledgments to other professional working in the same project. The finance entities deserve also our recognition of their importance. They are AES Uruguaiana Empreendimentos S/A, Barra do Brau´na Energe´tica S/A, Campos Novos Energia S/A, Companhia Energe´tica Rio das Antas, Energe´tica Barra Grande S/A, Foz do Chapeco Energia S/A and Itiquira Energe´tica S/A.

References [1] EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia-2029. ,http:// www.epe.gov.br.; 2020 [accessed 20.03.20]. [2] PSR. An´alise estrutural do suprimento. Energy Rep. 2020;157:15
16. [3] EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia-2015. ,http:// www.epe.gov.br.; 2005 [accessed 20.03.20]. [4] EPE-Empresa de Pesquisa Energe´tica. Plano decenal de expansa˜o de energia-2027. ,http:// www.epe.gov.br.; 2017 [accessed 20.03.20]. [5] ANEEL-Agencia Nacional de Energia Ele´trica. Resoluc¸o˜es tarifarias-diversos anos, 2001
2016. ,http://aneel.gov.br/biblioteca/resolucoeshomologatorias.; 2020 [accessed 20.12.17]. [6] IRENA-International Renewable Energy Agency. A new world
the geopolitics of energy transformation. ISBN 978-92-9260-097-6; 2019.

536

PART | V Policy

[7] IEA-International Energy Agency. World energy outlook. ,http://www.iea.org.; 2019 [accessed 20.01.20]. [8] GWEC-Global Wind Energy Council. Global wind statistics. ,http://gwec.net/publications/global-wind-report-2/.; 2018 [accessed 14.09.19]. [9] ABEEolica-Associac¸a˜o Brasileira de Energia Eo´lica. Boletim Anual de Gerac¸a˜o Eo´lica. ,http://abeeolica.org.br.; 2018 [accessed 27.09.19]. [10] UNEP-United nation Environment Programme. REN
Renewables Global Status Report. ,https://wedocs.unep.org/bitstream/handle/20.500.11822/28496/REN2019.pdf? sequence 5 1&isAllowed 5 y.; 2019 [accessed 14.01.20]. [11] Solar Power Europe. Global market outlook for solar power/2019-2023 ,http://www. solarpowereurope.org/wp-content/uploads/2019/05/SolarPower-Europe-Global-MarketOutlook-2019-2023.pdf. [accessed 14.09.19]. [12] Cohn JA. The grid: biography of an American technology. MIT Press; 2017. p. 336. [13] Bakke G. The grid: the fraying wires between American and our energy future. New York, NY: Bloomsbury; 2016. p. 364. [14] Power-Technology. The 10 worst blackouts of the last 50 years. ,https://www.powertechnology.com/features/featurethe-10-worst-blackouts-in-the-last-50-years-4486990/.; 2015 [accessed 09.02.20]. [15] Dame´ L. Dilma reage ao ONS. Jornal O Globo, publicado em 6 de fevereiro de. ,https:// oglobo.globo.com/economia/apos-ons-admitir-descarga-eletrica-como-causa-de-apagaodilma-reage-cobra-fiscalizacao-11527891.; 2014 [accessed 09.02.20]. [16] Hunt S. Making competition work in electricity. John Wiley & Sons; 2002. p. 450. [17] McLean B, Elkind P. Reprint The smartest guys in the room: the amazing rise and scandalous fall of Enron. Penguin Group; 2013. p. 480. [18] Munson R. From Edison to Enron: the business of power and what it means for the future of electricity. Northeast-Midwest Institute; 2005. p. 206. [19] Ackermann T, Carlini EM, Ernst B, Groome F, Orths A, Sullivan J, et al. Integrating variable renewables in Europe. IEEE Power Energy 2015;13:67
77. [20] CAISO-California ISO. What the duck curve tell us about managing a green grid. ,http:// www.caiso.com/Documents/FlexibleResourcesHelpRenewables_FastFacts.pdf. [accessed 09.02.20]. [21] Madrigal M, Porter K. Operating and planning electricity grids with variable renewable generation. World Bank; 2013. p. 125. [22] Asmus P. How California hopes to manage the intermittency of wind power? Electr. J. 2003;16(6):48
53. [23] Murray B. The paradox of declining renewable costs and the rising of electricity prices. ,https://www.forbes.com/sites/brianmurray1/2019/06/17/the-paradox-of-declining-renewable-costs-and-rising-electricity-prices/#4c27f16461d5. [accessed 20.03.20]. [24] Sioshansi F, editor. Innovation and disruption at the grid0 s edge. Academic Press; 2017. [25] Sioshansi F, editor. Future of utilities
utilities of future: how technological innovations in distributed energy resources will reshape the electricity power sector. Academic Press; 2016. [26] Trainer T. Can Australian run on renewable energy? The negative case. Energy Policy 2012;50:306
14. [27] Mills A, Wiser R. Changes in the economic value of variable generation at high penetration levels: a pilot case study of California. Ernest Orlando Lawrence Berkeley National Laboratory; 2012. p. 1
114.

How much is possible? Chapter | 17

537

[28] Hirth L. The optimal share of variable renewable. How the variability of wind and solar affects their welfare-maximum deployment. Energy J. 2015;36(1):127
62. [29] PSR. Ok Renov´avel voceˆ venceu. Energy Rep. 2018;143:15
16. [30] Barbour E, Grant W, Hall P, Radcliffe J. Can negative prices encourage inefficient electrical energy storage devices? Int. J. Environ. Stud. 2014;71(6):862
76. [31] Davis, L. Is solar really the reason for negative electricity prices? Energy Institute at Haas. ,https://energyathaas.wordpress.com/2017/08/28/is-solar-really-the-reason-for-negativeelectricity-prices/. [accessed 20.03.20]. [32] Ambec S, Crampes C. Negative prices for electricity. In: Working Paper, Florence School of Regulation; 2017. [33] Go¨tz P, Henkel J, Lenck T, Lenz K. Negative electricity prices: causes and effects. In: Working Paper Energiewende; 2014. [34] Morris C. Negative power prices: good or bad? In: Energy transition. The global Energiewende. ,https://energytransition.org/2018/02/negative-power-prices-good-or-bad/.; 2018 [accessed 09.02.20]. [35] Milligan M, Lew D, Corbus D, Piwko R, Miller N, Clark K, et al. Large-scale wind integration studies in the United States: preliminary results. In: Conference Paper NREL/CP550-46527; 2009. [36] McConnel D, Hearps P, Eales D, Sandiford M, Dunn R, Wright M, et al. Retrospective modeling of the merit order on wholesale electricity prices from distributed photovoltaic generation in the Australian National Electricity Model. Energy Policy 2013;58:17
27. [37] Benhmad F, Percebois J. An econometric analysis of merit order effect in electricity spot price: the German case. Cahier de Recherche No. 1801120. Centre de Recherche en L economie et droit de lenergie; 2018. p. 1
20. [38] Cohon J, et al. Hidden costs of energy: unpriced consequences of energy production and use. Academies of Sciences, Engineering and Medicine Press; 2010. p. 507. [39] Pigou AC. The economics of welfare. London: Macmillan; 1920. p. 876. [40] Hohmeyer O. Social costs of energy consumption: external effects of electricity generation in the Federal Republic of Germany. New York, NY: Springer; 1988. p. 126. [41] Hildmann M, Ulbig A, Andersson G. Revisiting the merit-order effect of renewable energy sources. 2015 IEEE Power & Energy Society General Meeting, Denver, CO. IEEE; 2015. p. 1. [42] Foster E, Contestabile M, Blazquez J, Manzano B, Workman M, Shah N. The unstudied barriers to widespread renewable energy deployment: fossil fuel price responses. Energy Policy 2017;1
7. [43] International Gas Union. A global review of price formation mechanism. Wholesale gas price survey-2014 edition. ,https://www.igu.org/sites/default/files/node-page-field_file/ IGU%20Wholesale%20Gas%20Price%20Survey%20Report%20-%202014%20Edition. pdf. [accessed 05.02.19]. [44] Li X. The impact of wind power generation on the wholesale electricity price- evidence from Swedish electricity market [Master thesis]. Umea University; 2017, 27 p. [45] Pham T, Lemoine K. Impacts of subsidized renewable electricity generation on spot markets prices in Germany: evidence from GARCH model with panel data. In: Chair European Electricity markets
CEEM Working Paper; 2015. p. 15. [46] Eltham DC, Harrison GP, Allen SJ. Change in public attitudes towards a Cornish wind farm: implication for planning. Energy Policy 2008;36:23
33.

538

PART | V Policy

[47] Rand J, Hoen B. Thirty years of North American wind energy acceptance research: what have we learned? Energy Analysis and Environmental Impacts Division, Lawrence Berkeley National Laboratory; 2017. [48] Bell D, Gray T, Haggett C, Swaffield J. Re-visiting the ‘social gap’: public opinion and relations of power in the local politics of wind energy. Environ. Politics 2013;22 (1):115
35. [49] Enevoldsen P, Sovacool BK. Examining the social acceptance of wind energy: practical guidelines for onshore wind project development in France. Renew. Sustain. Energy Rev. 2016;53:178
84. [50] Ellis G, Ferraro G. The social acceptance of wind energy: where we stand and the path ahead. Luxembourg: Publications Office of the European Union; 2016. Joint Research Centre (JRC), European Commission. [51] Carneiro FOM, Rocha HHB, Rocha PAC. Investigation of possible societal risk associated with wind power generation systems. Renew. Sustain. Energy Rev. 2013;19:30
6.

Chapter 18

Renewable energy technologies: barriers and policy implications Jyoti Prasad Painuly1 and Norbert Wohlgemuth2 1 2

UNEP DTU Partnership, Technical University of Denmark, Copenhagen, Denmark, Department of Economics, University of Klagenfurt, Klagenfurt, Austria

Chapter Outline 18.1 Introduction 539 18.2 Literature on barriers to renewable energy 541 18.3 Barriers identification and policy frameworks 544 18.3.1 Economic barriers 545 18.3.2 Technical barriers 546 18.3.3 Awareness and information barriers 547 18.3.4 Financial barriers 547 18.3.5 Regulatory and policy barriers 548 18.3.6 Institutional and administrative barriers 548 18.3.7 Social and environmental barriers 548

18.3.8 End-use/demand-side barriers 549 18.4 Barriers identification framework 550 18.4.1 Selection of renewable energy technologies for the study of barriers 553 18.4.2 Identification of barriers for the study 553 18.5 Measures to overcome barriers 554 18.5.1 Renewable energy targets 554 18.5.2 Renewable energy promotion measures 555 18.5.3 Net metering/net billing 557 18.6 Current challenges 558 References 560

18.1 Introduction Total renewable energy power capacity reached 2378 GW in 2018 (including 1246 GW hydropower), registering a growth of 8% in 2018 (15% excluding hydropower) [1], indicating countries’ interest and commitment to increased use of renewables to combat climate change. Renewable power growth was led by solar PV, wind and hydro with capacity addition at 100, 51 and 20 GW, respectively, demonstrating the maturity of these technologies in many countries. 30 countries had more than 10 GW of renewable power Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00018-2 © 2021 Elsevier Inc. All rights reserved.

539

540

PART | V Policy

while 90 countries had more than 1 GW indicating efforts by a large number of countries to promote renewable power [1]. The addition to renewable exceeded 50% of total power capacity addition in 2012, and in 2018 renewable power addition stood at 64% of new generation capacity
almost double of all other sources combined. $273 billion were invested in renewable capacity addition in 2018, with the decade 2010
19 investment reaching $2.6 trillion [2]. The cost competitiveness of renewable power also improved significantly with the levelized cost of electricity (LCOE) for solar photovoltaics (PV) and onshore wind, two key sources of renewable power, registering a decrease of 81% and 46%, respectively, during the decade. According to International Energy Agency statistics, between 1990 and 2017, total worldwide energy demand increased at an average annual rate of 1.7%, hardly outpaced by renewables which increased by 2% annually on average. Solar PV and wind energy were the fastest growing forms of renewable energy, with average annual growth rates of 37.0% and 23.4%, respectively (Fig. 18.1). The use of solid biofuels and charcoal increased by a mere 1.0%, dragging
as a result of their weight within the renewables category
the annual average growth rate of all renewables to 2.0%. With these impressive growth rates, can it still be said there are barriers to the growth of renewables? First, the share of renewable energy in total primary energy supply is still as low as 10.2% in the OECD (in 2017). This share ranges from 0.4% in the Middle East to 9.0% in China to 47.3% in Africa. Second, there is a huge potential for the development of renewables and increasing its share in the total primary energy supply. Third, even the

FIGURE 18.1 Average annual growth rates of world renewables supply, 1990
2017. IEA. Renewables information 2019. Paris: International Energy Agency; 2019a [3].

Renewable energy technologies Chapter | 18

541

current pace of renewable development owes to support from various quarters including governments and donors absence of which may throttle its development, and finally, the growth is uneven, with a few countries primarily contributing to renewable energy development and potential remains untapped in others due to a variety of barriers. Though barriers exist, these may vary in their intensity (importance) requiring major efforts where intensity is high, for example high subsidies to fossil fuels can make economic viability a major barrier, and coverage, for example some barriers are specific to a technology or a country/region. In addition, barriers have interdependencies and addressing the key barriers can help others getting addressed easily. For example once an economic barrier has been addressed, the financing barrier could also vanish in some cases, as the risk to financers is reduced. This calls for a new and differentiated approach across countries to address the barriers.

18.2 Literature on barriers to renewable energy A number of studies have gone into the issue of barriers renewables have to overcome to penetrate the market, including for example [1,4,5
12]. This section provides a short literature overview of recent publications on specific barriers. The European Union has an ambitious ‘Green Deal’ that requires almost complete decarbonization of the Union’s energy system by 2050. Shivakumar et al. [13] provide a meta-analysis of projections for the shares of renewables in selected EU states. They find that ‘a favourable investment climate also requires the absence of severe barriers in the noneconomic environment’. Their paper provides a detailed overview of the main barriers and enablers for individual EU countries. Not surprisingly, countries face similar barriers and can achieve deployment through common enablers. Consequently there is significant potential to learn from each other. Hindering barriers include primarily an unsustainable ratio of effectiveness and efficiency, access to finance and difficult spatial planning processes. Hu et al. [14] present a comprehensive literature review of barriers to large-scale market integration of intermittent renewables in the EU’s electricity market. Sen and Ganguly [15] provide an overview of public policy initiatives to promote renewable energy. They stress five pivotal actions for a sustainable energy future: strengthening policy commitment to renewable energy; mobilizing investment; institutional, technical and human capacity building; harnessing cross-cutting impacts of renewable energy on other sustainable development goals and enhancing regional engagement and international cooperation. Government policies that complement each other (e.g. in the areas of land use, transportation, employment, agricultural, food security, water, trade concerns, infrastructure) are more likely to be effective. Sen and Ganguly conclude that a shift towards renewables might begin with a

542

PART | V Policy

prominent role for energy efficiency, the ‘hidden fuel’. Conserving energy by increasing efficiency in its use increases, ceteris paribus, the share of renewables [16]. The integration of (low cost) renewables in the electricity system causes organizational problems in many cases, for example concerning utility resource planning methods and processes that ensure the least cost service provision. Go et al. [17] conclude that, to overcome these barriers, regulators and utilities ‘will need to expand and upgrade their analytical capabilities’. In regions with little growth in electricity demand, such as Europe, increasing generation by renewables depresses wholesale electricity prices, making renewables somehow victims of their own success, indicating the limits of marginal cost pricing [18]. Wholesale electricity prices slumped from about h80/MWh in 2008 to h30
40. In an industry where marginal costs are of great importance to the overall economics (merit order), the economics of renewables affects electricity markets to an extreme extent, leading electric utilities to separate their renewables (and grid) businesses from (loss-making) conventional generation. Blazquez et al. [19] claim that the world is caught in a vicious circle: renewable energy subsidies increase their deployment, which depresses prices, thereby further increasing the need for financial support. In the extreme case of 100% (nondispatchable) renewables, the marginal cost-driven market price would fall to zero, deterring any investment that is not fully subsidized. Therefore the more successful policies to support renewables are, the more expensive and less effective policies become. The utility model of generating electricity is in many cases broken, as are markets. The ‘zero marginal cost society’ (Jeremy Rifkin) may not be as easy to achieve as commonly thought but lessons can be learned from the communications industry [20]. Foster et al. [21] analyse the effect of higher penetration of renewables on the price of fossil fuels. Usually fossil fuel prices are assumed to be exogenous, however, a massive increase in the use of renewables is likely to depress fossil fuel prices, keeping all other factors constant. Lower prices for fossil fuels, as a result of the increasing use of renewables, are, therefore, likely to inhibit the economic competitiveness of renewables. They fall victim to their own success. The Green Paradox and Carbon Leakage are examples of this development. On the other hand, the crude oil price crash of March 2020 also makes renewables less cost-competitive as a result of changing relative prices between fossil fuels and renewables (‘Oil shock threatens to take wind out of sails for renewable energy shift’, Financial Times of 11 March 20201). The economics of individual fuels is to a large extent affected by market failure
by the lack of correct pricing. Owen [22] shows that internalization of negative environmental spillovers of fossil fuels ‘would serve to hasten

1. https://www.ft.com/content/24e9abd0-62ef-11ea-b3f3-fe4680ea68b5

Renewable energy technologies Chapter | 18

543

FIGURE 18.2 Global reported corporate R&D spending in energy-related sectors, 2012
18. IEA. World energy investment 2019. Paris: International Energy Agency; 2019c [23].

this transition process’ towards renewables. This internalization could take place, for example by fuel and carbon taxes and provide an important contribution to the much-cited levelling of the playing field between different types of energy. A suboptimal level of R&D expenditure can also be considered a case of market failure. Composition of public energy-related R&D expenditure changed substantially over time. In 1974 most of public energy-related R&D in OECD countries had been spent on nuclear energy research. Had this amount been spent on renewables, they would probably already have achieved full cost competitiveness. Concerning the private sector, Fig. 18.2 shows worldwide corporate R&D spending in energy-related sectors between 2012 and 2018. In contrast to public-funded energy R&D activities, private R&D activities account only for a modest share in total R&D expenditure. A literature review reveals a general consensus that technical and economic barriers are no longer the principal barriers preventing some renewables from achieving a greater role in energy systems in many countries. Seetharaman et al. [24], for example by collecting data through an online questionnaire sent to energy professionals, find that social, technological and regulatory barriers strongly affect the deployment of renewable energy all over the globe, while economic barriers strongly influence it indirectly. For a renewable energy policy to be successful, Blazquez et al. [25] conclude that policymakers have to prioritize their objectives: large deployment of renewables, speed of adoption and cost of the policy. All three objectives cannot be achieved by a single policy. Diesendorf and Elliston [26] argue in favour of the possibility of large-scale electricity systems based on 100% predominantly intermittent renewables that meet key requirements of reliability, affordability and energy security. In addition, such an energy transition can be implemented much faster than historical energy transitions.

544

PART | V Policy

In the early years, most of the barriers were common across countries and many, even across various types of renewables. The list of the barriers included awareness, economic and financial, technical, political and regulatory, institutional, capacity, market, and social and cultural. However, over a period of time, many of these barriers have been addressed for some renewables, particularly in developed countries and emerging economies, which the scale of investment in the renewables reflects. In many developing countries, most or all of these barriers still remain, but the nature of barriers in countries that are leaders in renewables has changed. Technology-specific barriers have emerged, that impede large-scale deployment. Increased use of bioenergy requires a sustainable framework that also considers environmental issues. For geothermal, economic viability and sustainability of the enhanced geothermal systems on a large scale is a challenge [7]. Ecological and social impacts in the case of new hydropower projects can have ecological and social issues while ocean energy development may require testing infrastructure, enabling policies and regulations. Environmental concerns and public acceptance in the case of wind energy can be an issue. In many countries, issues include complicated licensing procedures and difficulty with land acquisition and permissions.

18.3 Barriers identification and policy frameworks Despite significant progress, renewable potential is yet to be fully realized. The progress in developing countries picked up with a sudden spurt in capacity addition in China and India in 2006 and onwards. Though barriers still hamper the realization of potential both in developed and developing countries, the nature of barriers and their impacts vary. With the development in the deployment of renewable energy, a variety of technologies have been used that are at different stages of development and maturity. These have been developed in response to various challenges including the cost of technologies, efficiency, availability of resources and their characteristics. In case of solar energy, three primary technologies include PV, which converts solar light to electricity, concentrating solar power, which uses heat from solar radiation (thermal energy) primarily used to drive electric turbines, and solar heating and cooling systems, which use solar (thermal) energy to provide hot water, air heating and cooling. Similarly wind technologies include onshore and offshore, each with their own variations. Some barriers to solar and wind energy are technology-specific while others are common across technologies. The most common application from renewable is to generate power but renewables also lag in for application to end uses such as heating and cooling, and transport on account of a lack of regulatory measures [6]. At the same time, new challenges have emerged due to intermittent nature of renewable
at the current status of electricity grids and available technologies, most countries may find it difficult to absorb renewable power beyond

Renewable energy technologies Chapter | 18

545

a threshold, which depends on several factors including their grid power composition and available system flexibility. As the literature on barrier indicates, a variety of barriers to penetration of renewable energy technologies (RETs) have been identified, and several types of classifications have been made in the literature with some variations across the studies. The upfront cost of renewable technology, for example is categorized under economic barriers in a study while it is a part of the financial barrier in the other. Similarly a technical barrier includes the infrastructure barrier in some studies. The following barriers can be considered as key barriers to renewable energy based on the literature on barriers.

18.3.1 Economic barriers Primarily refer to the high cost of technology compared with other competing technologies; for example renewable power competing with coal-based power. The cost of renewable power has declined over the past years and with support through a variety of policies, it has become competitive in several countries. Fig. 18.3 shows the LCOE of distributed PV systems in selected countries, including their expected cost in 2024. Further cost digressions can be expected over the next years. There are a variety of reasons for renewable technologies yet to be economically viable in many countries. Market failure is a prominent reason, which refers to subsidies to competing fuels like fossil fuels and failure to

FIGURE 18.3 LCOE for distributed PV systems, and variable residential, small commercial and industrial retail electricity prices, 2018/19. LCOE, Levelized cost of electricity; PV, photovoltaics. IEA. Renewables 2019. Paris: International Energy Agency; 2019b.

546

PART | V Policy

FIGURE 18.4 Share of variable renewable energy
top 10 countries in 2018. REN21. Renewables 2018. Global status report. Renewable energy policy network for the 21st century. Paris; 2018. ,http://www.ren21.net/wp-content/uploads/2018/06/17-8652_GSR2018_FullReport_ web_final_.pdf. [5].

account for external costs such as pollution and health damage costs of fossil fuels. Subsidized electricity tariffs, the market power of the fossil fuel lobbies, high import duties on renewable technologies are other reasons that go against renewable power.

18.3.2 Technical barriers Also referred to as infrastructure and capacity barriers, they refer to a lack of knowledge about renewable technology and skill to maintain it. Capacity barriers
such as an inability to properly carry out operation and maintenance activities
can cause renewable systems to fail after implementation [6]. Access to the grid and capacity of the grid to carry renewable power are also constraints in some countries. With the increased generation of power from renewable energy (see Fig. 18.4), a lack of infrastructure to absorb the variable power in the system has emerged in many countries (see Box 18.1). Similarly renewable application to end uses such as district heating and cooling, where the technology has large potential to contribute, faces infrastructure barriers. In the case of transport, biofuels require re-designing the engines for optimum performance.

BOX 18.1 Energy systems integration Energy from wind and solar is referred to as variable renewable energy (VRE) due to their intermittent nature of availability, leading to challenges to integrate it into the existing energy systems. Though a type of technical barrier, it is one of the fast-emerging barriers, which is already causing a serious problem in upscaling RETs to supply power to grid
for example in India, where some state (Continued )

Renewable energy technologies Chapter | 18

547

BOX 18.1 (Continued) utilities have refused to accept VRE beyond a limit due to its intermittent nature. With the increasing share of VRE in the system, the problem gets compounded in cases where the electricity grid is primarily supplied from coal and nuclear, since backing them down creates a problem for the system. At part loads, efficiency is low, and shutting down and re-starting a unit can take several hours, if not days. In the case of grids that have hydro and gas-based electricity, integration is relatively manageable. For example Denmark is able to dispatch its surplus production from wind to the Nordic grid, which has a good proportion of hydro (from Norway). Depending on the mix of power plants in a grid, or its access to flexible grids (with hydro and gas plants), the issue of integration puts a limit on the percentage of renewable energy a grid can absorb. This area is at an early stage of development for application to VRE management with work in progress on electrical storage systems, load management, and load shifting, smart grids to manage supply and demand better, pump storage, new technologies and uses such as heat pumps and electric vehicles (EVs), better forecasting techniques for wind power generation, etc. [27].

18.3.3 Awareness and information barriers These include a lack of proper information about technology, its potential and benefits. In many cases, the data on available resources is weak and a lack of capacity to understand the technology and its economics may lead to the wrong perception about it. RETs have seen significant technological advancements, but this progress is not always reflected in public perceptions. The falling cost of wind and solar technologies has brought down their costs substantially, but users may not be aware of that. In the public debate, either information on current costs is absent or outdated numbers are cited [4].

18.3.4 Financial barriers RETs have yet to be a standard product for financial institutions to fund in many countries. The risk perception is high and therefore access to finance is an issue and even where it is available, it is at a higher cost. Project finance is not available for RET projects and risk mitigation instruments such as guarantees also not easily available. Most RETs also face high upfront costs, which in turn adds to the financial risk and thereby to the cost of finance. Besides access issues, the high cost of finance can also make the project economically unviable in some cases. In the case of utility-scale power projects, grid strengthening may require substantial investment, which distribution companies may find difficult to raise, particularly in developing countries.

548

PART | V Policy

18.3.5 Regulatory and policy barriers Since RETs are still in the early stage of development in many countries and facing many barriers, they require policies and regulations favouring their use in various sectors/for various end uses. Appropriate enabling policies for RETs can help create stable and predictable investment environments; through feed-in tariff for example which can help overcome financial and economic barriers by ensuring predictable revenue streams. Similarly the expansion of renewable power may need regulations to absorb the same in the system. Renewable usage in new areas
for example in transport, space heating and cooling may also require initial help through incentive policies/ regulations. New technologies such as battery storage may also need to be supported to help RETs. A lack of enabling policies/regulation or bad policy design, policy uncertainty, ineffective implementation of regulations, etc. create barriers to the development of RETs. Some locations, good to set up RET projects from a resource perspective
availability of good wind speeds, for example may get ruled out due to zoning regulations.

18.3.6 Institutional and administrative barriers Institutional barrier refers to a lack of appropriate institutions to facilitate and take various measures related to the promotion of RETs. A weak institutional capacity that is unable to support RETs, monitor and enforce regulations can also be a barrier. On the administrative side, a lack of clearly defined responsibilities, complicated licensing procedures, difficulty with land acquisition and permission, inadequate planning guidelines, bureaucratic licensing procedure, etc. are barriers to RETs adoption [6].

18.3.7 Social and environmental barriers A lack of public acceptance can delay and increase the cost of the RET projects and sometimes even lead to cancellation. Modern wind turbines are highly visible structures, usually located on open and elevated sites, making a visual impact on the landscape. Even offshore wind parks can face strong public opposition. Similarly if not planned properly, solar farms’ visual impact can also be an issue. Wind turbines also emit noise during operation, which can be irritating for the nearby properties. Socio-cultural reasons can also be strong barriers in some cases
acceptance of improved cookstoves, biogas for cooking for example. There is resistance also to shift to a new technology where current practice is to be abandoned and new technology requires learning. RETs may also face environmental barriers, regulations may rule out siting projects at favourable locations. The project may have to satisfy site-specific environmental requirements, which may increase its cost.

Renewable energy technologies Chapter | 18

549

18.3.8 End-use/demand-side barriers Two major end-use areas, where there is huge potential to use renewable energy, viz., renewable-based heating and cooling, and transport, have a very low level of penetration of RETs. These end uses are emerging areas for use of RETs and offer the potential for upscaling RETs but face several barriers, both from demand and supply side. Though with economic growth, particularly in emerging economies, demand for space heating and cooling is expected to go up exponentially, a lack of district heating and cooling infrastructure is a barrier to the use of RETs in this area. Following Ref. [28], typical supply-side barriers for renewable heating and cooling include infrastructure development, the high capital cost of infrastructure, technical suitability, availability of skilled manpower and regulatory issues. On the demand side, the barriers include initial retrofitting costs that may face typical ‘investor-user dilemma’ (owner of the building may not have any incentive as benefit goes to the tenant), long payback period, tariff levels (low tariffs may discourage), a lack of awareness of the technology, etc. The transport sector is one of the major sectors where there is a big potential for the substitution of fossil fuels by climate friendly biofuels directly and through the use of renewable electricity (to enhance electric mobility) indirectly. A relatively new area for application, the transport sector faces a variety of barriers from economic, finance and technology to policy, institutional capacity and consumer preference/acceptance (see Box 18.2).

BOX 18.2 Barrier to renewable energy in the transport sector There are several barriers to use of renewable energy in transport sector including technology, policies, and behavioural change. In case of biofuels, the technological challenges for example include deployment of vehicles capable of running on biofuels, development of infrastructure to dispense bio-fuel, removing policy distortions such as subsidy to fossil fuels (if any), supportive policies such as mandatory blending of fuels with biofuels, etc. EVs seen as one of the major solutions to reduce GHG emissions in transport sector, is a relatively new technology and faces economic, financial, technological and social barriers. High upfront cost of the vehicles, high cost of batteries, limited driving range, low speed and low carrying capacity have been found to be major barriers in the literature on barriers to EVs [29]. A lack of awareness and a lack of charging infrastructure were found as major barriers by Dhar and Cherla [30]. These barriers restrict mobility of EV users leading to a lack of acceptance. Therefore there is a need for integrated policies, including pollution standards that motivate shift to electric transport in urban areas.

It can be seen that several barriers discussed above are noneconomic in nature. Solar energy, which has witnessed a large expansion in the last

550

PART | V Policy

decade with falling costs making it competitive in many countries, still faces a variety of noneconomic barriers. Some of these barriers along with options to address them have been described, for example in Ref. [31]. Some other barriers such as political
leading to inappropriate market intervention, remote and difficult location in case of geothermal and hydropower, etc. have also been mentioned in the barrier literature but do not seem widely prevalent and hence not included here. Similarly as already mentioned, a suboptimal level of R&D expenditure can lead to market failure due to its impact on the economic barrier. A summary of the barriers indicating their dimensions and impact on the RET project is given in Table 18.1. Several types of barriers and categorization can be found in literature beyond this general categorization. In addition, the presence and intensity of barriers may vary across countries. However, in general, though many early barriers had been addressed in most developed countries through governmental policy support, new barriers such as VRE absorption in the grid, end-use barriers in some sectors and similar have emerged with expansion of RETs, requiring further governmental support. Financial barriers also remain in some cases due to upfront invest requirement and environmental issues may also come up when installation increase. Emerging economies have also been at the forefront of promoting renewable energy with China and India among countries with large-scale installations. Some of the emerging economies still face many of these barriers including economic, technical, financial, etc., and may require international support besides national efforts to achieve their renewable energy ambition. Finally most of the other developing countries still face all these barriers and are heavily dependent on international support to address the same.

18.4 Barriers identification framework A variety of barriers have been discussed in the literature and several frameworks have been used to identify barriers and measures to address them. Painuly [32] categorizes barriers and suggests identification of relevant RETs for a country, and site visits and stakeholders’ surveys as important components of barriers identification methodology. Reddy and Painuly [12] use stakeholder surveys and use nonparametric analysis to identify barriers as well as measures to overcome them, while Nguyen et al. [33] use the analytical hierarchy process to analyse data from 37 stakeholders to identify barriers. Richards et al. [34] use a multi-dimensional approach to analyse barriers and identify the most significant barriers with a special focus on wind energy. They use AKTESP theoretical framework, which looks at environmental problems and their potential solutions in the agreement, knowledge, technology, economic, social or political factors. Seetharaman et al. [24] group various barriers and thereafter from a survey of 223 experts

Renewable energy technologies Chapter | 18

551

TABLE 18.1 Barriers and their dimensions.

Economic barriers

Technical and infrastructure barriers

Dimensions

Remarks

Cost of technology is high (investment, and/or cost of power in case of power production)

Typically when technology is in the initial stage of development, ocean thermal for example

Market failure
competing technologies are subsidized or their costs not internalized

Subsidy to fossil fuels, external costs of fossil fuels not considered

High import duties/taxes

Sometimes levied to promote local industries

Technology availability

Technology patents, especially for new technologies

A lack of knowledge about technology and skill to maintain

Need for training

Access to the grid (for renewable power)

Due to distance, it can be expensive

Grid capacity (to carry renewable power)

Strengthening of the grid may be needed

System capability to absorb VRE

The system may not be flexible

A lack of skilled personnel and training facilities Infrastructure barriers for some end-use Awareness and information barriers

A lack of information about technology, its economics and its features

A lack of appropriate institution to promote RE

A lack of information about benefits compared to the existing technology

Need for capacity building

A lack of capacity to understand the technology and its benefits Financial barriers

High-risk perception of RET projects

This leads to high financing cost and also makes access to finance difficult

Difficult to get project finance

This is due to a lack of familiarity with RETs with mainstream finance institutions as well as the high-risk perception of RET projects (Continued )

552

PART | V Policy

TABLE 18.1 (Continued)

Regulatory and policy barriers

Institutional and administrative barriers

Social and environmental

End-use barriers

Dimensions

Remarks

High upfront cost

Increases financial risk and cost of finance; can make project unviable

May require additional investment in the system (grid strengthening) by distribution companies

Distribution company may not be interested or financially sound to invest

Absence of enabling policies/ regulations

RETs may need regulations such as feed-in tariffs, quotas, etc. to compete with a traditional technology (fossil fuels) since external costs not charged to the latter

Uncertain policies

Policy uncertainty can lead to a lack of confidence and consequently affect investment in RE

A lack of institutional capacity to promote RETs

Leads to a lack of awareness about REs, their true costs and benefits

Inadequate support
for example for land acquisition

Can increase the cost of RE

A lack of involvement of stakeholders in policymaking

This can lead to misplaced priorities and policies that do not work

Complicated and bureaucratic licensing and other procedure

These lead to delays and increased cost

A lack of public acceptance due to social reasons

It can lead to increased cost and smaller market

Environmental barriers

Noise pollution, zoning regulations for example

Aesthetic considerations

Visibility issues leading to public opposition in case of wind turbines for example

A lack of infrastructure and high cost of its development for renewable energy heating, charging stations for electric vehicles for use of renewable electricity

These are usually technology and end-use specific barriers

RETs, Renewable energy technologies; VRE, variable renewable energy.

Renewable energy technologies Chapter | 18

553

analyse the factors affecting the deployment of renewable energy, and the impact of breaking the barriers on deployment. They also test the applicability of Rogers’ theory of diffusion to the barriers to renewable energy. Using a case study approach, Dio´genesa et al. [35] interview 41 stakeholders to identify barriers and removal measures. Countries/experts can select any methodology or develop new methodologies to capture the changing dynamics of barriers and design measures to remove the barriers. However, meaningful consultation with stakeholders should be a key component of any methodology to ensure that stakeholders’ concerns are addressed, leading to acceptance and successful implementation of policies and measures. In many developing countries, before barrier analysis, there may be a need to identify RETs that the country should develop, Therefore in the section below, building on Ref. [32], a simple approach has been enunciated starting with the identification of the RETs, identification of the barriers and measures to overcome the identified barriers.

18.4.1 Selection of renewable energy technologies for the study of barriers The objective of this step is to identify the RETs that can contribute substantially to a country’s target for renewable energy development. Following criteria are suggested for identification of RETs for barrier removal. G

G

G

Adequate potential: Unlike fossil fuels that can be imported, most of the RETs development in the countries centres around local availability of the RET resource; availability of solar energy, wind energy, geothermal, rivers (for hydro energy), land availability, etc. In the case of biofuels, availability of land and climatic conditions to grow biofuels could also be important, though import is also an option. In some cases where regional cooperation is feasible, resource availability could be considered at the regional level. Availability of technologies and their costs: Despite the potential, some technologies may not be suitable for a country depending on other factors like the maturity of a technology which in turn gets reflected in its economic viability and availability of skilled personnel; offshore wind for example. Other factors: Such as the availability of infrastructure, financing, socioeconomic impacts, environmental impacts, etc.

18.4.2 Identification of barriers for the study Proper identification of barriers is important to ensure that appropriate policies and support actions are designed to promote RETs. A lot of resources are spent on designing and implementing policies and actions and any errors

554

PART | V Policy

in the identification of barriers may not fulfil the objectives for which these were designed. Barriers identification can be done using a combination of the literature survey with particular focus on local studies and on countries/regions similarly placed, site visits of RETs projects, wherever feasible and wider stakeholders’ consultations. The stakeholders, for example could include equipment manufacturers, project developers, regulators and policymakers, consumers, utilities, financial institutions, experts (including consultants and academic/research institutions) and other relevant NGOs. The interaction can be through structured interviews and/or questionnaires and it may involve qualitative as well as quantitative assessment based on methodology selected; ranking of barriers or scoring the barriers on a given scale for example. The stakeholders’ consultation is very crucial to the identification of the barriers as the perception of stakeholders on barriers may reveal the lacunae in existing policies and help in the identification of measures to overcome the barriers. Finally these approaches
literature survey, site visits, and stakeholders’ consultations
complement each other and therefore it is recommended to use all the three approaches for the identification of barriers. Feedback from stakeholders should be an important component in designing action/policies to promote RETs.

18.5 Measures to overcome barriers Renewable energy targets are an important indicator of a government’s intention to promote renewable energy. Governmental intervention is necessary due to a variety of imperfections and distortions in the market, unfavourable regulatory environments and a lack of institutional capacity to promote RETs. The government’s role could include actions to remove barriers through building institutional capacity, creating an enabling environment, providing information and setting up required mechanisms to promote RETs. Policy approaches to promote the RETs can either remove the barriers or create conditions where the market is forced to act, ignoring the barriers. The former normally works at the micro-level addressing the barriers directly, and the latter mostly at the macro level addressing the barriers indirectly. For example setting up information centres, establishing codes and standards, etc. address the barriers directly, whereas increasing energy prices through pollution taxation address the barriers indirectly. Various measures to address barriers are described in this section.

18.5.1 Renewable energy targets Political commitment to promoting renewable energy is reflected in the renewable target that a country or subnational jurisdiction set for themselves. This is then followed by various measures to achieve the targets. Target

Renewable energy technologies Chapter | 18

555

setting, therefore, is one of the best indicators of the intention and a facilitating measure for RETs promotion. It brings together stakeholders and sets in motion other actions. Almost all countries and several subnational jurisdictions had adopted renewable energy targets by 2018. Of these, a few of them (less than 10) had economy-wide targets for at least 50% renewable energy [1]. Most (92 countries) had the target for the use of renewables in the power sector. Some countries also had renewable energy targets for heating, cooling and transport sectors. EU came out with a target of meeting at least 32% of its final energy consumption from renewable sources by 2030. Though some countries and cities have 100% renewables for power targets, globally, only Denmark has a target for 100% renewables in total final energy. If taken as a sector, heating and cooling is the biggest end-use energy consumption sector, followed by transport and power and some countries now have targets for renewable heating and cooling also [1]. EU had a 19.5% share of renewable heat in 2017 and the target is an annual increase of 1.3% in the share through 2030. Barring a few countries, most of the targets for the share of renewable heating and cooling are for 2020, notable among them Denmark, which has a target of 100% renewable heating and cooling by 2050, and Lithuania 90% by 2030. Paris Agreement on climate in 2015 was affirmed by countries through submission of the Nationally Determined Contributions (NDCs) specifying actions they commit to support efforts related to climate change. More than 100 countries included renewable energy in their NDCs, which is a strong indicator of commitment to promoting renewable.

18.5.2 Renewable energy promotion measures The measures required to promote RETs thus follow from (a) identification of barriers through the administration of questionnaires/interview of the stakeholders and (b) feedback from stakeholders on the measures to overcome the barriers, obtained by extending the questionnaire/interview to include questions related to the possible measures. Finally policy actions need to be designed and implemented to operationalize the measures identified to overcome the barriers. Some of the measures taken by the EU and various other governments to remove barriers and promote RETs are listed below. Several possibilities may exist and the one that best suits a country should be chosen.

18.5.2.1 Support mechanisms The most common mechanisms to support the deployment of renewable electricity on a large scale include are feed-in tariffs (FITs), auctions (or tendering) and tradable green certificates (TGCs) along with quota, Net

556

PART | V Policy

metering, and renewable obligations are also now used in many countries. Tax incentives and cash grants were other popular support mechanisms.

18.5.2.1.1 A feed-in tariff The FIT policy guarantees renewable generators specified payments per unit of energy supplied to the grid (USD per kWh for example) over a fixed period. Access to the grid is an important part of the policy, and regulation may be made if needed. 111 countries including a few subnational jurisdictions had a feed-in tariff or its variations by the end of 2018 [1]. There have been issues related to feed-in tariff; though needed to address uncertainty for investors, long-term contracts can lead to technology lock-in and expensive power for consumers in the falling technology cost market. Some countries have put annual caps on the amount of capacity for which FIT is provided for a certain time. Germany introduced the concept of ‘breathing cap’ for solar PV which refers to the programmed tariff degression linked to the deployment of the solar PV capacity in the year before. More the capacity addition in the previous year, the faster the tariffs go down. 18.5.2.1.2 Auctions or tendering schemes Competitive bids are invited from project developers for the installation of a certain capacity for renewable energy supply from one or more technologies. Tenderers can typically ask also to quote for per unit rate of the power generated and accept the lowest bid. In a variant of this, referred to as a reverse auction, tenderer declares the lowest bid received and then goes for auction among eligible bidders to further reduce the rate. The Tendering system offers the facility to include specific requirements such as the share of local manufacturing, the maximum price per unit of energy, etc. Tenders typically help overcome economic barriers through the creation of market and cost recovery assuming that the lowest cost bidder is efficient and recovers the cost of generation. Auction system also ensures that the benefits of technological developments resulting in falling costs of renewable are passed on to consumers through competitive bidding. 48 countries, including a few subnational jurisdictions, had renewable power auction in 2018, with many preferring this over FIT. China, for example, shifted auction for utility-scale and distributed solar projects in 2018 and plans to shift to auction in case of wind also in the near future. Overall 98 countries, including a few subnational jurisdictions, had the experience of this mechanism (auction) to promote renewable energy [1]. 18.5.2.1.3 Renewable energy certificates Also referred to as TGCs, used normally in conjunction with a quota or obligation to use renewable power. The certificate is awarded to the generator of renewable energy to certify the generation of renewable energy, a certificate

Renewable energy technologies Chapter | 18

557

typically for 1 MWh. The certificates are used to meet renewable energy obligations, which are put on the power producers and big consumers of power. The certificates are tradable and obligation creates the market for them. There is a penalty if an entity is not able to meet the obligation, which determines the price of the certificate. Thirty-one countries (including one subnational jurisdiction) had renewable energy certificates (REC) mechanism operating to promote renewable power generation in 2018 [1].

18.5.2.1.4 Renewable portfolio standard Popular in the United States (and also referred to as renewable energy standards) and in some other countries, renewable portfolio standard (RPS) is similar to obligation or quota mentioned earlier in the case of RECs. RPS is also an obligation applicable to a utility company to provide a predetermined minimum targeted renewable share of power in their supply. They can buy renewable power from certified entities to meet the obligation. 33 countries, including a few subnational jurisdictions, had an RPS mechanism to promote renewable power generation in 2018 [1]. Some countries have made renewable heat for specific technology mandatory for new buildings
for example Israel introduced solar collector obligations, making solar collectors obligatory in new residential buildings starting 1980. The solar obligation helped make solar thermal systems mainstream technology in the Israel water heater market without any financial support [4]. Some more countries including Spain, Italy, Brazil and India also subsequently launched their specific renewable technology obligations. This mechanism has also been applied in the case of biofuels, where several countries have biofuel blend obligations/mandates, requiring a certain percentage of biofuel blended with fossil fuel (diesel or petrol) by a target date.

18.5.3 Net metering/net billing Net metering has emerged an important mechanism that engages a large number of stakeholders including households and the commercial sector in the generation and use of renewable power. Primarily designed to promote the decentralized generation of renewable power by household and commercial entities, the regulation allows them to generate and consume renewable power (primarily in context with PV Solar in case of households), and feed excess generation to the grid, for which they get credit. The customer receives credit at the same rate as they pay in retail to the utility under net metering, while the rate may be lower in case of net billing. Net metering is quite popular with the consumers and 66 countries including a few subnational jurisdictions had net metering to promote decentralized renewable power generation in 2018 [1].

558

PART | V Policy

18.5.3.1 Fiscal incentives This includes tax incentives, and renewable power production credits or payment. G

G

Tax incentive: The most popular mechanism to promote renewable energy, tax incentives are used by more than 100 countries with their own variations of the tax incentive schemes. Some countries exempt or reduce taxes on renewable power, while others may provide tax credits. In the United States for example the federal government provided tradable tax credits for renewable power generation. In China, a reduced rate of VAT and income tax are applied to wind energy and biogas. In India, wind and solar power are offered the lowest rate of tax on equipment and tax-free income for 10 years. Hybrid incentive scheme for renewable heat: The UK government introduced a hybrid incentive scheme in 2011 consisting of grant and tariff support to promote renewable heat. Renewable Heat Incentive policy, as it is known, is an initiative for designing a FIT policy for the heat market. For the households, it includes a grant in the first year on the installation of a renewable heat technology with support through long-term tariffs from the second year onwards.

18.5.3.2 Public financing of renewable energy A popular way to promote renewable energy, public financing, that includes public investments, loans, grants and capital subsidies or rebates, is used by most of the countries. Public financing plays an important role, particularly in the initial stage of technology development and demonstration. It is also important in countries where the private sector does not have sufficient capacity to invest or venture into new areas with risk and uncertainty, or the financial market is not well developed. Grants and subsidies help address risk as well as economic and financial barriers in the initial stage of technology diffusion. Even in the United States, grant schemes where renewable energy project developers get back 30% of the investment costs in cash were introduced after the economic and financial crisis in 2009. This helped to lower the effective price of technology to project developers, making it more competitive [4].

18.6 Current challenges From the literature on barriers, the following conclusions can be drawn: 1. Despite the growth of renewable and support by governments and other stakeholders, there are barriers to its further development.

Renewable energy technologies Chapter | 18

559

2. Though broadly, the category of barriers has not changed, they vary across regions, technologies, end uses, and in their dimension in terms of intensity and coverage. It is therefore important to examine the barriers for technology not only for their existence but also importance, intensity and similar other features. 3. Barriers also vary in terms of their importance
some core barriers lead to the presence of other barriers, and therefore once the core barriers are addressed, other related barriers may simply disappear. For example the economic barrier can be on account of the high cost of technology (cost barrier), market failure (barrier due to direct or indirect subsidy to competing technologies), or high import tariffs (trade barriers). Addressing any one of these three barriers or a breakthrough in technology leading to lower cost of technology may help address not only an economic barrier but make other barriers also irrelevant. Similarly a lack of awareness/ information about technology can lead to its nonacceptance. For example the high upfront cost of a solar home system can dissuade users not knowing that it has a short payback period, particularly in regions where there is good solar radiation and alternate source of hot water is high tariffs electricity and its life cycle costs can be much smaller. It is therefore important to identify and focus on the removal of such barriers. 4. With maturing renewable technologies and their increased penetration, new barriers have emerged in countries with significant contributions from renewables. From ‘no more wind turbines in my backyard’ to ‘no more of this intermittent power in my grid’ new challenges have emerged, some requiring sophisticated approaches, complexities of which are themselves a barrier. Danish integration of higher wind energy in the grid, for example has been possible due to flexibility in conventional production, strong transmission and distribution networks, and a larger exchange of power with neighbouring countries. Replicating Danish experience in a different setting can be very challenging. Similarly battery storage, vehicle to grid (V2G) technologies have been suggested as opportunities to address the barrier related to the absorption of intermittent renewable power in the grid. The current state of renewable energy development indicates the success of efforts made by various countries to address various barriers that renewable energy faced. The results, however, are mixed in countries depending on the availability of resources and their capacity to handle the challenges faced by them. As a result, the development of renewable energy is dissimilar across countries. Several countries where renewable energy has made good inroads have now reached to stage where they face new challenges that include absorption of VRE in the grid and increased application of renewable energy to heating and cooling, and transport sector, two crucial sectors to increase its penetration. In the transport sector, there is a need for integrated

560

PART | V Policy

policies to address three issues; use of renewable fuel, vehicles that use renewable fuels, and infrastructure development for renewable fuel distribution [6]. Similarly for greater penetration to heating and cooling, integrated policies for renewable fuel use and infrastructure development may be needed. The integration of VRE in the system requires increased flexibility in the system as well as better forecasting techniques for VRE generation. Better system operations and market designs that have incentives to absorb VRE may be other measures. Increased flexibility may require integration with other regional grids that have more flexibility, energy storage and innovative measures such as vehicles to a grid system that can provide flexibility. Policies that promote these measures will be needed. Some areas that hold the good potential to push renewable energy production and use but difficult to implement include a carbon tax, energy subsidy reform and behavioural changes. The first two send price signals to the producers as well as consumers impacting their behaviour. Behavioural changes can also be brought through better awareness and information, in addition to taxes and penalties.

References [1] REN21. Renewables 2019 global status report. Paris: Renewable Energy Policy Network for the 21st Century; 2019. ,https://www.ren21.net/reports/global-status-report/.. [2] GTR. Global trends in renewable energy investment 2019. Frankfurt School-UNEP Centre/ BNEF; 2019. ,http://www.fs-unep-centre.org.. [3] IEA. Renewables information 2019. Paris: International Energy Agency; 2019. ¨ lz S. Renewable energy
policy considerations for deploying [4] Mu¨ller S, Brown A, O renewables. Information Paper, International Energy Agency, Paris; 2011. ,https://www. iea.org/publications/freepublications/publication/Renew_Policies.pdf.. [5] REN21. Renewables 2018. Global status report. Renewable energy policy network for the 21st century. Paris; 2018. ,http://www.ren21.net/wp-content/uploads/2018/06/17-8652_ GSR2018_FullReport_web_final_.pdf.. [6] IRENA, IEA, REN21. Renewable energy policies in a time of transition. IRENA, OECD/ IEA, and REN21; 2018. ,http://www.irena.org/-/media/Files/IRENA/Agency/Publication/ 2018/Apr/IRENA_IEA_REN21_Policies_2018.pdf.. [7] IPCC. Renewable Energy Sources and Climate Change Mitigation; Special Report of the Intergovernmental Panel on Climate Change; 2012. ,https://www.ipcc.ch/pdf/presentations/Rio20/Rio20_puc_aivanova.pdf.. [8] Wiseman J, Taegen E, Luckins K. Post Carbon Pathways report; towards a just and resilient post-carbon future. Discussion paper, Center for Policy Development; 2013. ,https:// cpd.org.au/wp-content/uploads/2013/04/Post-Carbon-Pathways-Report-2013_Revised.pdf.. [9] UCS. Barriers to Renewable Energy Technologies. Union of Concerned Scientists; 2017. ,https://www.ucsusa.org/clean-energy/renewable-energy/barriers-to-renewable-energy#. W1tS9MInapo.. [10] Kariuki D. Barriers to renewable energy technologies development; 2018. ,https://www. energytoday.net/economics-policy/barriers-renewable-energy-technologies-development/..

Renewable energy technologies Chapter | 18

561

[11] ADBI. Financial barriers to development of renewable and green energy projects in Asia. Asian Development Bank Institute Working Paper Series No. 862; 2018. [12] Reddy S, Painuly JP. Diffusion of renewable energy technologies
barriers and stakeholders’ perspectives. Renew Energy 2004;29(9):1431
47. [13] Shivakumar A, Dobbins A, Fahl U, Singh A. Drivers of renewable energy deployment in the EU: an analysis of past trends and projections. Energy Strategy Rev 2019;26:100402. [14] Hu J, Harmsen R, Crijns-Graus W, Worrell E, van den Broek M. Identifying barriers to large-scale integration of variable renewable electricity into the electricity market: a literature review of market design. Renew Sustain Energy Rev 2018;81:2181
95. [15] Sen S, Ganguly S. Opportunities, barriers and issues with renewable energy development
a discussion. Renew Sustain Energy Rev 2017;69:1170
81. [16] IEA. Renewables 2019. Paris: International Energy Agency; 2019. [17] Go R, Kahrl F, Kolster C. Planning for low-cost renewable energy. Electr J 2020;33: 106698. [18] Edenhofer O, Lion H, Brigitte K, Michael P, Steffen S, Eva S, et al. On the economics of renewable energy sources. Energy Econ 2013;40:S12
23. [19] Blazquez J, Fuentes-Bracamontes R, Bollino CA, Nezamuddin N. The renewable energy policy Paradox. Renew Sustain Energy Rev 2018;82:1
5. [20] Lo H, Blumsack S, Hines P, Meyn S. Electricity rates for the zero marginal cost grid. Electr J 2019;32:39
43. [21] Foster E, Contestabile M, Blazquez J, Manzano B, Workman M, Shah N. The unstudied barriers to widespread renewable energy deployment: fossil fuel price responses. Energy Policy 2017;103:258
64. [22] Owen AD. Renewable energy: externality costs as market barriers. Energy Policy 2006;34 (2006):632
42. [23] IEA. World energy investment 2019. Paris: International Energy Agency; 2019. [24] Seetharaman, Moorthy K, Patwa N, Saravanan, Gupta Y. Breaking barriers in deployment of renewable energy. Heliyon 2019;5(2019):e01166. [25] Blazquez J, Nezamuddin N, Zamrik T. Economic policy instruments and market uncertainty: exploring the impact on renewables adoption. Renew Sustain Energy Rev 2018;94:224
33. [26] Diesendorf M, Elliston B. The feasibility of 100% renewable electricity systems: a response to critics. Renew Sustain Energy Rev 2018;93:318
30. [27] Painuly JP, Wohlgemuth N. Economics of renewable energy. Handbook of energy economics. Routledge; 2019. [28] Chassein E, Roser A, John F. Using renewable energy for heating and cooling: barriers and drivers at local level; an analysis based on a literature review and empirical results from local case studies. IREES; 2017. ,http://www.progressheat.eu/IMG/pdf/progressheat_wp3.2_report_publication.pdf.. [29] Vimmerstedt L, Brown A, Heath G, Mai T, Ruth M, Melaina M, et al. High penetration of renewable energy in the transportation sector: scenarios, barriers, and enablers. In: Presented at the 2012 World Renewable Energy Forum, Denver, CO, May 13
17; 2012. ,https://www.nrel.gov/docs/fy12osti/54442.pdf.. [30] Dhar S, Cherla S. A study of electric mobility for city of Hyderabad; 2019. ,https://backend.orbit.dtu.dk/ws/portalfiles/portal/141857029/UMI_Electric_Mobility_ Hyderabad_Final.pdf.. [31] IEA, ISA. Solar energy; mapping the road ahead. Paris: International Energy Agency; 2019.

562

PART | V Policy

[32] Painuly JP. Barriers to renewable energy penetration; a framework for analysis. Renew Energy 2001;24(1):73
89. [33] Nguyen NT, Ha-Duong M, Tran TC, Shrestha RM, Nadaud F. Barriers to the adoption of renewable and energy-efficient technologies in the Vietnamese power sector. GMSARN Int J 2010;4(2):89
104 halshs-00444826. [34] Richards G, Bram N, Ken B. Barriers to renewable energy development: a case study of large-scale wind energy in Saskatchewan”, Canada. Energy Policy 2012;42(2012):691
8. [35] Dio´genes JRF, Claro J, Rodrigues JC. Barriers to onshore wind farm implementation in Brazil. Energy Policy 2019;128:253
66.

Chapter 19

Policies for a sustainable energy future: how do renewable energy subsidies work and how can they be improved? Qin Bao, Jiali Zheng and Shouyang Wang Center for Forecasting Science, Chinese Academy of Sciences, University of Chinese Academy of Sciences, Beijing, P.R. China

Chapter Outline 19.1 Introduction 563 19.2 Renewable energy development and renewable energy subsidies 567 19.2.1 The development of renewable energy varies across countries 567 19.2.2 A brief review of the renewable energy subsidy policies in United States 570 19.2.3 A brief review of the renewable energy subsidy

policies in European Union 571 19.2.4 A brief review of renewable energy subsidies in China 575 19.3 The mechanism of how renewable energy subsidy works 577 19.3.1 A model of renewable energy generation 577 19.3.2 Discussion and policy implications 580 19.4 Conclusions 581 References 582

19.1 Introduction Renewable energy is playing an increasingly important role in the energy system. As over-reliance on fossil energy has caused severe problems of the environment and climate change, it is significant to restructure a smart energy system by shifting from fossil fuels to renewable energies [1]. The development of renewable energy, especially the advancement of renewable energy related technologies, would lead the energy system and the economic system to a more sustainable future [2]. Due to the fact that renewable Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00019-4 © 2021 Elsevier Inc. All rights reserved.

563

564

PART | V Policy

energy has played significant role in the energy system from both supply and demand sides, the related sustainable development has become the centre of national policies and strategies for many countries all over the world. There are mainly three predominant motivations for the development of renewable energy, including mitigating global climate change by reducing greenhouse gas (GHG) emissions, improving security of energy supply by relaxing the dependence on fossil energy, promoting energy diversification and stimulating related industries and employment [3]. For example, the development of renewable energy in Europe has been proved to provide ‘triple dividend’ including lower emissions, higher GDP and higher employment [4]. The development of renewable energy has made great progress in the past two decades. From the demand side, as shown in Fig. 19.1, the renewable energy consumption has quickly increased from 63.9 million tonnes oil equivalent in 2003 to 561.3 million tonnes oil equivalent in 2018, with an average annual growth rate of 15.6%. The overall share of renewable energy in total primary energy consumption has quickly increased to 4.0%. From the supply side, as shown in Table 19.1, the capacity of global renewable energy generation grew to around 2378 GW in the end of 2018 with around 181 GW newly installed in 2018, and the share of renewable energy in global installed electricity generating capacity grew to accounting for more than 33% [5]. As a result, an increasing number of countries have varied their electricity mixes with more renewables and the shares of wind power and solar photovoltaic have increased further. The development of renewable energy relies heavily on policies. The reasons can be concluded mainly from two aspects. First, climate and environmental issues have strong externalities, which requires the government

FIGURE 19.1 Renewable energy consumption and its share in total primary energy consumption. Source: BP Statistical Review of World Energy.

Policies for a sustainable energy future: how do renewable Chapter | 19

565

TABLE 19.1 Renewable energy indicators (existing at end-2018). Power

Unit

2018

%

Renewable power capacity

GW

2378

100

Hydro power capacity

GW

1132

47.62

Wind power capacity

GW

591

24.86

Solar photovoltaics capacity

GW

505

21.24

Bio-power capacity

GW

130

5.47

Geothermal power capacity

GW

13.3

0.56

Concentrating solar thermal power capacity

GW

5.5

0.23

Ocean power capacity

GW

0.5

0.02

CSP, Concentrating solar thermal power. Source: REN21. Renewables 2019 Global Status Report. Paris: REN21 Secretariat; 2019.

regulation by policies to achieve better results. The appropriate mechanism design of energy policy can achieve the internalization of environmental benefits of renewable energy [6]. Second, the renewable energy is mainly technology-intensive, and these technologies usually require large input and have strong spillover effect, whereas policies can make them profitable by adjusting and rebalancing costs and revenues [7]. As revealed by a majority of research, a well-designed portfolio of policies could achieve better mitigation effect at a relative lower economic cost. Amongst these candidate policies, renewable energy subsidy and carbon price or tax policies are the main instruments used by government to provide effective incentives for restructuring the energy sources towards renewables. By changing the relative prices of different energy sources, these policies can help mitigate market distortions and amend market failures [8]. Since the end of the 1990s, countries worldwide have paid increasing attention to the development of renewable energy. The strategic goals for the renewable energy development are formulated and various related policies including subsidies are introduced to support the renewable energy [9]. In general, there is no consensus definition for ‘renewable energy subsidy’. From a broad approach, all efforts taken by the government targeting both energy sector or other sectors to improve the development of renewables can be viewed as renewable energy subsidies, including tax expenditures, direct expenditures, research and development (R&D) support, loan guarantees and so on. For this chapter, we focus on the subsidies that are directly related to the renewable energies serving as primary energy. In this sense, renewable energy subsidies work as one of the most popular policy instruments in promoting the renewable energy generation and improving the power systems.

566

PART | V Policy

There is a majority of literature related with renewable energy subsidy policies aiming at electricity sector. These researches can be narrowed down from general subsidy to specific policies [10]. The transmission mechanism of these subsidy policies and related consequences are the core focus. Unlike imposing an additional cost on the use of fossil energies or on the carbon emissions to cover the negative externality, either by a carbon tax or by certain emission trading systems, the subsidy policies aim to provide incentives for the development of renewable energies directly. In the ideal economic system, these policies might be equivalent. However, in the real economic system, due to various frictions of the market, the effects of these policies might differ greatly. A large body of the literature focuses on the policy evaluations for various renewable energy generation supporting policies, and subsidies are suggested to be effective in certain aspects, especially when coupled with other policies [8,11]. There are also critiques for renewable energy subsidies, mostly from the efficiency perspective. For example, Palmer and Burtraw [12] implied that the renewable energy subsidy policy with a production tax credit is less effective than a portfolio standard policy. Lapan and Moschini [13] suggested that biofuels mandates outperform subsidies from the welfare perspective. Amongst these subsidy policies, feed-in tariff (FIT) is widely used to advance renewable energy generation. By the end of 2018, the FIT scheme for renewable power has been adopted by a total of 111 states, provinces or countries [5]. The FIT policies provide guaranteed prices for renewable power suppliers, which are higher than the market price for nonrenewable power. It helps to amend competitive disadvantage caused by relatively high cost of renewable energy generation. Thereby, it fosters investment and development of renewable energy in the power systems. The existing literature on FIT concentrates on investigating the impact of FIT on different renewable sources in different countries. Some studies focused solely on the analysis of FIT policy [14], while others compared FIT with other policies, such as renewable portfolio standard (RPS) [15
18], emissions tax [19], etc. Due to the complex institutional constraints, the eventual policy design and funding of FIT policy is suggested to be based on a careful and comprehensive policy assessment. To assess the effectiveness of FIT and other renewable energy subsidy policy, various methods have been employed. First, the most commonly used method is econometric modeling. For example, the instrumental variables approach was used to estimate the causal relationship of FIT policy and renewable energy generation in 26 industrialized countries, indicating that FIT is an effective method to increase renewable energy generation [20]. However, due to the different selections of variables and indicators, the econometric studies might get different results. The FIT policy is implied to be effective in several aspects, including stimulating renewable energy generation [21], showing advantages over portfolio standard in certain aspects

Policies for a sustainable energy future: how do renewable Chapter | 19

567

[22], and improving the deployment of renewable energy worldwide [23]. Second, input
output analysis is another frequently used tool for quantitative analysis of renewable energy policies [24
26]. The impact of FIT policy on the environment, economy and society can be quantified based on the extended environmental input
output technologies. For example, Behrens et al. [27] studied the impacts of FIT in Portugal over the period 2000
2010 by using a hybrid energy-economic input
output model and suggested that the FIT policy reduced the emissions, while increased GDP and employment. Third, the computable general equilibrium (CGE) model also played as a useful tool to assess the general impacts of FIT policy under economic equilibrium. For example, Bo¨hringer et al. [28] suggested that the impact of FIT policy in Canada on employment is negative. Tabatabaei et al. [29] implied that the FIT policy in Iran is most effective in a technology-neutral scenario, with smaller negative impact on GDP and less financial expenditure burden. Chatri et al. [30] indicated that the FIT policy in Malaysia contributes to the renewable energy expansion of related industries. As disclosed by a huge amount of existing literature, the renewable energy subsidy policies vary in different economies and in different time, and the impacts of them are suggested to be different. In this chapter, we take a historical perspective to draw a map of the development of renewable energies all over the world. The stylized economies are analysed, and the clues between the development of renewable energies and renewable energy subsidy policies are clarified. Despite the general equilibrium model is a good way to provide an overall estimation of the policies [8], due to the complexity of the real economy, a simple model with fewer parameters would provide us better understanding for the interactions between different variables. Thereby, a two-stage behavioural model is used to analyse how the renewable energy subsidy policies can be improved to promote the development of renewables and the implications are further discussed. The following part of this chapter is organized accordingly as follows: in Section 19.2, the development of renewable energy all over the world is reviewed, and the renewable energy subsidy policies and their effects are reviewed for the US, the EU and China. In Section 19.3, a two-stage behavioural model is built to discuss the optimal design of renewable energy subsidy policies. Section 19.4 provides concluding remarks.

19.2 Renewable energy development and renewable energy subsidies 19.2.1 The development of renewable energy varies across countries The development of renewable energy accelerates in these two decades. As shown in Fig. 19.2, the total world renewable energy generation was

568

PART | V Policy

FIGURE 19.2 Renewable energy generation (TWh) and its annual growth rate. Source: BP Statistical Review of World Energy.

FIGURE 19.3 Total world renewable energy generation by source (TWh). Source: BP Statistical Review of World Energy.

2480.4 TWh in 2018, more than 10 times of that in 2000. The renewable energy generation has been developed rapidly from 2004 both in OECD and non-OECD countries, with the annual growth rate ranging from 13% to 19%. In detail, the source of renewable energy generation can be generally divided into solar, wind, geothermal, biomass and other. As shown in Fig. 19.3, the wind power dominates global renewable energy system, followed by solar. In 2018, the renewable energy generation by wind, solar, and geothermal, biomass and other were 1270, 585 and 626 TWh, respectively, accounting for 51%, 24% and 25%, respectively.

Policies for a sustainable energy future: how do renewable Chapter | 19

569

FIGURE 19.4 The US renewable energy generation by source (TWh) and its share. Source: BP Statistical Review of World Energy.

FIGURE 19.5 The EU renewable energy generation by source (TWh) and its share. Source: BP Statistical Review of World Energy.

Amongst all the economies, the renewable energy development in the US, the EU and China contributes the most. The US started its development of renewables at a relatively early stage. As shown in Fig. 19.4, the renewable energy generation in the US was 72.8 TWh in 2000, which contributes to a share of 33.4% in the world. However, as the EU implemented a series of policies to improve the development of renewables, the share of the US declined. In 2018, the renewable energy generation of the US was 6.3 times of that in 2000, contributing to 18.5% of the world. Compared to the US, as shown in Fig. 19.5, the renewable energy generation in the EU was 706 TWh in 2018, almost 11.3 times of that in 2000, contributing to 28.4% in the total world generation. The peak of the EU’s share in the total world

570

PART | V Policy

FIGURE 19.6 China renewable energy generation by source (TWh) and its share. Source: BP Statistical Review of World Energy.

generation was 42.7% in 2007. As China took a fast development pace in renewable energies, the renewable energy generation in China has quickly increased since 2008, from 15.5 TWh in 2007 to 634.2 TWh in 2018, with its share in the world increasing from 3.3% to 25.6% (Fig. 19.6). The differences of the development pattern for renewable energies in these economies can be attributed to the different renewable energy policies, especially the subsidy policies. Both the EU and China have undertaken ambitious projects for renewable energy development with active supporting policies, while the US took a relatively slower pace. In the following part of this section, we will discuss the renewable energy policies in these areas in detail.

19.2.2 A brief review of the renewable energy subsidy policies in United States As one of the economies that started the renewable energy development at a relatively early stage, the US has systematically formed a range of policy instruments in support of renewable energy development, including the direct subsidy, tax credit, accelerated depreciation, fund support, loan guarantee, etc. The overall policy framework for promoting renewable energy development in the US is a combination of federal policies and state policies. The federal policy system provides incentives for the renewables to effectively reduce project costs, promote technological progress and accelerate industrialization. The main policy instruments include long-term preferential tax policies based on investment tax credit and production tax credit. The Public Utility Regulatory Policies Act in 1978 implemented the investment tax credit on renewable energy for the first time with the deregulation of the electricity market, which facilitated power generation from renewables [31]. The Energy Policy Act of 1992 introduced the production tax credit policy, which provided financial incentives for electricity produced from new

Policies for a sustainable energy future: how do renewable Chapter | 19

571

qualifying renewable energy facilities [32]. The American Recovery and Reinvestment Act of 2009 provided a large financial subsidy for the renewable projects. Furthermore, the US department of Energy announced a series of funding programs, focusing on advanced renewable technology research including solar, biomass, geothermal, etc. In the state level, the policies seek to promote the development of renewable energy through market mechanisms. The RPS is the popular policy scheme for promoting the development of renewables. The RPS policy has been widely adopted by a majority of the US states since the late 1990s, and 37 states have implemented it by 2017. By requiring the electricity providers to meet a growing portion of their electricity supplies with eligible forms of renewable energy, the RPS policy drives the growth of renewable energy in the US. For example, Barbose et al. [33] assessed the benefits and impacts of RPS at the US national level and suggested that the estimated benefits wellexceed its costs. Bhattacharya et al. [34] studied the market and welfare effects of RPS in the US and concluded that this policy might reduce the total quantity of green power used. In general, the effects of the renewable energy subsidies in the US were disclosed to be twofolds [21,35]. On the one hand, these policies do stimulate renewable power generation, create jobs or improve energy efficiency. On the other hand, the subsidies might be ineffective in reducing carbon emissions due to the rebound effect or calculated by the life cycle analysis. Newell et al. [10] provided a systematic overview for the US subsidy policies for clean energy and draw several highlights for related policy design, such as providing subsidies based on physical outcomes.

19.2.3 A brief review of the renewable energy subsidy policies in European Union As the leader of the renewable energy technologies worldwide, the EU governments have implemented a various of policies to promote energy from renewable energy sources at an early stage. In December, 1997, the ‘White Paper for a Community Strategy and Action Plan, Energy for the Future: Renewable Sources of Energy’ was issued, which set specific targets for renewable energy sources to promote the development of renewable energy. Later, the Directive 2001/77/EC was put into forth on the promotion of electricity produced from renewables, and the Directive 2003/87/EC established a scheme for GHG emission allowance trading within the Community. In 2009, the European Commission’s Directive 2009/28/EC has established the mandatory national targets, that is ‘a 20% share of energy from renewable sources’ by 2020 [36]. It allows each member state to implement proper renewable energy policies according to their own situations. Under this Directive, a majority of policies are designed and executed at the national level to meet the targets. The supporting policies include subsidies, tax

572

PART | V Policy

credits, preferential loans for investment, and FITs for renewable energy generation, etc. [37]. These policies have brought significant effects on promoting renewables. By 2014, the EU had achieved a share of renewable energy by 16%, of which 9 member states had achieved the 2020 target. By 2016, that number had increased to 17%, with 11 member states meeting the 2020 target. The latest EU-wide Renewable Energy Progress Report was published in 2017 by the European Commission, which was based on the 2015 data. The report summarized several contributions of renewables, including improving energy security, enhancing market integration, increasing energy efficiency, reducing GHG emissions, stimulating innovation, encouraging economic growth and promoting sustainable jobs. Studies of the EU member states indicate that the renewable energy policies can stimulate modest economic growth and have a positive impact on the employment, which depends largely on the burden of cost rise. For example, Ragwitz et al. [38] studied the impacts of renewable energy policy in the EU and found that the increase in energy costs resulting from renewable energy policies can be offset by economic growth. Boeters and Koornneef [39] studied the effects of the implementation of the EU renewable energy policy by using a CGE model and implied that these policies increase energy security, bring more jobs and improve technology leadership. As for the renewable energy subsidy policy, a majority of literature has proved their positive role on promoting the development of renewable energy generation [40,41]. As revealed by the literature, the policy design for renewable energy generation is suggested to be core for its effects in the EU [42]. Li et al. [43] studied the effectiveness of different policy instruments in 21 EU countries and confirmed that FIT is more efficient than RPS in advancing the development of solar photovoltaic (PV) and wind power. For the FIT policy, it may differ in various aspects, including fix-price or premium tariff, cost allocation mechanism, cost containment, contract duration from 10 to 25 years, tariff amount, digression rate, etc. [44]. Campoccia et al. [45] analysed the FIT policies for solar PV in France, Germany, Greece, Italy, Spain and the UK and suggested that different ways of implementing FIT policies could have significantly different results. Jenner et al. [44] suggested that policy design together with electricity price and production cost are all important determinants for the effectiveness of FIT. Since Germany, Spain and Denmark are the three representative states in the EU who adopted the FITs policy, we will further discuss their policies in detail.

19.2.3.1 Germany As a major energy consumer in the world, Germany has developed rapidly in wind power and solar PV in recent years. It has achieved successful energy transition through the top-down regulation and policy constraints as well as

Policies for a sustainable energy future: how do renewable Chapter | 19

573

the promotion of incentive mechanisms. The national overall target of Germany for the share of energy from renewable sources in gross final consumption in 2020 is 18% [36]. The FIT for renewables was first implemented in 1991 under the Electricity Feed-in Law (Stromeinspeisungsgesetz, SEG). The FIT policy significantly stimulated the development of renewables, and the installed wind power capacity grew fast. Later, the evolution of FIT policy in Germany can be phased basically by the development process of the Renewable Energy Act (Erneuerbare
Energien
Gesetz, EEG), which was passed in 2000 and modified in 2004, 2009, 2012, 2014 and 2017 [46]. Accordingly, the development of EEG can be roughly divided into six phases [47]. The FIT scheme was maintained from SEG with minor modifications. The EEG2000 (2000
2003) marked the official launch of the German renewable energy electricity generation market with the FIT scheme. The EEG2004 (2003
2008) improved the FIT to promote the development of renewables. The EEG2009 (2009
2012) proposed a transition to market approach, and the EEG2012 (2012
2014) revised the market mechanism. The EEG2014 (2014
2016) took an overall transition of renewable subsidies from FIT to feed-in premium. The EEG2017 (2017-) was the latest revision, which adopted the auction model to determine the subsidies for renewable electricity. It further responds to the EU’s request for member states to adjust their policies and gradually reduce renewable energy subsidies. The FIT policies implemented by Germany have achieved significant effects in promoting renewables. For example, Hitaj and Lo¨schel [48] estimated the impact of FIT on wind power in Germany and found that 1 h-cent/ kWh increase in the FIT rate would on average increase 796 MV additional capacity per year from 1996 to 2010. However, the FIT policy is expensive. As estimated, the cost of EEG amounts can be up to about 26 billion Euro in 2016, with the surcharge paid by electricity consumers rising to 6.35 cents/ kWh in 2016, roughly a fourth of the average households’ consumer price [49]. The financial burden of FIT might increase income inequality [49], and it is suggested that there is no significant difference between EEG tariff and SEG tariff with lower tariff rate on innovation impact [50].

19.2.3.2 Spain Spain is one of the leading countries to develop renewable resources such as wind and solar. The Spanish government has introduced a large number of policies to stimulate and support the development of the renewable energy industry. For example, the Spanish government formulated the Spain’s National Renewable Energy Action Plan in 2010, and approved the 2005
2007 Action Plans, the 2008
2012 Action Plans and the 2008
2011 Activation Plans. These policies have effectively promoted the change of its energy structure and enhanced the use of renewable energy. The national

574

PART | V Policy

overall target of Spain for the share of energy from renewable sources in gross final consumption of energy in 2020 is 20% [36]. The Spanish government introduced the feed-in policies since 1998. The main body of the system includes two alternative schemes: one is a fixed FIT and the other is a premium payment based on the electricity market price. The electricity producers can choose between these two options. Besides, there is no time limit for the supporting, but the fixed tariffs are reduced after either 15, 20 or 25 years depending on the technology [37]. The feed-in policies, together with other soft loans, tax incentives and regional investment incentive policies, greatly promoted the development of renewables. Furthermore, the RD661/2007 established a cap and floor system in 2007, and the premium was adjusted to a variable payment depending on the price in the spot market with cap and floor [51]. However, the financial crisis in 2008 brought significant impact on the renewable development model in Spain. The renewable energy generation in Spain greatly improved before 2008 with large private investment in the renewable industries. Nevertheless, due to the fiscal shock, Spain experienced a sudden transformation in its renewable energy development model in 2012, with the national government support and financial incentives for renewables removed. The research on the effectiveness of the policies in Spain is controversial. For example, by carrying out a case study of wind generation in Spain, De Miera et al. [52] indicated that the reduction in wholesale electricity caused by more renewable energy generation is greater than the costs increase for the consumers due to the FIT scheme, thereby providing positive effects. It suggests that the three goals of the renewable energy deployment, carbon emission reductions and moderate burden on consumers can be achieved simultaneously with careful policy design. However, Bean et al. [53] studied the renewable energy policies in Spain by using a cost-benefit analysis and suggested that the investment credit policy would yield similar results as FIT but with lower costs. Ciarreta et al. [54] compared FIT and tradable green certificates and implied that the latter could help achieve the 2020 targets with lower regulatory costs.

19.2.3.3 Denmark Denmark has an outstanding performance in promoting the renewable energy development. With clear development plans and goals, the Danish government has provided strong policy supporting for renewable energy utilization. For example, in 1996, the national plan set the goal that the energy consumption in 2005 to be 12%
14% depended on the renewables and to be 35% by 2030 [55]. These policies encouraged the public to actively participate in the process and promoted the international transaction of renewable energy, which together effectively improved the Danish energy structure towards renewables. The national overall target of Denmark for the share of energy

Policies for a sustainable energy future: how do renewable Chapter | 19

575

from renewable sources in gross final consumption of energy in 2020 is 30% [36], and the Danish government sets an ambitious goal in ‘Energy Strategy 2050
from coal, oil and gas to green energy’, that is more than 60% of electricity production will be renewable sources by 2020. The development of renewables in Denmark started at a relatively early stage, which could be sourced back to the 1970s. After the oil crisis, the shift of the energy system towards renewables has gradually taken place, motivated by the urgent need to improve energy security and provide sustainable energy supply. A series of policies and measures have been adopted to promote the development of renewables, including FIT, investment subsidies, tax exemption, etc. Denmark was amongst the leading countries that introduced the FIT scheme at a relatively early stage. In 1993, the FIT scheme was introduced in Denmark, as utilities were obligated to purchase wind-generated electricity at 85% the price paid by consumers [56]. This favourable FIT scheme was designed under a table legal framework, which effectively promoted the development of renewable power. In 2001, the FIT policy was modified towards a free-market approach, with FIT abandoned and the RPS mechanism with tradable green certificates promoted. This shift was due to the increasing of cost burden of FIT as well as the pressure from the European Commission on liberalization of energy markets [57]. The policy shift has an obvious effect on slowing the development of renewables in Denmark. Thereafter, as the ambitious goal of 100% renewable energy was set, the Danish government introduced the feed-in premium policy in the Law on the Promotion of Renewable Energy, which was put in forth in 2009 and amended in 2014.

19.2.4 A brief review of renewable energy subsidies in China The development of China’s renewable energy started relatively late compared to the developed countries. The Chinese government establishes an ambitious target in the ‘Strategy for Energy Production and Consumption Revolution (2016
2030)’, that is increasing the proportion of nonfossil energy consumption to 15% by 2020, to about 20% by 2030 and to above 50% by 2050. Driven by this great ambition, strong policies are implemented and China has made great progress in renewable energies in recent years. The installation for wind and solar power has roared at high speed, and the proportion of China’s renewable energy generation continues to rise. The Renewable Energy Act came into force in 2006, which provided legal framework for the development of renewables. Both direct and indirect subsidies are provided in China to stimulate the investments in renewables. As an important direct subsidy, the FIT scheme was set up in 2009 for the wind power and in 2013 for the solar PV [58,59]. There are several key points. First, different FIT tariffs relative to the baseline price, that is the regulated coal-fired price level, are applied for different resources. Initially for

576

PART | V Policy

TABLE 19.2 The feed-in tariff for on-shore wind power in different resources areas (Yuan/kWh). Time

I

II

III

IV

Referential price in 2020

0.29

0.34

0.38

0.47

Referential price in 2019 after 2019.7.1

0.34

0.39

0.43

0.52

After 2018.1.1

0.40

0.45

0.49

0.57

After 2016.1.1

0.47

0.50

0.54

0.60

After 2015.1.1

0.49

0.52

0.56

0.61

After 2009.8.1

0.51

0.54

0.58

0.61

Source: Based on the information from website of National Development and Reform Commission (NDRC).

the wind power, four resources areas were set with FIT tariff from 0.51 to 0.61 Yuan/kWh. For the solar PV, three resources areas were set with FIT tariff from 0.90 to 1.0 Yuan/kWh, and the tariff for the distributed solar PV was 0.42 Yuan/kWh. Second, the cost sharing system is implemented, as the Renewable Energy Development Fund is built and the renewable surcharge is charged from the final electricity consumers. Third, the tariffs for the feed-in policy were adjusted according to the development of renewables and related technologies. The tariffs for the on-shore wind power and solar PV are as shown in Tables 19.2 and 19.3, respectively. As the development of renewables in China is dependent heavily on the renewable subsidies, China is now facing great challenges. The renewable subsidies have brought a major financial burden on the Renewable Energy Development Fund. As estimated, by the end of 2018, the fund had already accumulated a deficit of over RMB 100 billion. China’s National Energy Administration is now aiming to promote wind and solar power grid parity policy, which will reduce the subsidy burden as well as maintain a stable market. As a measure stepping towards the promotion of subsidy-free renewables, the Chinese government has changed the fixed price tariff into the auction premium under the referential price since 2019. It is supposed that after 1 January, 2021, the new installed on-shore wind power will be subsidy-free. The renewable subsidy policies in China are proved to be effective in the promotion of renewables and the economy. For example, Lin and Moubarak [60] suggested that there is a long-term bi-directional causality between renewable energy consumption and economic growth in China. Mu et al. [61] studied the impacts of renewables development on employment by using the input
output analysis and estimated that every 1 TWh of solar PV and wind power will create 45,100 direct jobs and 15,800 indirect jobs in China,

Policies for a sustainable energy future: how do renewable Chapter | 19

577

TABLE 19.3 The feed-in tariff for solar PV in different resources areas (Yuan/kWh). Time

I

II

III

Distributed

Referential price in 2020

0.35

0.40

0.49

0.05/0.08

Referential price in 2019 after 2019.7.1

0.40

0.45

0.55

0.10/0.18

After 2018.6.1

0.5

0.6

0.7

0.32

After 2018.1.1

0.55

0.65

0.75

0.37

After 2017.1.1

0.65

0.75

0.85

0.42

After 2016.1.1

0.80

0.88

0.98

0.42

After 2013.9.1

0.90

0.95

1.0

0.42

Source: Based on the information from website of NDRC.

respectively. Specially, the FIT policy are implied to have positive effects on China’s economy, employment as well as emission reductions [62,63].

19.3 The mechanism of how renewable energy subsidy works 19.3.1 A model of renewable energy generation The existing literature employed various methods to analyse the impacts of renewable energy subsidies on the economy and environment; however, few of them paid attention to the mechanism how subsidy policies affect the promotion of renewable energies. As disclosed by the literature, government plays an important role in the initial phase of the introduction of renewable energy technologies, and the subsidy policies should try to balance the cost burden and the benefit from renewables while stimulating related innovation. Hereafter, we provide a behavioural two-stage model to study the optimal policy design of renewable energy subsidy. In the first stage, the government sets the policy scheme to promote renewable energy generation, including the FIT surcharge, the tax, and others. In the second stage, the electricity generation enterprises make optimal decisions for both the proportion of renewable energy generation and R&D investment in renewable technologies. In this model, the development of the renewable energy generation will be determined by both the government policies and the behaviours of electricity generation enterprises. In detail, the model setting is as follows.

19.3.1.1 Government It is assumed that the government takes a basket of policies to encourage renewable energy generation, including the subsidy for the renewables and

578

PART | V Policy

tax for the fossil fuels. The FIT scheme is employed by the government under a constraint of balance between revenue and expenditure. The detailed policies are assumed to be implemented as follows: first, a renewable energy development fund is established by collecting renewable surcharge from the end users of electricity, which is denoted as TN per kWh. For simplicity, the electricity demand is assumed to be exogenous, which is represented by Q. Second, it is assumed that the subsidy for renewable energy is carried out through a fixed FIT scheme. The benchmark price for fossil energy power is denoted as Pc and the price for renewable energy is assumed to be Pc 1 β, where β is the subsidy for the renewables. Third, the government is supposed to impose a special environmental tax on the electricity produced of fossil energy from the electricity generation enterprises, with a tax rate of TE. The tax can also be viewed as emission tax or carbon tax, which is poured into the renewable energy development fund in a dedicate way. Fourth, in each period, the government is required to achieve a balance between the total revenue and the expenditure for the renewable energy development fund. If the proportion of renewable energy generation in the total electricity generation is denoted as λ, then the constraint of the government can be expressed as Eq. (19.1). TN UQ 1 ð1 2 λÞUQUTE 5 βUλUQ

ð19:1Þ

Thereby, the renewable surcharge satisfies Eq. (19.2), which means that the surcharge should be levied depends on the subsidy rate, the tax rate, as well as the proportion of renewables. TN 5 βUλ 2 ð1 2 λÞUTE

ð19:2Þ

The renewable surcharge (TN) increases if the renewable energy subsidy rate (β) increases or the tax rate (TE) decreases. In addition, the required renewable surcharge also increases when the proportion of renewable energy generation (λ) increases. This well reflects the trouble faced by China and other countries when the implementation of FIT policy effectively promotes the renewables and greatly raises this proportion. With higher proportion of renewables but unchanged renewable subsidy and environmental tax rate, the renewable surcharge needs to be continuously raised to meet the balance, which will increase burden to the economy. Therefore, for the sustainability of renewable policies, it implies that the policy needs to be adjusted by reducing the subsidy rate or raising the environmental tax rate.

19.3.1.2 Electricity generation enterprises In the second stage, it is assumed that the electricity generation enterprises will make optimal decisions by maximizing their profits. For simplicity, we assume that the total electricity generation equals the demand. If we denote the unit cost of electricity generation from renewables and fossil energy as

Policies for a sustainable energy future: how do renewable Chapter | 19

579

CN and CC, respectively, then the cost difference between renewables and fossil energy can be denoted as ΔC 5 CN 2 CC. As revealed by the literature, the relatively high cost of renewables can be reduced through technological progress driven by R&D investment. Thus, the cost difference is assumed to be reduced by R&D investment represented by M per unit of output, that is @ΔC @M , 0. Simultaneously, the cost of renewables can also be reduced through learning by doing. Thereby, the relative cost is assumed to be lower with C higher proportion of renewables, that is @Δ @λ , 0. To be specific, we take the most popular way of using a power function to denote these effects [64]. It is assumed that ΔC 5 ΔUM 2εM  λ2ελ , where Δ represents the basic cost difference, and εM and ελ are the efficiency parameters of the technological progress caused by R&D investment and learning by doing, representatively, satisfying εM $ 0 and ελ $ 0. Thereafter, as the proportion of renewable energy generation is λ, the total cost TC of electricity generation enterprises can be calculated by: TC 5 λUQUCN 1 ð1 2 λÞUQUðCC 1 TE Þ 1 MUQ 5 QUCC 1 λUQUΔC 1 ð1 2 λÞUQUTE 1 MUQ. Besides, the total revenue TR of the electricity generation enterprises can be calculated by: TR 5 λUQUðPC 1 βÞ 1 ð1 2 λÞUQUPC 5 QUPC 1 λUQUβ. The electricity generation enterprises are supposed to make optimal decisions by maximizing their profits: maxΠ 5 TR 2 TC 5 QUPC 1 λUQUβ 2 ½QUCC 1 λUQUΔUM 2εM Uλ2ελ λ;M

1 ð1 2 λÞUQUTE 1 MUQ

s:t: 0 # λ # λ

where λ is the assumed upper limit for the proportion of renewables. @Π By solving @M 5 0 we can get the optimal choice of R&D investment, that
 1  12ελ εM 11 . It implies that the optimal R&D investment for is M 5 εM UΔUλ renewables technology improvement depends on the two efficiency parameters, the basic cost difference, and the proportion of renewables. Since   1 2 ελ M @M @λ 5 εM 1 1 U λ , the relationship between the optimal R&D investment and the proportion of renewable energy generation is closely dependent on the  value of ελ . If ελ $ 1, then @M @λ # 0, which means that the optimal R&D investment would decrease with higher proportion of renewables. If 0 # ελ , 1, then  @M @λ . 0, which means that the optimal R&D investment would increase with higher proportion of renewables. This implies that the level of learning efficiency determines the growing of R&D investment. By strengthening management and improving learning efficiency, the efficiency of R&D investment can be effectively improved, which will help promote the R&D investment level. Furthermore, by solving @Π @λ 5 0 we can get the optimal proportion of M

2εεM1ε11 UΔε 1ε1 ε U2ε ε1ε  M λ M M λ M λ E renewables, that is λ 5 β1T . It indicates that the opti12ελ mal proportion of renewable energy generation depends on the efficiency

580

PART | V Policy

parameters of technological progress, the basic cost difference between renewables and fossil fuels in power generation, the subsidy rate, and the environmental tax rate. As we assumed 0 # λ # λ, it can be seen that @Π @λ . 0 holds for any λ whenever ελ $ 1; therefore, the optimal proportion of renewable energy generation is the upper bound. If 0 # ελ , 1, the optimal proportion of renewable energy generation is solved as:

2εεM1ε11 εM 1  2 M λ E λ 5 β1T ΔεM1ελ εM εM 1ελ . 12ελ

19.3.2 Discussion and policy implications The two-stage model provides insights for the design of renewable energy subsidy policies. This can be summarized from four aspects. 1. The design of renewable subsidy policy should balance the cost and the benefit. A basket of policies including both subsidies for the renewables and tax for the fossil fuels would help better to reduce the cost burden of renewable surcharges, especially with the increasing share of renewables in the total electricity generation. 2. The efficiency of technological progress resulting from learning by doing is the key parameter that determines both the optimal R&D investment strategy and the development of renewables. With a relatively higher efficiency, that is ελ $ 1, the electricity generation enterprises will have more motivation to pursue the upmost proportion of renewables con strained by the current circumstances. In the case λ 5 λ, the optimal

1  12ελ εM 11 R&D investment per unit of output is M 5 εM Δλ . 3. The renewable energy policy with FIT for renewables and tax for fossil energy could have its bottlenecks caused by low efficiency of technological progress. With relatively lower efficiency, that is 0 # ελ , 1, the electricity generation enterprises will choose the optimal proportion of renewables following the cost-benefit principle. When the marginal reve nue equals the marginal cost, the optimal proportion is solved as: λ 5 ε 11

2ε M1ε ε   1 2 M M λ β1TE εM 1 1 λ ΔεM 1ελ εM εM1ελ . In this case, since @λ 12ελ @β 5 2 εM 1 ελ β 1 TE # 0 



εM 1 1 λ and @λ @TE 5 2 εM 1 ελ β 1 TE # 0, the optimal proportion of renewables will decrease rather than increase with higher subsidy rate or higher environmental tax rate. 4. With relatively lower efficiency caused by learning by doing effect, the efficiency of technological progress caused by R&D investment matters for theh promotion When 0 # ελ , 1, we can iderive that of renewables.



@λ @εM 

@λ @ελ

5

5

h

1 2 ελ ðεM 1ελ Þ2

ln

β 1 TE 1 2 ελ

εM 1 1 ðε M 1 ε λ Þð1 2 ε λ Þ



ελ 1 lnΔ 2 εM 1 and ελ 2 ðεM 1ελ Þ2 lnεM λ

i  1 TE 11 εM 1 1 ðεεM1ε ln β1 2 ελ 2 ðε 1ε Þ2 lnΔ 2 ðε 1ε Þ2 lnεM λ . Þ2

2 ðε

M

1

M 1ελ Þ

λ

2

M

λ

M

λ

Policies for a sustainable energy future: how do renewable Chapter | 19 

Hereby,

@λ @εM

ελ 1 ελ lnεM

. 0 and and

581



TE . 0 holds for ð1 2 ελ Þln β1 1 2 ελ . lnΔ 1 εM 1

ðεM 1 1ÞðεM 1 ελ Þ β 1 TE 1 ðε 1 1Þln M ð1 2 ε λ Þ 1 2 ελ . lnΔ 1 εM lnεM , 

@λ @ελ

respectively. It implies that only with the promotion of both efficiency of R&D investment and learning can we effectively promote the development of renewables. Following the rules, we can easily understand why the renewables development paths are different for different countries with suitable policies. For example, for the FIT scheme, the renewable subsidy rate and surcharge are different, according to the different costs in different countries for different technologies. Thereby, the advice for the general rule of the renewable energy subsidy policy design is to be adaptive and systematic. This can be clarified from following aspects. First, the renewable energy subsidy rate should be appropriately reduced with the development of renewable energy technologies. This will provide effective incentives for the enterprises to promote renewables, as well as reduce the cost burden for the end users. Second, the subsidy for the renewables should be carried out together with the tax policy on fossil energy. Particularly, as the energy subsidy on traditional fossil energy still accounts for a large amount in a majority of countries, policies should be coordinated with each other. For example, as Li et al. [43] suggested, an integrated policy of increasing clean energy subsidies and reducing fossil energy subsidies would be more effective to bring co-benefits for the environment and the economy. Thereby, better effects would be achieved by phasing out fossil energy subsidy or increasing tax on fossil energy while providing renewable subsidies. Third, the renewable energy subsidy policy might bring distortions to the market by providing wrong incentives. Finally, the relationship between renewable energy subsidy and renewable energy development is non-linear. How renewable energy subsidy can effectively promote the development of renewables depends on various institutional parameters and technological parameters. Thereby, a carefully designed and dynamically evolved policy is suggested to outperform a rigid one. Actually the successful cases in Germany and China well prove so.

19.4 Conclusions The development of renewable energy provides an effective way to reduce carbon emissions and mitigate climate change. As market itself failing to provide enough renewables as expected due to externality problems, policies amend it by reshaping the incentives for the agents in the economy. Amongst the policies, the renewable energy subsidy effectively promoted the development of renewables. Specifically, the FIT policy has been proved to

582

PART | V Policy

be successful in many countries, such as Germany, Denmark, Spain and China. There are huge amount of studies on the impacts of renewable energy policies, and the results depend heavily on the evaluation method, the date sample as well as the variables and indicators. Despite there are critiques for the renewable energy subsidy policy, especially for the FIT scheme, we cannot deny its positive role in promoting the development of renewables. However, with the renewable energy generation growing fast, the cost burden of the subsidy policy has attracted more attention. A sustainable policy leads a sustainable renewable energy future. Based on the two-stage behaviour model, we derive several principles for the renewable subsidy policy. The core principle is that the renewable subsidy policy should be adaptive and systematic, accordingly with the development of renewable technologies and other energy related policies. The renewable subsidy policies across the world are now under the pressure of change. For example, China is stepping towards reducing the renewable subsidies and improving the market power in determining the development of renewables, and a basket of energy policies are implemented and reformed, including the national-wide carbon emission trading market, the reform of fossil energy resource tax, and gradually decreasing or removing the renewable subsidies. With a careful design and evaluation, these policies are supposed to work together to shape a bright future.

References [1] Dincer I, Acar C. Smart energy systems for a sustainable future. Appl Energy 2017;194:225
35. [2] Østergaard PA, Duic N, Noorollahi Y, Mikulcic H, Kalogirou S. Sustainable development using renewable energy technology. Renew Energy 2020;146:2430
7. [3] Ydersbond IM, Korsnes MS. What drives investment in wind energy? A comparative study of China and the European Union. Energy Res Soc Sci 2016;12:50
61. [4] Duscha V, Fougeyrollas A, Nathani C, Pfaff M, Ragwitz M, Resch G, et al. Renewable energy deployment in Europe up to 2030 and the aim of a triple dividend. Energy Policy 2016;95:314
23. [5] REN21. Renewables 2019 Global Status Report. Paris: REN21 Secretariat; 2019. [6] He Y, Pang Y, Shu H. The synergy mechanism of promoting renewable energy consumption in China. Energy Procedia 2017;105:327
34. [7] Fischer C, Newell RG. Environmental and technology policies for climate mitigation. J Environ Econ Manag 2008;55(2):142
62. [8] Kalkuhl M, Edenhofer O, Lessmann K. Renewable energy subsidies: second-best policy or fatal aberration for mitigation? Resour Energy Econ 2013;35:217
34. [9] Ackermann T, Andersson G, So¨der L. Overview of government and market driven programs for the promotion of renewable power generation. Renew Energy 2001;22(1
3):197
204.

Policies for a sustainable energy future: how do renewable Chapter | 19

583

[10] Newell RG, Pizer WA, Raimi D. U.S. federal government subsidies for clean energy: design choices and implications. Energy Econ 2019;80:831
41. [11] Morris J, Reilly JM, Paltsev S. Combining a renewable portfolio standard with a cap-andtrade policy: a general equilibrium analysis. MIT Joint Program on the Science and Policy of Global Change, Report No. 187; 2010. [12] Palmer K, Burtraw D. Cost-effectiveness of renewable electricity policies. Energy Econ 2005;27:873
94. [13] Lapan H, Moschini G. Second-best biofuel policies and the welfare effects of quantity mandates and subsidies. J Environ Econ Manag 2012;63:224
41. [14] Doherty R, O’Malley M. The efficiency of Ireland’s renewable energy feed-in tariff (REFIT) for wind generation. Energy Policy 2011;39(9):4911
19. [15] Choi G, Huh S-Y, Heo E, Lee C-Y. Prices versus quantities: comparing economic efficiency of feed-in tariff and renewable portfolio standard in promoting renewable electricity generation. Energy Policy 2018;113:239
48. ´ lvarez MT, Cabeza-Garc´ıa L, Soares I. Analysis of the promotion of onshore [16] Garc´ıa-A wind energy in the EU: feed-in tariff or renewable portfolio standard? Renew Energy 2017;111:256
64. [17] Sun P, Nie P. A comparative study of feed-in tariff and renewable portfolio standard policy in renewable energy industry. Renew Energy 2015;74:255
62. [18] Zhang Q, Wang G, Li Y, Li H, McLellan B, Chen S. Substitution effect of renewable portfolio standards and renewable energy certificate trading for feed-in tariff. Appl energy 2018;227:426
35. [19] Lehmann P. Supplementing an emissions tax by a feed-in tariff for renewable electricity to address learning spillovers. Energy Policy 2013;61:635
41. [20] Smith MG, Urpelainen J. The effect of feed-in tariffs on renewable electricity generation: an instrumental variables approach. Environ Resour Econ 2014;57(3):367
92. [21] Kilinc-Ata N. The evaluation of renewable energy policies across EU countries and US states: an econometric approach. Energy Sustain Dev 2016;31:83
90. [22] Dong C. Feed-in tariff vs. renewable portfolio standard: an empirical test of their relative effectiveness in promoting wind capacity development. Energy Policy 2012;42:476
85. [23] Couture TD, Jacobs D, Rickerson W, Healey V. Next generation of renewable electricity policy: how rapid change is breaking down conventional policy categories. Golden, CO (United States): National Renewable Energy Lab (NREL); 2015. [24] Garrett-Peltier H. Green versus brown: comparing the employment impacts of energy efficiency, renewable energy, and fossil fuels using an input
output model. Econ Model 2017;61:439
47. [25] Guenther-Lu¨bbers W, Bergmann H, Theuvsen L. Potential analysis of the biogas production
as measured by effects of added value and employment. J Clean Prod 2016;129:556
64. [26] Tourkolias C, Mirasgedis S. Quantification and monetization of employment benefits associated with renewable energy technologies in Greece. Renew Sustain Energy Rev 2011;15(6):2876
86. [27] Behrens P, Rodrigues JFD, Br´as T, Silva C. Environmental, economic, and social impacts of feed-in tariffs: a Portuguese perspective 2000
2010. Appl Energy 2016;173:309
19. [28] Bo¨hringer C, Rivers NJ, Rutherford TF, Wigle R. Green jobs and renewable electricity policies: employment impacts of Ontario’s feed-in tariff. BE J Econ Anal Policy 2012;12 (1) Article 25.

584

PART | V Policy

[29] Tabatabaei SM, Hadian E, Marzban H, Zibaei M. Economic, welfare and environmental impact of feed-in tariff policy: a case study in Iran. Energy Policy 2017;102:164
9. [30] Chatri F, Yahoo M, Othman J. The economic effects of renewable energy expansion in the electricity sector: a CGE analysis for Malaysia. Renew Sustain Energy Rev 2018;95:203
16. [31] Gan L, Eskeland GS, Kolshus HH. Green electricity market development: lessons from Europe and the US. Energy Policy 2007;35(1):144
55. [32] Menz FC. Green electricity policies in the United States: case study. Energy Policy 2005;33(18):2398
410. [33] Barbose G, Wiser R, Heeter J, Mai T, Bird L, Bolinger M, et al. A retrospective analysis of benefits and impacts of U.S. renewable portfolio standards. Energy Policy 2016;96:645
80. [34] Bhattacharya S, Giannakas K, Schoengold K. Market and welfare effects of renewable portfolio standards in United States electricity markets. Energy Econ 2017;64:384
401. [35] Murray BC, Cropper ML, de la Chesnaye FC, Reilly JM. How effective are US renewable energy subsidies in cutting greenhouse gases? Am Econ Rev 2014;104(5):569
74. [36] Union E. Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC. Off J Eur Union 2009;5:2009. [37] Haas R, Panzer C, Resch G, Ragwitz M, Reece G, Held A. A historical review of promotion strategies for electricity from renewable energy sources in EU countries. Renew Sustain Energy Rev 2011;15:1003
34. [38] Ragwitz M, Schade W, Breitschopf B, Walz R, Helfrich N, Fraunhofer I, et al. The impact of renewable energy policy on economic growth and employment in the European Union. Brussels, Belgium: European Commission, DG Energy and Transport; 2009. [39] Boeters S, Koornneef J. Supply of renewable energy sources and the cost of EU climate policy. Energy Econ 2011;33:1024
34. [40] Nicolini M, Tavoni M. Are renewable energy subsidies effective? Evidence from Europe. Renew Sustain Energy Rev 2017;74:412
23. [41] Pablo-Romero M d P, S´anchez-Braza A, Salvador-Ponce J, S´anchez-Labrador N. An overview of feed-in tariffs, premiums and tenders to promote electricity from biogas in the EU-28. Renew Sustain Energy Rev 2017;73:1366
79. [42] Batlle C, Pe´rez-Arriaga IJ, Zambrano-Barrag´an P. Regulatory design for RES-E support mechanisms: learning curves, market structure, and burden-sharing. Energy Policy 2012;41:212
20. [43] Li H, Bao Q, Ren X, Xie Y, Ren J, Yang Y. Reducing rebound effect through fossil subsidies reform: a comprehensive evaluation in China. J Clean Prod 2017;141:305
14. [44] Jenner S, Groba F, Indvik J. Assessing the strength and effectiveness of renewable electricity feed-in tariffs in European Union countries. Energy Policy 2013;52:385
401. [45] Campoccia A, Dusonchet L, Telaretti E, Zizzo G. An analysis of feed’ in tariffs for solar PV in six representative countries of the European Union. Sol Energy 2014;107:530
42. [46] Leiren MD, Reimer I. Historical institutionalist perspective on the shift from feed-in tariffs towards auctioning in German renewable energy policy. Energy Res Soc Sci 2018;43:33
40. [47] Bu¨sgen U, Du¨rrschmidt W. The expansion of electricity generation from renewable energies in Germany: a review based on the Renewable Energy Sources Act Progress Report 2007 and the new German feed-in legislation. Energy Policy 2009;37(7):2536
45.

Policies for a sustainable energy future: how do renewable Chapter | 19

585

[48] Hitaj C, Lo¨schel A. The impact of a feed-in tariff on wind power development in Germany. Resour Energy Econ 2019;57:18
35. [49] Winter S, Schlesewsky L. The German feed-in tariff revisited
an empirical investigation on its distributional effects. Energy Policy 2019;132:344
56. [50] Bo¨hringer C, Cuntz A, Harhoff D, Asane-Otoo E. The impact of the German feed-in-tariff scheme on innovation: evidence based on patent filings in renewable energy technologies. Energy Econ 2017;67:545
53. [51] Schallenberg-Rodriguez J, Haas R. Fixed feed-in tariff versus premium: a review of the current Spanish system. Renew Sustain Energy Rev 2012;16:293
305. [52] De Miera GS, del R´ıo Gonz´alez P, Vizca´ıno I. Analysing the impact of renewable electricity support schemes on power prices: the case of wind electricity in Spain. Energy Policy 2008;36(9):3345
59. [53] Bean P, Blazquez J, Nezamuddin N. Assessing the cost of renewable energy policy options
a Spanish wind case study. Renew Energy 2017;103:180
6. [54] Ciarreta A, Espinosa MP, Pizarro-Irizar C. Optimal regulation of renewable energy: a comparison of feed-in tariffs and tradable green certificates in the Spanish electricity system. Energy Econ 2017;67:387
99. [55] Pyrgou A, Kylili A, Fokaides PA. The future of the feed-in tariff (FiT) scheme in Europe: the case of photovoltaics. Energy Policy 2016;95:94
102. [56] Lipp J. Lessons for effective renewable electricity policy from Denmark, Germany and the United Kingdom. Energy Policy 2007;35(11):5481
95. [57] Meyer NI, Koefoed AL. Danish energy reform: policy implications for renewables. Energy Policy 2003;31(7):597
607. [58] NDRC. Notice on enhancing the feed-in tariff policy of wind power generation. National Development and Reform Commission, Price [2009] No. 1906; 2009. ,https://www.ndrc. gov.cn/xxgk/zcfb/tz/200907/t20090727_965206.html.. [59] NDRC. Notice on exerting price mechanism to promote the development of photovoltaic industry. National Development and Reform Commission, Price [2013] No. 1638; 2013. ,https://www.ndrc.gov.cn/xxgk/zcfb/tz/201308/t20130830_963934.html.. [60] Lin B, Moubarak M. Renewable energy consumption
economic growth nexus for China. Renew Sustain Energy Rev 2014;40:111
17. [61] Mu Y, Cai W, Evans S, Wang C, Roland-Holst D. Employment impacts of renewable energy policies in China: a decomposition analysis based on a CGE modeling framework. Appl energy 2018;210:256
67. [62] Hui J, Cai W, Ye M, Wang C. Clean generation technologies in Chinese power sector: penetration thresholds and supporting policies. Energy Procedia 2015;75:2807
12. [63] Wei W, Zhao Y, Wang J, Song M. The environmental benefits and economic impacts of fit-in-tariff in China. Renew Energy 2019;133:401
10. [64] Wang R, Leuthold F. Feed-in tariffs for photovoltaics: learning by doing in Germany? Appl Energy 2011;88:4387
99.

This page intentionally left blank

Chapter 20

Renewable energy-based power generation and the contribution to economic growth: the case of Portugal Joa˜o Paulo Cerdeira Bento GOVCOPP - Research Unit in Governance, Competitiveness and Public Policy, Department of Economics, Management, Industrial Engineering and Tourism, University of Aveiro, Campus of Santiago, Aveiro, Portugal

Chapter Outline 20.1 Introduction 20.2 Methodology 20.2.1 Econometric model and data 20.2.2 Testing for unit roots and detecting outliers

587 589 589 591

20.2.3 Testing for cointegration and estimating parameters 20.3 Empirical results 20.4 Conclusions References

591 592 599 606

20.1 Introduction The European Union (EU) has set itself long-term objectives consisting of decarbonizing the energy system, increasing the share of renewable energy and efficient energy use. The European approach of a common energy market is expected to result in lower costs, increasing competitiveness and more secure energy supplies by reducing the dependence on imported fossil fuels for power generation. For example the Energy Roadmap 2050 is compatible with the EU long-term goal of reducing greenhouse gas emissions by 80%
95%, when compared to 1990 levels. The purpose of this chapter is to contribute to the understanding of the long-term benefits of renewable energy-based power generation in the EU. The EU gets more than 30% of its electricity from renewable sources, up from 12% in 2000, against a target of 20% for 2020. However, the speed Renewable-Energy-Driven Future. DOI: https://doi.org/10.1016/B978-0-12-820539-6.00020-0 © 2021 Elsevier Inc. All rights reserved.

587

588

PART | V Policy

at which the renewable energy share is growing has slowed down since 2014. At the current rate of growth, the EU can reach a share of renewables in its electricity mix of 50% by 2030, but a sustained effort will be required to meet 2050 targets. In this context, it is important to follow future prospects and trends of renewable energy sources in the electricity sector. Although renewable energy technologies have reached a point of maturation and costs have declined sharply, their rapid development poses new challenges to the energy system. The framework for future energy will take into account the international situation and expected developments in the long run in a changing and challenging energy landscape. It is expected that energy demand will increase worldwide due to electric energy consumption, and this rise will have to be covered by existing green technologies, as well as new developments in renewable energy and technologies for renewable storage. The 2030 EU framework for climate and energy sets the promotion of renewable sources of energy as one of its top targets to help its member states achieve a more competitive, secure and sustainable energy system and to meet its long-term greenhouse gas mitigation and decarbonization targets by 2050. This strategy includes policies aimed at increasing the share of energy from renewable sources such as solar, wind, hydro, biomass and geothermal. In compliance with EU legislation and Directives, renewable energy sources have been substantially promoted mainly not only through the use of subsidies in the form of feed-in tariffs, green markets and green certificates but also through tax exemptions or deductions. The purpose of this chapter is to contribute to the understanding of the long-term benefits of renewable energy-based power generation in the EU by taking into consideration the case of Portugal. Portugal, which previously relied heavily on coal and natural gas, has shown rapid progress in converting to renewable electricity and increasing its use of solar, wind and hydro power. This achievement was the result of the combination of different policy instruments in the Portuguese renewable energy market. Hence it is key to focus on the feasibility of policies because the participation of several levels of government with potentially conflicting objectives is one of the most important obstacles to the promotion of renewable sources of energy. One example can be reported in this respect. In the wake of the financial crisis of 2008 and the austerity imposed on Portugal by the EU, the country lost access to international bond markets, bailout measures were adopted, leading to a debt crisis and cuts in public expenditures deferring feed-in tariffs and tax credits, hence challenging environmental, economic and social impacts of clean energy [1]. This reiterates the need for additional research on the energy and economic growth relationship in Portugal for at least two contribution motivations. First, rather than just presenting cointegration and Granger causality results, it is important to actually calculate the magnitude, sign and statistical

Renewable energy-based power generation Chapter | 20

589

significance of the cointegrating form and long-run parameters from this relationship. Second, it is central to assess how multiple structural breaks affect this relationship to overcome the omitted variable bias that plagued earlier studies. This uncertainty is a catalyst for this study and therefore it uses methods that lend greater insight into the nature of the energy
growth nexus by taking into account possible outliers in the econometric analysis to provide more robust estimates than the conventional methods used so far in the literature. The existence of multiple structural breaks not only affects the unit root properties of energy variables but also the cointegration relationship among variables, and ignoring the existence of structural breaks can introduce biases in estimation [2]. Contrary to the existing empirical studies on the energy
growth relationship, this study examines this relationship in the case of Portugal by including multiple structural breaks related to some episodes of financial crisis over the period 1970
2014. As shown in Table 20.1, Granger causality and long-run estimations in the case of Portugal validate the growth hypothesis [3,8], while most studies consistently confirm the feedback hypothesis [4
7,10,11,13]. In this literature energy variables share a long-run relationship with nonenergy variables, and empirical studies almost always report cointegration and Granger causality results. However, only a small number of studies have estimated long-run parameters, albeit with mixed results. A study infers that energy consumption has a negative effect on economic growth [9] while the deployment of renewable energy has a positive impact on economic growth [12]. In this study, time-series data for the period from 1970 to 2014 is used to empirically investigate the energy
growth relationship in the case of Portugal, namely to test for cointegration and causality, and to estimate short- and long-run elasticities. The results are obtained from a supply-side model with structural breaks coupled with several episodes of energy and financial crisis. The way in which these events interact with the energy
growth nexus is of interest to policy makers and often ignored in the energy economics literature. Moreover Portugal’s experience is almost unique within the EU. When the financial crisis struck the EU in the second half of 2008, Portugal obtained a bailout and engaged in many economic and fiscal reforms. The economic crisis has prompted an immediate response by governments and harsh austerity measures constrain and cut-off investments in renewable energy.

20.2 Methodology 20.2.1 Econometric model and data This study uses a production function approach and draws upon the empirical model proposed by [14]. The model adopts the form:

590

PART | V Policy

TABLE 20.1 Selected empirical studies on the energy
growth nexus in the case of Portugal. Study

Time period

Explanatory variables

Key findings

Single-country [3]

1971
2009

Per capita electricity consumption, employment, per capita income

E2Y

[4]

1965
2009

Income, energy consumption

E2Y

[5]

1974
2008

Electricity consumption, income, price of energy, employment

E2Y

[4]

1965
2009

Income, oil consumption crude oil prices

E2Y

[6]

1974
2009

Electricity consumption, income, price of energy and nonenergy goods

E2Y

[7]

2007
12

Industrial production, electricity generation regimes, electricity imports

E2Y

Multi-country [8]

1960
2002

Electricity consumption, income

E-Y

[9]

1980
2006

Energy consumption, income

Negative effect of E on Y

[10]

1990
2007

Income, gross fixed capital formation, labour force, renewable and nonrenewable electricity production

RE2Y; NRE2Y

[11]

1980
2010

Electricity consumption, income, net electricity imports

E2Y

[12]

1991
2012

Income, gross fixed capital formation, labour force, renewable and nonrenewable energy consumption

Positive effect of RE on Y

Notes: The symbols -, 2, and 6¼ represent causality in one direction, in two directions and absence of causality, respectively. E, total electricity or energy; Y, income; RE, renewable electricity or energy; NRE, nonrenewable electricity or energy.

Y 5 f ðK; L; NRE; RE; FDÞ

ð20:1Þ

where Y is the level of income, measured by the real gross domestic product at constant national prices (in millions of 2011 U.S. dollars); K is the capital stock at constant national prices (in millions of constant 2005 U.S. dollars); L is the labour force (number of persons engaged in millions); NRE is the electricity produced from nonrenewable sources of energy such as oil, gas and coal (in megawatt hours); RE is the electricity

Renewable energy-based power generation Chapter | 20

591

produced from renewable sources of energy such as solar, wind and hydroelectric (in megawatt hours); and FD stands for financial crisis dynamics proxied by various dummy variables that take the value 1 in periods of financial crisis and 0 otherwise. The financial crisis episodes are taken from Reinhart and Rogoff [15] and include currency crisis (1981
84 and 2005), inflation crisis (1974 and 1982
84), stock market crashes (1974, 1978, 1980, 1983, 1988, 1990
92 and 2000-02) and banking crisis (2008
14). The nonenergy variables are national-account based and obtained from the Penn World Table PWT 9.0 [16]. The energy variables are retrieved from the World Bank World Development Indicators. This study uses a sample that covers the period 1970
2014 and all data are log-transformed to make skewed distributions less skewed. The time period is justified by the availability of the data and particularly the data on economic crisis as shown in Appendix 1.

20.2.2 Testing for unit roots and detecting outliers Conventional unit root tests are biased toward a false unit root null when the data contain a structural break point. Hence it is important to identify potential “additive” or “innovative” outliers by modifying the Augmented Dickey
Fuller test [17]. The “additive” outlier model allows for a sudden change in mean while the “innovative” outlier model assumes more gradual changes. The break date selection is obtained by minimizing the Dickey
Fuller t-statistic over all possible break dates. These unit root tests are consistent whether there is a break or not, invariant to the break parameters so that the test performance does not depend on the magnitude of the break [18]. This approach captures the single most significant break in the data where breaks are treated as exogenous events and not correlated with the data. If the break date is correlated with the data, the test might indicate a break even though no break exists [19]. The single break approach is also preferred over testing for multiple structural breaks because of the distortions in both the size and power of these tests in smaller samples [20].

20.2.3 Testing for cointegration and estimating parameters The magnitude, sign and significance of the energy elasticities with respect to income are calculated through the Autoregressive Distributed Lag (ARDL) model [21]. The ARDL approach to cointegration has a superior small-sample performance and can be implemented no matter if the involved variables are integrated of order zero or one, but not of higher order [22]. The ARDL model used to test for cointegration is given by the following equation:

592

PART | V Policy

ΔlnYt 5 α1 1 β 1 lnYt-1 1 β 2 lnKt-1 1 β 3 lnLt-1 1 β 4 lnNREt-1 1 β 5 lnREt-1 k k k k X X X X 1 β 6i ΔlnYt2i 1 β 7i ΔlnKt2i 1 β 8i ΔlnLt2i 1 β 9i ΔlnNREt2i j51

j50

1

j50

k X

j50

β 10i ΔlnREt2i 1 ηi FD 1 εt

j51

ð20:2Þ where Δ is the first difference operator, ln denotes the natural logarithm and k is the lag order selected by Akaike Information Criterion. The residuals are assumed to be normally distributed and white noise. An F-test ascertains the joint significance of the subset of coefficients of the lagged level variables. The null hypothesis of having no cointegration is tested against the alternative hypothesis of cointegration. If the computed F-statistic is greater than the upper bound critical value, then the null hypothesis is rejected, hence the variables are cointegrated. If cointegration is established, the conditional ARDL model can be estimated to obtain the long-run coefficients through the following equation: ΔlnYt 5 α2 1

k X

δ1i ΔlnYt2i 1

k X

j51

δ1i ΔlnKt2i 1

j50

k X

δ2i ΔlnLt2i

j50

k k X X 1 δ3i ΔlnNREt2i 1 δ4j ΔlnREt2j 1 ηi FD 1 μt j50

ð20:3Þ

j50

The error correction or cointegrating form contains the short-run dynamics, where ψ is the coefficient of the error correction term (ECTt-1) obtained from the long run ARDL estimation and is expressed by the following equation: ΔlnYt 5 α3 1

k X

δ1i ΔlnYt2i 1

j51

1

k X j50

δ1i ΔlnKt2i 1

k X

δ2i ΔlnLt2i

j50

k k X X δ3i ΔlnNREt2i 1 δ4j ΔlnREt2j 1 ηi FD 1 ψECTt21 1 vt j50

ð20:4Þ

j50

To validate all coefficients of the variables, the error correction term has to be negative and statistically significant, and the ARDL model has to pass the residual diagnostics and stability check.

20.3 Empirical results The results of the unit root tests which allow for a structural break are reported in Table 20.2. In the case of the additive outlier model with

TABLE 20.2 Breakpoint unit root test results. Innovative outlier Variable

At level

Additive outlier

Break year

At first difference

Break year

At level

Break year

At first difference

Break year

1986

5.11

2001

2.35

1985

5.16

2001

Model with intercept Y K L NRE RE

3.64 3.67 3.21 3.18 

4.39

1985 2011 1978 2005



4.58



5.19



7.28



9.26

2005 2002 1977 1981

2.76 2.22 3.14 

4.53



2005



2002



1989



9.62

1980

2009

5.94

1987

2009



2009



2007

1985 1985 1978 2005

Significance level

1%

5%

10%

Critical values

4.94

4.44

4.19

4.58 5.32 7.72

Model with intercept and trend Y K L

3.18 2.11 2.13

2008

5.84

2009



2008

5.51



5.63

1986 2008 2008

3.73 2.12 3.21

2007

5.53 5.69

(Continued )

TABLE 20.2 (Continued) Innovative outlier

Additive outlier

Variable

At level

Break year

At first difference

Break year

At level

Break year

At first difference

Break year

NRE

2.38

2006

7.50

1999

2.66

2006

7.63

1984

RE

5.92

2005

9.23

1981

6.57

2009

9.57

1980

Significance level

1%

5%

10%

Critical values

5.35

4.86

4.61

Notes: The lag order is selected based on Akaike Information Criterion. Augmented Dickey
Fuller test statistic is reported for each variable in their level and in their differenced forms.  , Significance at the 5% and 10% levels, respectively. The break selection minimizes the t-statistic. K, capital stock at constant national prices; L, the labour force; Y, income; RE, renewable electricity or energy; NRE, nonrenewable electricity or energy.

Renewable energy-based power generation Chapter | 20

595

intercept and trend, they indicate that electricity generated from renewable energy sources is integrated of order zero since the calculated t-statistic for this variable in its level form is greater than the tabulated critical values. Otherwise the calculated t-statistics of the variables are less than the critical values in their level forms and greater than the critical values in their differenced forms, hence they become stationary after differencing suggesting that they are integrated of order one. The breakpoint unit roots tests detect various break points that coincide with the episodes of financial crisis in Portugal. The structural breaks identified do not coincide with world economic events such as the OPEC oil crisis 1973
74, the 1979
80 energy crisis and the 1997 Asian financial crisis. The inclusion of a dummy variable related to Portugal’s accession to the European Economic Community in 1986 and another dummy variable modelling the effect of the electric-connection with the Spanish grid in 2007 did not alter the cointegration test results reported in Table 20.3. Evidence of a long-run equilibrium relationship is found between income, capital stock, labour force, electricity from nonrenewables and renewables and financial crisis episodes. When all dummy variables are included in the production function, the calculated F-statistic is 11.67 and greater than the critical values of the top level of the bound. This is also the case for the remaining set of model specifications. The null hypothesis of no cointegration is rejected and the bounds cointegration test is regarded as conclusive since the computed F-statistic falls outside the upper bound critical values using

TABLE 20.3 Cointegration test results. Episodes of financial crisis

F-statistic

None

Currency

Inflation

Stock

Banking

All

9.81

9.32

18.41

7.95

8.44

11.67

Critical values Unrestricted intercept and no trend (unrestricted trend) Significance level (%)

1%

5%

10%

Lower bound

4.92 (5.95)

3.48 (4.33)

3.37 (3.63)

Upper bound

6.55 (7.60)

4.78 (5.63)

4.02 (4.79)

Notes: Critical values are calculated from Tuner [23] to address the short time-series.

596

PART | V Policy

conventional significance levels. The results imply that the variables are cointegrated and have similar stochastic trends. The estimated cointegrating form and long-run elasticities are reported in Table 20.4. All models are free of serial correlation, heteroscedastic and normally distributed at the default value of 5% statistical significance. The Jarque
Bera test indicates that the residuals are normally distributed. The serial correlation Lagrange Multiplier test cannot reject the null hypotheses of no serial correlation. There is also an absence of heteroskedasticity problems in the residuals. The Ramsey regression equation specification error tests pass at the statistical significance level of 1% and convey that all models have the correct functional form. The long-run parameters of the stock of capital, labour force, conventional and renewable electricity variables present all positive values which attain statistical significance at conventional levels. As expected, the financial crisis dummies have a negative impact upon income, and the results show statistically significant economic effects associated especially with inflation crisis. The comparison of the energy elasticities with respect to income reveals that their magnitudes are lower in the crisis models when compared to the noncrisis models although their signs and statistical significance remain unchanged. The results for model 1 indicate that a 1% increase in renewables and nonrenewables will enhance income by 0.05% and 0.03% in the short run and 0.07% and 0.04% in the long run, respectively. The results for model 6 shows instead that a 1% increase in renewables and nonrenewables will enhance income by 0.04% and 0.03% in the short run and 0.06% and 0.03% in the long run, respectively. The error correction terms in all models are statistically significant at the 1% significance level and present the correct negative sign. If the system is exposed to a shock, the speed of adjustment to the long-run equilibrium occurs at a relatively high convergence speed. In model 1, the coefficient of the error correction term suggests that the deviation of income from the short run to the long run is corrected by approximately 75% to each year. For example in model 6 the speed of adjustment parameter is 73%. Overall model comparison shows that the magnitude of the error correction mechanism is slightly higher in the noncrisis model. Governments typically take measures to mitigate adverse economic shocks on the economy to reduce the impact of crisis. This leads to a slowdown of the error correction mechanism. The time-series results in this study are very similar to those obtained with panel data by [12] and emphasize the importance of renewable energy to sustain long-term economic growth in Portugal. Similar results are found in earlier studies [12,24]. In addition, these findings also suggest that economic growth improves with the energy paradigm shift toward renewables despite the high costs of promoting renewables, such as increasing costs of electricity tariffs, placed upon the Portuguese economy [25].

TABLE 20.4 Cointegration regression results. Model

1

2

3

4

5

6

D (K)

3.29

3.10

2.35

3.28

3.25

2.29

D (L)

0.07

0.15

0.45

0.07

0.04

0.42

D (NRE)

0.05

0.05

0.04

0.05

0.05

0.04

D (RE)

0.03

0.02

0.02

0.03

0.03

0.02

Short-run elasticity

Currency

0.01

0.01 

Inflation

0.02

0.02

Stock

0.01

0.01

Banking

0.01 



0.01

The negative sign is missing in every reported coefficient for ECT(21) in all models. The first coefficient should be corrected as follows: 20.75 0.75

-0.74

-0.74

-0.75

-0.74

-0.73

K

0.61

0.59

0.57

0.60

0.62

0.58

L

0.09

0.18

0.21

0.10

0.06

0.18

ECT ( 2 1)





Long-run elasticity

NRE



0.07



0.06



0.06



0.07



0.07

0.06 (Continued )

TABLE 20.4 (Continued) Model

1

2 

RE

0.04

Currency

3 

0.03

4 

5 

0.03

0.04

6 

0.04

0.02

0.01 

Inflation

0.07

0.07

Stock

0.01

0.01

Banking 1.72

Intercept 2

Adjusted R

2.09

0.99

2.48

0.99 

1.72

0.99 

0.99 

0.01

1.59

2.36

0.99

1.58 (.21)

1.56 (.22)

0.17 (.69)

1.23 (.54)

1.58 (.45)

1.34 (.51)

1.07 (.59)

0.01 (.94)

1.93 (.16)

2.26 (.17)

0.06 (.81)

0.50 (.62)

0.21 (.65)

0.08 (.77)

0.33 (.57)

16.25

1.68 (.20)

0.63 (.43)

0.16 (.69)

Normality

1.64 (.44)

1.55 (.46)

ARCHc

1.83 (.17)

0.21 (.65)

d

0.11 (.74)

0.64 (.43)



0.99 17.69

24.92

19.48

0.01

16.99

15.94

F-statistic

0.03



Residual and stability diagnostics LMa b

RESET

Notes:  ,  ,   Significance at the 1%, 5% and 10% levels, respectively. The variables in parenthesis are expressed in their first differences. The probability values are reported in parenthesis. K, capital stock at constant national prices; L, the labour force; RE, renewable electricity or energy; NRE, nonrenewable electricity or energy a Breusch Godfrey Lagrange multiplier test for serial correlation. b Jarque
Bera skewness and Kurtosis test for normality. c Engel’s test for residual heteroscedasticity. d Ramsey regression equation specification error test.

Renewable energy-based power generation Chapter | 20

599

20.4 Conclusions This study sheds some light on the energy
growth relationship in Portugal by addressing the existence of multiple structural breaks linked to several episodes of financial crisis over the period 1970
2014. It estimates a supply-side model augmented with financial crisis episodes to analyze whether those episodes affect the statistical significance, sign and magnitude of electricity generated from renewable energy sources. This study has investigated the effect of nonrenewable and renewable electricity on economic growth in Portugal by taking into account several episodes of financial crisis during the period under scrutiny. The applied cointegration technique has confirmed the presence of cointegration among the variables. The findings emphasize the role of renewable electricity to sustain longterm economic growth. The short-run and long-run elasticities of renewable electricity do not diverge in terms of statistical significance and sign in the noncrisis and crisis models, but the elasticities tend to present lower magnitudes in the latter. The Granger causality results further confirm the cause
effect relationship among the renewable energy renewable and economic variables proposed in the study. Taken together, the findings highlight the need for energy policy instruments and promotional mechanisms for renewable energy in Portugal. In a time of crisis, policy makers have to take into account that managing a crisis goes beyond the immediate emergency response and that the EU energy-policy targets could be compromised in the long run.

Appendix 1 Table 20.A1

TABLE 20.A1 Dataset. Year

Y

K

L

NRE

RE

Currency

Inflation

Stock

Banking

All

1970

86356.5

319105.813

3.1494081

1,448,000

5,990,000

0

0

0

0

0

1971

92083.3672

342167.156

3.24261189

1,583,000

6,313,000

0

0

0

0

0

1972

99464.4844

371681.125

3.26619244

1,641,000

7,199,000

0

0

0

0

0

1973

110605.18

402907.031

3.28275871

2,267,000

7,525,000

0

0

0

0

0

1974

111869.242

429306.281

3.51809859

2,677,000

8,008,000

0

1

0

0

0

1975

107005.578

450871.531

3.61120987

4,055,000

6,646,000

0

0

0

0

0

1976

114389.211

471576.125

3.6314857

5,029,000

5,017,000

1

0

1

0

1

1977

120797.977

496147.531

3.68616104

3,575,000

10,200,000

1

0

0

0

0

1978

124199.609

524961.313

3.70971727

3,556,000

11,043,000

0

0

0

0

1

1979

131203.141

550040.313

3.78731775

4,644,000

11,471,000

0

0

0

0

1

1980

137224.5

579271.5

3.88523698

6,871,000

8,335,000

0

0

1

0

1

1981

139444.938

610654.938

3.90223575

8,494,000

5,330,000

1

0

0

0

0

1982

142422.609

643502.75

3.91585851

8,083,000

7,274,000

1

1

0

0

0

1983

142176.063

671489.563

4.10412407

9,622,000

8,431,000

1

1

1

0

1

1984

139503.172

688564.188

4.09560394

9,057,000

10,138,000

1

1

0

0

1

1985

143419.641

704991.125

4.07481909

7,569,000

11,241,000

0

0

1

0

1

1986

149358.594

727641.563

4.07641125

11,207,000

9,120,000

0

0

0

0

1

1987

158889.75

758756.563

4.17461014

10,336,000

9,767,000

0

0

0

0

0

1988

170789.172

797069.313

4.27509737

9,534,000

12,886,000

0

0

0

0

0

1989

181789.094

834856.688

4.36621046

19,067,000

6,482,000

0

0

0

0

0

1990

188970.703

874276.188

4.46493721

18,503,000

9,852,000

0

0

1

0

1

1991

197225.344

911345.938

4.60833549

19,881,000

9,858,000

0

0

0

0

0

1992

199374.063

947584.063

4.66151237

24,122,000

5,538,000

0

0

1

0

1

1993

195300.297

978158.875

4.56844425

21,552,000

9,455,000

0

0

1

0

1

1994

197184.625

1010301.63

4.55704021

19,695,000

11,643,000

0

0

1

0

1

1995

205629.609

1044371

4.53477907

23,764,000

9,390,000

0

0

0

0

0

1996

212819.828

1082328.5

4.61401939

18,633,000

15,792,000

0

0

0

0

0

1997

222239.688

1129462.5

4.72706079

19,906,000

14,232,000

0

0

0

0

0

1998

232888.906

1185198.5

4.84553051

24,760,000

14,153,000

0

0

0

0

0

1999

241944.125

1243701.63

4.91127396

34,214,000

8,650,000

0

0

0

0

0

2000

251107.75

1303029.75

5.0140276

30,247,000

12,868,000

0

0

0

0

0

2001

255987.531

1360037.63

5.09734917

30,172,000

15,741,000

0

0

0

0

0

2002

257955.547

1410249.25

5.11447191

35,656,000

9,733,000

0

0

1

0

1

2003

255545.719

1451436.75

5.06265926

28,540,000

17,703,000

0

0

1

0

1

2004

260175.141

1491903.13

5.02486706

32,245,000

12,314,000

0

0

1

0

1

2005

262170.125

1531547.25

4.9998703

37,623,000

8,260,000

1

0

0

0

0

(Continued )

TABLE 20.A1 (Continued) Year

Y

2006

266241.781

2007

K

L

NRE

RE

Currency

Inflation

Stock

Banking

All

1569157

5.01824331

32,548,000

15,722,000

0

0

0

0

1

272876.531

1608016.38

5.01775885

30,392,000

16,218,000

0

0

0

0

0

2008

273420.281

1645318.13

5.03281021

30,543,000

14,638,000

0

0

0

0

0

2009

265277.531

2010

270314.344

1672989.75

4.88738823

30,893,000

18,291,000

0

0

0

1

1

1698433.5

4.80284643

25,008,000

28,353,000

0

0

0

1

1

2011

265376.094

1712079.88

4.68732595

27,448,000

24,115,000

0

0

0

1

1

2012

254686.078

1713377.63

4.46822834

25,947,000

19,370,000

0

0

0

1

1

2013

251807.719

1712239

4.30996513

20,764,000

29,471,000

0

0

0

1

1

2014

254088.594

1713123.5

4.34134722

20,143,000

31,560,000

0

0

0

1

1

Year

Y

L

NRE

Currency

Inflation

Stock

Banking

All

1970

11.3662394

K 12.673278

1.14721453

14.1856939

RE 15.605602

0

0

0

0

0

1971

11.4304496

12.7430547

1.17637914

14.2748323

15.6581216

0

0

0

0

0

1972

11.5075559

12.8257916

1.18362491

14.3108164

15.7894527

0

0

0

0

0

1973

11.6137222

12.9064611

1.18868414

14.6339679

15.8337414

0

0

0

0

0

1974

11.625086

12.9699259

1.25792067

14.8002073

15.8959516

0

1

0

0

0

1975

11.5806362

13.0189377

1.28404286

15.2154612

15.7095257

0

0

0

0

0

1976

11.647362

13.0638358

1.28964185

15.4307317

15.4283427

1

0

1

0

1

1977

11.7018748

13.1146286

1.30458555

15.0894757

16.1378983

1

0

0

0

0

1978

11.7296453

13.1710798

1.31095567

15.0841469

16.2173073

0

0

0

0

1

1979

11.7845021

13.2177468

1.33165805

15.3510866

16.2553327

0

0

0

0

1

1980

11.8293735

13.2695266

1.35718398

15.7428202

15.9359741

0

0

1

0

1

1981

11.8454251

13.3222873

1.36154966

15.9548706

15.4888618

1

0

0

0

0

1982

11.866554

13.3746816

1.36503459

15.9052736

15.7998169

1

1

0

0

0

1983

11.8648214

13.4172538

1.41199234

16.0795627

15.9474259

1

1

1

0

1

1984

11.8458426

13.4423638

1.40991419

16.0190485

16.1318013

1

1

0

0

1

1985

11.8735302

13.4659405

1.40482635

15.8395715

16.2350784

0

0

1

0

1

1986

11.9141054

13.4975638

1.40521701

16.2320491

16.0259804

0

0

0

0

1

1987

11.9759658

13.5394363

1.42902097

16.1511435

16.0945199

0

0

0

0

0

(Continued )

TABLE 20.A1 (Continued) Year

Y

K

L

NRE

1988

12.0481852

13.5886969

1.45280688

16.0703749

1989

12.1106025

13.6350154

1.47389546

1990

12.1493473

13.6811516

1991

12.1921022

1992

12.2029381

1993

RE

Currency

Inflation

Stock

Banking

All

16.371652

0

0

0

0

0

16.7634697

15.6845397

0

0

0

0

0

1.49625515

16.7334434

16.103185

0

0

1

0

1

13.7226778

1.52786673

16.8052751

16.1037939

0

0

0

0

0

13.7616709

1.53933994

16.9986348

15.527144

0

0

1

0

1

12.1822936

13.7934274

1.51917272

16.8859792

16.0620543

0

0

1

0

1

1994

12.1918958

13.8257595

1.51667334

16.7958754

16.2702157

0

0

1

0

1

1995

12.2338318

13.8589253

1.51177637

16.9836824

16.0551559

0

0

0

0

0

1996

12.2682012

13.8946253

1.52909936

16.7404448

16.575014

0

0

0

0

0

1997

12.3115118

13.9372524

1.55330361

16.8065318

16.4710035

0

0

0

0

0

1998

12.3583168

13.9854208

1.57805674

17.02474

16.4654372

0

0

0

0

0

1999

12.3964621

14.0336027

1.59153337

17.3481455

15.9730699

0

0

0

0

0

2000

12.4336374

14.0802027

1.6122395

17.2249076

16.3702542

0

0

0

0

0

2001

12.452884

14.1230229

1.62872063

17.2224249

16.5717793

0

0

0

0

0

2002

12.4605426

14.159277

1.63207415

17.389428

16.0910327

0

0

1

0

1

2003

12.4511566

14.1880645

1.62189189

17.1668172

16.6892447

0

0

1

0

1

2004

12.4691103

14.2155631

1.614399

17.2888736

16.3262474

0

0

1

0

1

2005

12.4767489

14.2417891

1.60941197

17.4431261

15.9269351

1

0

0

0

0

2006

12.4921601

14.2660491

1.61307993

17.2982265

16.5705716

0

0

0

0

1

2007

12.5167747

14.2905119

1.61298339

17.22969

16.6016323

0

0

0

0

0

2008

12.5187654

14.3134443

1.61597852

17.2346461

16.4991314

0

0

0

0

0

2009

12.4885318

14.3301229

1.58665806

17.2460402

16.7219197

0

0

0

1

1

2010

12.5073408

14.3452169

1.56920875

17.0347063

17.1602434

0

0

0

1

1

2011

12.4889033

14.3532195

1.54486226

17.1278039

16.9983446

0

0

0

1

1

2012

12.447787

14.3539772

1.49699199

17.0715666

16.779236

0

0

0

1

1

2013

12.4364211

14.3533124

1.46092981

16.8487313

17.1989173

0

0

0

1

1

2014

12.4454383

14.3538289

1.46818472

16.8183674

17.2674011

0

0

0

1

1

Notes: Columns “Currency”, “Inflation”, “Stock”, “Banking” and “All” contain the dummy variables that take only the value 0 or 1 to indicate the absence or presence of crisis. Bold values correspond to impulse dummies that indicate the presence of crisis.

606

PART | V Policy

References [1] Behrens P, Rodrigues J, Br´as T, Silva C. Environmental, economic, and social impacts of feed-in tariffs: a Portuguese perspective 2000
2010. Appl Energy 2016;173:309
19. [2] Smyth R, Narayan P. Applied econometrics and implications for energy economics research. Energy Econ 2015;50:351
8. [3] Shahbaz M, Tang C, Shahbaz S. Electricity consumption and economic growth nexus in Portugal using cointegration and causality approaches. Energy Policy 2011;39 (6):3529
36. [4] Fuinhas J, Marques A. Energy consumption and economic growth nexus in Portugal, Italy, Greece, Spain and Turkey: an ARDL bounds test approach (1965
2009). Energy Econ 2012;34(2):511
17. [5] Tang C, Tan E. Electricity consumption and economic growth in Portugal: evidence from a multivariate framework analysis. Energy J 2012;33(4):23
48. [6] Tang C, Shahbaz M, Arouri M. Re-investigating the electricity consumption and economic growth nexus in Portugal. Energy Policy 2013;62:1515
24. [7] Marques AC, Fuinhas JA. The role of Portuguese electricity generation regimes and industrial production. Renew Sustain Energy Rev 2015;43:321
30. [8] Narayan P, Prasad A. Electricity consumption
real GDP causality nexus: evidence from a bootstrapped causality test for 30 OECD countries. Energy Policy 2008;36(2):910
18. [9] Narayan P, Popp S. The energy consumption-real GDP nexus revisited: empirical evidence from 93 countries. Econ Model 2012;29(2):303
8. [10] Apergis N, Payne J. Renewable and non-renewable energy consumption-growth nexus: Evidence from a panel error correction model. Energy Econ 2012;34(3):733
8. [11] Karanfil F, Li Y. Electricity consumption and economic growth: exploring panel-specific differences. Energy Policy 2015;82:264
77. [12] Bhattacharya M, Paramati S, Ozturk I, Bhattacharya S. The effect of renewable energy consumption on economic growth: Evidence from top 38 countries. Appl Energy 2016;162:733
41. [13] Fuinhas J, Marques A. An ARDL approach to the oil and growth nexus: Portuguese evidence. Energy Sources B: Econ Plan Policy 2012;7(3):282
91. [14] Apergis N, Payne J. Renewable energy consumption and economic growth: evidence from a panel of OECD countries. Energy Policy 2010;38(1):656
60. [15] Reinhart C, Rogoff K. This time is different: a panoramic view of eight centuries of financial crises. Ann Econ Finance Soc 2014;15(12):1065
188. [16] Feenstra R, Inklaar R, Timmer M. The next generation of the Penn World Table. Am Econ Rev 2015;105(10):3150
82. [17] Perron P. The Great Crash, the oil price shock, and the unit root hypothesis. Econometrica 1989;57(6):1361
401. [18] Vogelsang T, Perron P. Additional tests for a unit root allowing for a break in the trend function at an unknown time. Int Econ Rev 1998;39(4):1073
100. [19] Hansen B. The new econometrics of structural change: dating changes in U.S. labor productivity. J Econ Perspect 2001;15(4):117
28. [20] Bai J, Perron P. Multiple structural change models: a simulation analysis. Corbae D, Durlauf S, Hansen B, editors. Econometric theory and practice: frontier of analysis and applied research (essays in honour of Peter Phillips). Cambridge University Press; 2006.

Renewable energy-based power generation Chapter | 20

607

[21] Pesaran M, Shin Y, Smith R. Bounds testing approaches to the analysis of level relationships. J Appl Econom 2001;16(3):289
326. Available from: https://doi.org/10.1002/ jae.616. [22] Pesaran B, Pesaran M. Time series econometrics using Microfit 5.0. Oxford University Press; 2009. [23] Tuner, P. Response surfaces for an F-test for cointegration. App Econ Lett 2006;13 (8):479
82. [24] Marques A, Fuinhas J, Pereira A. On the dynamics of generating electricity from diversified sources: evidence from Portugal. Energy Environ 2016;26(4):587
600. [25] Marques AC, Fuinhas JA. Is renewable energy effective in promoting growth? Energy Policy 2012;46:434
42.

This page intentionally left blank

Index Note: Page numbers followed by “f” and “t” refer to figures and tables, respectively.

A a-Si. See Amorphous silicon (a-Si) ABE. See Acetone
butanol
ethanol (ABE) Abiotic depletion potential (ADP), 448 Absorption, 29, 135 AC. See Alternating current (AC) Acetone
butanol
ethanol (ABE), 55
56 ACF. See Autocorrelation function (ACF) Acid-producing bacteria, 47
48 Acidogenesis, 46, 55
56 Active system, 136 AD. See Adsorption desalination (AD); Anaerobic digestion (AD) ADP. See Abiotic depletion potential (ADP) Adsorption desalination (AD), 337
339 AEC. See Alkaline electrolysis (AEC) AGC systems. See Automatic Generation Control systems (AGC systems) Agricultural waste, 49
54 AHP. See Analytic hierarchy process (AHP) Air mass (AM), 4 Air pollutant emissions reduction, 271
273, 271f, 272f Akaike Information Criterion, 591
592 Alcohol fermentation, 57
58 wastewater, 51
52 Aldehydes, 45
46 Algae, SCWG of, 392
393 Alkaline electrolysis (AEC), 379
380 Alternating current (AC), 439 AM. See Air mass (AM) Ammonia (NH3), 436
437 production, 440
448 cost index, 449
450 effects of global warming, 456
462 effects of HT, 462
466 efficiency index, 448
449 environmental impact index, 450
452 effect of exergy efficiency, 456

sustainability index values of, 453 weighting scheme, 452 thermophysical properties of, 438t Amorphous silicon (a-Si), 297 Anaerobic digestion (AD), 142 Anaerobic digestion units, 49
51 Analytic hierarchy process (AHP), 84
85, 492 Aquifers, 136 ARDL model. See Autoregressive Distributed Lag model (ARDL model) ARIMA model. See Autoregressive integrated moving average model (ARIMA model) Autocorrelation function (ACF), 411 Automatic Generation Control systems (AGC systems), 522 Autoregressive Distributed Lag model (ARDL model), 591
592 Autoregressive integrated moving average model (ARIMA model), 410
411, 416 Auxiliary equations, 81t Awareness barriers, 547

B Balanced InDS, 194f, 195 Band gap energy, 7 Barriers identification, 544
550 awareness barriers, 547 economic barriers, 545
546 end-use/demand-side barriers, 549
550 financial barriers, 547 framework, 550
554 information barriers, 547 institutional and administrative barriers, 548 measures to overcome barriers, 554
555 regulatory and policy barriers, 548 social and environmental barriers, 548 technical barriers, 546
547 Batch fermentation, 54

609

610

Index

Batteries, 245
247, 246f BAU. See Business-as-usual scenario (BAU) Benders’ decomposition, 210
211 Benefit-type criteria, 495
496, 500
501 Binary cycle power plant, 141 Bio-oil reforming, 383 GHG footprints of, 397
398 inventory analysis of, 391
392 Bioalcohols, 48 fermentation, 54 Biobutanol fermentation, 43
44 Biodiesel, 143 Bioenergy for sustainability challenges, 61
62 future prospects, 62
63 properties of bioalcohols and gasoline, 44t technologies feedstocks, 48
54 fermentation technologies, 54
61 microorganisms, 45
48 Bioethanol, 143 Biofilm, 56
57 Biofuels, 142
144, 145f biodiesel, 143 bioethanol, 143 biogas, 144 biomethanol, 143 straight vegetable oils, 143 Biogas, 43
44, 62
63, 144 Biological conversion processes, 142 Biological digestion process, 62 Biomass, 43 biomass-fuelled CCHP systems, 144 energy, 142
144, 302 biofuels, 142
144 biomass-fuelled CCHP systems, 144 technologies, 142 generation, 249 to H2, 380
383 thermal gasification of, 386
391 transportation, 389
391 Biomethanol, 143 BIPV. See Building-integrated photovoltaics (BIPV) Boron (B), 8 Brackish water, 335 Brine, 335 production, 359 Building construction, 295 life-cycle systems, 292, 293f material extraction and transportation, 292
294

mix, 323 operation, 296
304 PV-T, 299
304 solar PV, 297
298 solar thermal, 298
299 Building-integrated photovoltaics (BIPV), 297
298 Business-as-usual scenario (BAU), 326 Butanol, 43 Butyraldehyde, 46
47 Butyryl phosphate, 46 Butyryl-CoA, 46

C Cadmium (Cd), 5, 13 Cadmium sulphide (CdS), 13
14 Cadmium telluride (CdTe), 9
10, 13
14, 14f, 297 Calcium titanate (CaTiO3), 11
12 Capacity barriers. See Technical barriers Capital recovery factor (CRF), 76
77 Carbon and air pollutant emissions reduction, 271
273 Carbon capture and storage (CCS), 258 Carbon dioxide (CO2), 142, 234
235, 377
378 emission, 23, 359
360 Carbon footprint analysis, 82
83 of ORC, 99
108 Carbon monoxide (CO), 377
378 CC. See Closeness coefficient (CC) CCHP system. See Combined cooling, heating and power system (CCHP system) CCP. See Chance-constrained programming (CCP) CCS. See Carbon capture and storage (CCS) CdS. See Cadmium sulphide (CdS) CdTe. See Cadmium telluride (CdTe) Cell immobilization, 56
57 Cellulose, 49 CES. See Constant elasticity of substitution (CES) CFCs. See Chlorofluorocarbons (CFCs) CGE model. See Computable general equilibrium model (CGE model) CGS. See Copper
gallium
diselenide (CGS) Chance-constrained programming (CCP), 208
210 Chemical conversion processes, 142 Chemical energy storage, 436
438 Chemical oxygen demand (COD), 52
54 Chemical storage, 137 Chlorofluorocarbons (CFCs), 126

Index CIGS. See Copper indium gallium selenide (CIGS) CIS. See Commonwealth of Independents States (CIS); Copper indium diselenide (CIS) Climate change, 233
238, 236f, 237f, 258 Closeness coefficient (CC), 481 Clostridia, 47
48 Clostridium, 46 Coal combustion, 247, 377
378 Coal phase-out, 247
248, 248f COD. See Chemical oxygen demand (COD) Coefficient of performance (COP), 24
25, 141, 320
321 Cogeneration, 126
134, 318, 320 combined cooling, heating and power operation strategies, 133 DESs and polygeneration microgrids, 132 distributed/decentralized energy system, 130
131 energy tools/software in energy systems, 133
134 microgeneration, 130 polygeneration, 130 Cointegration, 591
592 Combined cooling, heating and power system (CCHP system), 126 generation, 320 Combined heat and power systems (CHP systems). See Cogeneration Combined heat storage, 137 Combustion of coal, 295 Commercial software, 207 Commonwealth of Independents States (CIS), 239 Computable general equilibrium model (CGE model), 258
259, 566
567 Concentrating PV (CPV), 5
6 Concentrating solar collectors, 146 Concentrating solar plant (CSP), 147 CSP-geothermal hybrids, 152 Condensation process, 37 Condenser, 73 Constant elasticity of substitution (CES), 260 Constraints, 178 Continuous fermentation, 54
55 Continuous stirred tank reactor process (CSTR process), 49
51, 55f Conventional waste-driven CCHP, 151
152 Conversion coefficients of straw, 416
419 Conversion efficiency, 9
11 Cooling strategy, 133

611

Cooling technologies cooling applications in trigeneration systems, 135
136 types, 135 desiccant technology, 135 sorption technology, 135 COP. See Coefficient of performance (COP) Copper (Cu), 5 Copper indium diselenide (CIS), 9
10, 14 Copper indium gallium selenide (CIGS), 9
10, 14
15, 15f, 297 Copper indium gallium selenium. See Copper indium gallium selenide (CIGS) Copper zinc tin sulfide (CZTS), 12
13 Copper
gallium
diselenide (CGS), 14 Copper
indium
gallium diselenide. See Copper indium gallium selenide (CIGS) Cost index, 449
450 Cost-benefit analysis, 574 Cost-type criteria, 495
496, 500
501 CPV. See Concentrating PV (CPV) CRF. See Capital recovery factor (CRF) CSP. See Concentrating solar plant (CSP) CSTR process. See Continuous stirred tank reactor process (CSTR process) CZTS. See Copper zinc tin sulfide (CZTS)

D DC. See Direct current (DC) Death Spiral, 531 Decision-making, 84
85, 492 processes, 471 tools, 133
134 Defuzzification, 495 Dehumidification desalination, 34
35, 35f Demand response (DR), 253, 323 Demand-side management, 323
324 building mix, 323 demand response, 323 land use change, 323
324 Denmark, RE subsidy policies in, 574
575 DERs. See Distributed energy resources (DERs) Desalination direct type, 37
38 indirect type, 34
37 systems, 140 techniques, 335
345, 335f, 337f barriers, issues and opportunities, 357
360 energy and, 345
346

612

Index

Desalination (Continued) global status of, 340
342 membrane desalination techniques, 339
340 research trends in, 343
345 thermal, 337
339 Desiccant technology, 135 DESs. See District energy systems (DESs) DG system. See Distributed generation system (DG system) Diesel
mahua oil, 149
151 Digitalization and smart grids, 253 Direct current (DC), 26 Direct emissions, 82 Direct type desalination, 37
38. See also Indirect type desalination solar still desalination system, 38f Distributed energy resource management problem, 178
183 Distributed energy resources (DERs), 169
170, 191t, 213
214 Distributed generation system (DG system), 130 Distributed/decentralized energy system, 130
131 Distribution system (DS), 169
170 Distribution system operator (DSO), 172 District energy systems (DESs), 132, 132f District heating systems cascade and upgrade use of heat energy, 316
317 and cooling system, 312, 313f energy efficiency and exergy efficiency, 314
316 Divergence measures for FS, 475
479 measures-based fuzzy TOPSIS method, 480
485 Donghai Bridge Offshore Wind Farm, 301
302 Doping process, 8 Double flash power plant, 141 DR. See Demand response (DR) Dry steam power plant, 141 DS. See Distribution system (DS) DSO. See Distribution system operator (DSO) DSSCs. See Dye-sensitized solar cells (DSSCs) Dual-loop ORC, 68 Dual-pressure ORC system, 68, 118
119, 119f

Dye-sensitized solar cells (DSSCs), 13, 15
16, 16f, 297

E ECE. See Emission of carbon dioxide equivalent (ECE) Econometric model and data, 589
591 Economic analysis, 76
81 assessment of renewable energy, 260
261 barriers, 545
546 growth, 588
589 impacts of renewable energy development, 273
277 on industrial output, value-added and employment, 274
277 investment, 273
274 systems, 68 Economic performance design parameters effects on, 89
92 costs of components in basic ORC, 93f optimization working conditions, 91t total cost and electricity production cost, 92f sensitivity analysis on, 95
99 ECTs. See Energy conversion technologies (ECTs) ED. See Electrodialysis (ED) EDM. See Electrical demand management (EDM) EEG process. See Erneuerbare
Energien
Gesetz process (EEG process) Efficiency index, 448
449 energy efficiency, 448
449 exergy efficiency, 449 EHP. See Electron-hole pair (EHP) EJ. See Exajoules (EJ) Electric field, 10 Electric photo-generated load current, 9 Electric Regulation, 522 Electric vehicle (EV), 234 Electrical demand management (EDM), 133 Electricity, 234, 377
378 feed-in law, 573 generation enterprises, 578
580 Electricity production cost (EPC), 69 Electrochemical ammonia synthesis, 440
442, 445
448 Electrodialysis (ED), 339 Electrolysis, 379
380 Electrolytic process, 377
378

Index Electromagnetic waves (EM waves), 3
4 Electron transporting layer (ETL), 12 Electron volt (eV), 3
4 Electron-hole pair (EHP), 7 EM waves. See Electromagnetic waves (EM waves) Embden-Meyerhof-Parnas pathway, 45
46 Emission of carbon dioxide equivalent (ECE), 82 Emission reductions analysis, 105
106, 106f End-use/demand-side barriers, 549
550 Energy consumption, 311
312 conversion, 320
321 and desalination, 345
346 efficiency, 314
316, 448
449 energy potential of straw, 410
411 and environmental challenges, 289
290 gap, 7 sector, 234 sequence, 130 storage, 244
247, 435
436 batteries, 245
247 hydrogen, 247 pumped-storage hydropower, 245 system, 267
271 integration, 546
547 power structure, 270
271 primary energy, 269
270 tools/software in energy systems, 133
134 Energy conversion technologies (ECTs), 407
408, 419
421 Energy Policy Act, 570
571 Environmental benefits of straw, 421
424 Environmental evaluation of components, 103
104 emissions of CO2 equivalent of components, 104f of life cycle, 99
103 emissions of CO2 equivalent and corresponding proportions of components, 103f optimal parameters of basic ORC, 101t total emissions of CO2 equivalent, 102f of working fluids, 104
105 Environmental impact index, 450
452 EPC. See Electricity production cost (EPC) Erneuerbare
Energien
Gesetz process (EEG process), 573 Ethanol, 43 ETL. See Electron transporting layer (ETL)

613

ETSC. See Evacuated type tube solar collector (ETSC) EU. See European Union (EU) EU Emission Trading System (EUETS), 237
238 2030 European energy transition strategies coal phase-out, 247
248 decrease in renewable energy costs, 248
251 digitalization and smart grids, 253 DR, 253 international interconnections, 251
253 European Union (EU), 237, 587 Eutrophication potential, 69 EV. See Electric vehicle (EV) Evacuated type tube solar collector (ETSC), 20
21 Evaporation process, 37 Evaporator, 73 Exajoules (EJ), 376 Exergoeconomic(s), 68
69 analysis, 76
81 coefficients in equations, 80t life cycle boundary of an ORC system, 81f design parameters effects on performance, 92
95 Exergy cost rate balances, 81t efficiency, 314
316, 449 flow rates, 94t parameters, 96t rate balance, 74 Exponential entropy, 472
473 Extracellular enzymes, 47
48 Extraction, 58

F Fan chart, 241
242 FCCP framework. See Fuzzy credibility constrained programming framework (FCCP framework) Feasibility level (FL), 84
85 Fed-batch fermentation, 54 Federal Power Commission (FPC), 517 Feed-in tariffs (FITs), 555
556, 566 Feedstocks, 48
54 bioalcohols by fermentation from corn, sugar, and cellulose, 50f potential pathways from microalgae to fuels, 50f production, 387
389

614

Index

Feedstocks (Continued) world bioethanol production, 48t FEL. See Following electrical load (FEL) Fermentation bacteria, 47
48 fermenting acid-producing bacteria, 47
48 technologies, 43, 54
61 pervaporation membrane separation coupled with fermentation, 61f principle of pervaporation membrane separation, 60f FF. See Fill factor (FF) FFP. See Fuzzy flexible programming (FFP) Fick’s law, 10 Fill factor (FF), 9 Financial barriers, 547 Financial crisis, 588
591 Finned-tube heat exchanger, 68
69, 76 Fired units, 134 First-stage unit commitment model, 175
177 Fiscal incentives, 558 FITs. See Feed-in tariffs (FITs) Five generation of district heating system, 319f FL. See Feasibility level (FL) Flashing, 338 Flat plate solar air heater (FPSAH), 21 Flat plate solar water heater (FPSWH), 20
21 Fluorinated gases, 235 FMP. See Fuzzy mathematical programming (FMP) FO. See Forward osmosis (FO) Following electrical load (FEL), 133 Following thermal load (FTL), 133 Forward osmosis (FO), 339
340 Fossil energy, 311
312 Fossil energy crisis, 43
44 Fossil fuels, 44
45, 127, 203
204, 237
238 hybrid energy systems, 139 Fossil-firing technologies, 260
261 4-dimensional Geographic Information System (4d-GIS), 326
328 4th Generation District Heating (4GDH), 314, 318 FPC. See Federal Power Commission (FPC) FPP. See Fuzzy possibilistic programming (FPP) FPSAH. See Flat plate solar air heater (FPSAH) FPSWH. See Flat plate solar water heater (FPSWH) FRP. See Fuzzy robust programming (FRP)

FS. See Fuzzy set (FS) FTL. See Following thermal load (FTL) Fuel cell vehicles, 234 Fuel combustion process, 134 Fuzzification, 494
495 Fuzzy credibility constrained programming framework (FCCP framework), 219
220 Fuzzy flexible programming (FFP), 214
216 Fuzzy inference system, 493
494, 494f application, 496
503 benefit-type criteria inference system, 501f cost-type criteria inference system, 502f defuzzification, 495 fuzzification, 494
495 operation rules, 495 Fuzzy mathematical programming (FMP), 207
208, 214
218 FFP, 214
216 FPP, 216
217 FRP, 217
218 Fuzzy MCDM method, 492
493 Fuzzy possibilistic programming (FPP), 214, 216
217 Fuzzy robust programming (FRP), 214, 217
218 Fuzzy set (FS), 470, 472 divergence measures for, 475
479 divergence measures-based fuzzy TOPSIS method, 480
485 prerequisites, 472
474 theory, 494
495

G GA. See Genetic algorithm (GA) GaAs. See Gallium arsenide (GaAs) Gain output ratio (GOR), 350 Gallium (Ga), 5, 13 Gallium arsenide (GaAs), 9
10 Gas striping, 58
59, 59f Gasification GHG footprints of, 395
397 process, 391 Gasolin
ethanol mixtures, 43
44 GBEP. See Global Bioenergy Partnership (GBEP) 4GDH. See 4th Generation District Heating (4GDH) GDP. See Gross domestic product (GDP) Genetic algorithm (GA), 206 Geographic Information System (GIS), 322
323, 408

Index Geothermal energy, 303 geothermal energy-driven desalination, 354
356 technologies, 141
142 Geothermal field, 69
70 Geothermal heat pumps (GHP), 304 Geothermal heat source applications, 68 Geothermal plant, 68 Germany, RE subsidy policies in, 572
573 GHG. See Greenhouse gas (GHG) GHP. See Geothermal heat pumps (GHP) GIS. See Geographic Information System (GIS) Global Bioenergy Partnership (GBEP), 409
410 Global renewable power deployment, 238
239 electricity generated by renewable sources at worldwide level, 240f share of electricity generation from variable renewable energy, 239f Global warming, 44
45 Global warming potential (GWP), 69, 448 GOR. See Gain output ratio (GOR) Grain yield, influential factors of, 410 ‘Green Deal’, 541
542 Greenhouse gas (GHG), 138, 233, 267 emissions, 43, 289
290, 292, 376, 408
409, 563
564 footprints of H2 production, 395
400 trends, 234
238 Gross domestic product (GDP), 257
258 Ground source heat pump (GSHP), 25, 27
28, 28f GWP. See Global warming potential (GWP)

H Haber
Bosch ammonia synthesis, 440, 441f hydropower-based electrolysis and, 445 SMR and, 442 wind power-based electrolysis and, 442
444 Haber
Bosch process, 437
438 HAWT. See Horizontal axis wind turbine (HAWT) HCFCs. See Hydrochlorofluorocarbons (HCFCs) HCPV. See High concentrating photovoltaics (HCPV) Heat distribution, 321 Heat energy, cascade and upgrade use of, 316
317

615

Heat exchanger, 29, 134 model, 76 capital cost model, 79t heat-transfer coefficient correlations for heat exchangers, 78t parameters of heat exchangers, 77t Heat pumps, 135 Heat storage, 321 Heat transfer enhancement techniques, 138 Heat transfer fluid (HTF), 18
19, 136 Heat-driven cooling technologies, 135 Heat-recovery steam generators (HRSGs), 134 Heat-recovery units, 134
135 heat pumps, 135 types, 134 fired units, 134 unfired units, 134 Heat-transfer coefficient, 76 Heating and power operation strategies, 133 Hemicellulose, 49 HEV. See Hybrid electric vehicle (HEV) HFCs. See Hydrofluorocarbons (HFCs) High concentrating photovoltaics (HCPV), 6 High penetration of renewable sources in power sector, 239
247 energy storage, 244
247 2030 European energy transition strategies, 247
253 optimal development of nondispatchable resources, 241
243 seasonal hourly wind generation distribution, 241f surplus and backup powers, 243
244 High-renewable hybrids, 147
148 Hole transporting layer (HTL), 12 HOMER. See Hybrid Optimization Model for Multiple Energy Resources (HOMER) Horizontal axis wind turbine (HAWT), 300
301, 301f HPP. See Hydroelectric Power Plants (HPP) HRES. See Hybrid renewable energy system (HRES) HRSGs. See Heat-recovery steam generators (HRSGs) HT. See Human toxicity (HT) HTF. See Heat transfer fluid (HTF) HTL. See Hole transporting layer (HTL) Human toxicity (HT), 448 Humidification desalination, 34
35, 35f Hybrid electric vehicle (HEV), 234 Hybrid energy systems, 139 ZEBs, 139

616

Index

Hybrid inexact mathematical programming, 219
220 Hybrid Optimization Model for Multiple Energy Resources (HOMER), 207 Hybrid photovoltaic-thermal systems, 146 Hybrid renewable energy system (HRES), 204, 204f deterministic optimization techniques, 204
207 classical techniques, 205
206 commercial software, 207 metaheuristic algorithm, 206
207 inexact mathematical programming methods, 207
220 integrated inexact optimization framework, 220
222 system-based autonomous energy systems, 220
221 HYBRID2, 207 Hydrochlorofluorocarbons (HCFCs), 126 Hydroelectric Power Plants (HPP), 512, 514 Hydrofluorocarbons (HFCs), 126 Hydrogen (H2), 247, 321, 376, 436 applications, 377
383 inventory analysis of bio-oil reforming, 391
392 of SCWG, 392
394 of solar-based water electrolysis, 386 of thermal gasification of biomass, 386
391 of wind-based water electrolysis, 385
386 LCA, 384
385 production, 439
440 cost index, 449
450 effects of exergy efficiency, 456 effects of global warming, 456
462 effects of HT, 462
466 efficiency index, 448
449 environmental impact index, 450
452 sustainability index values of, 453 weighting scheme, 452 production pathways, 377
383, 378f biomass to, 380
383 GHG footprints, 395
400 water electrolysis, 379
380 sensitivity and uncertainty analyses, 394
395 thermophysical properties of, 437t Hydropower conversion technologies, 138 Hydropower-based electrolysis, 439 and Haber
Bosch ammonia synthesis, 445

I ICEs. See Internal Combustion Engines (ICEs) ICT. See Information and Communication Technologies (ICT) IEA. See International Energy Agency (IEA) IEEE 118-bus network results, 195
196 unit commitment results, 197t IEEE 33-bus network, 192
193 IEEE 33-bus radial DS, 190 IEEE six-bus system, 185, 185f IEMs. See Ion-exchange membranes (IEMs) IF-THEN rules, 495 IHX-ORC. See ORC with internal heat exchanger (IHX-ORC) IHX-ORC system, 118
119, 119f IMED/CGE model. See Integrated model of energy, environment and economy for sustainable development/ computable general equilibrium model (IMED/CGE model) IMP. See Interval mathematical programming (IMP) Implementation period (IP), 496
497 In situ separation technologies, 57
58 Independent System Operators (ISOs), 519 Indirect emissions, 82 Indirect type desalination, 34
37. See also Direct type desalination humidification and dehumidification desalination, 34
35 multistage flash desalination, 35 osmotic desalination driven by solar energy, 36
37 vapour compression desalination, 36 Indium (In), 5, 13 InDS. See Integrated distribution systems (InDS) Industrial output, 274
277, 275f Industrial symbiosis, 324
326, 325f Industrial waste, 49
51 Inexact mathematical programming methods, 207
220 FMP, 214
218 HRES, 219
220 IMP, 218
219 RO, 212
214 SMP, 208
212 Information and Communication Technologies (ICT), 323 Information barriers, 547 Infrared rays (IR rays), 3
4

Index Infrastructure barriers. See Technical barriers Injection opening pressure (IOP), 149
151 Inlet temperature of geothermal source, 95
99 Input-output table (IOT), 260 Insoluble organic matters, 47
48 Institutional and administrative barriers, 548 Insulator, 7 InTDS. See Integrated transmission and distribution system (InTDS) Integrated distribution systems (InDS), 174
175, 194
195, 196t InD-1, 193 InD-3, 193 Integrated inexact optimization framework, 220
222, 222f Integrated model of energy, environment and economy for sustainable development/ computable general equilibrium model (IMED/CGE model), 259
260 Integrated power TS and DS, 169
171 conventional transmission and distribution systems, 170f mathematical model, 174
184 modern transmission and distribution systems, 171f numerical results, 184
196 Integrated solar combined cycles (ISCCs), 148 Integrated transmission and distribution system (InTDS), 169
170, 192
195 characteristics of three integrated distribution systems, 193t IEEE 118-bus network results, 195
196 integrated mode vs. isolated mode, 192t, 194t sensitivity analyses, 194
195 Integrated urban planning modelling strategic urban renewal for promoting district heating, 326
328 for renewable-energy-based district heating, 324
328 urban and industrial symbiosis, 324
326 Intergovernmental Panel on Climate Change (IPCC), 267 Intermittent renewable resources (IRRs), 511, 516
533 complexity, 517
521 economic effects, 526
529 externalities and MOE, 529
533 operation problem, 521
526 social acceptance of, 533
534 Internal Combustion Engines (ICEs), 128

617

International Atomic Energy Agency, 350
351 International Energy Agency (IEA), 126, 158, 311
312, 540 International interconnections, 251
253 Internet of Things (IoT), 323 Interval mathematical programming (IMP), 207
208, 218
219 Interval-valued random variable, 219
220 Inventory analysis of bio-oil reforming, 391
392 of SCWG, 392
394 of solar-based water electrolysis, 386 of thermal gasification of biomass, 386
391 of wind-based water electrolysis, 385
386 Investment, 273
274 in nonfossil power generation, 274f Investment ration (IR), 496
497 Ion-exchange membranes (IEMs), 339 IOP. See Injection opening pressure (IOP) IOT. See Input-output table (IOT) IoT. See Internet of Things (IoT) IP. See Implementation period (IP) IPCC. See Intergovernmental Panel on Climate Change (IPCC) IR. See Investment ration (IR) IR rays. See Infrared rays (IR rays) Iron (Fe), 3
4 Irresistible expansion, 515
516 IRRs. See Intermittent renewable resources (IRRs) ISCCs. See Integrated solar combined cycles (ISCCs) IsDS. See Isolated distribution systems (IsDS) Isolated distributed energy resource management problem, 190
192 Isolated distribution systems (IsDS), 174
175, 184
185 Isolated unit commitment problem, 185
189 net loads including solar generations, 189f including wind generations, 188f transmission line data, 186t unit commitment decisions, 187t units data, 186t ISOs. See Independent System Operators (ISOs) Isothermal process, 137

K Kirchhoff current law, 181

618

Index

L Land use change, 323
324 Landfill gas (LFG), 142 Large-scale integration of variable renewable resources climate change and greenhouse gas emissions trends, 234
238 global renewable power deployment, 238
239 high penetration of renewable sources in power sector, 239
247 Latent heat storage, 137 LCAs. See Life-cycle assessments (LCAs) LCB. See Low-carbon building (LCB) LCOE. See Levelized cost of energy (LCOE) LCPV. See Low concentrating photovoltaics (LCPV) LCZ. See Lower convective zone (LCZ) Lead (Pb), 13 Levelized cost of electricity. See Levelized cost of energy (LCOE) Levelized cost of energy (LCOE), 249, 539
540 evolution of LCOE on renewable sources, 250
251 LFG. See Landfill gas (LFG) LFR. See Linear Fresnel Reflector (LFR) Life-cycle and carbon footprint analysis of ORC analysis of emission reductions, 105
106 environmental evaluation of components, 103
104 of life cycle, 99
103 of working fluids, 104
105 sensitivity analysis, 106
108 Life-cycle assessments (LCAs), 292, 376
377, 384
385, 409 Life-cycle boundary, 81
82 Life-cycle environmental analysis, 81
84 carbon footprint analysis, 82
83 data sources, 83
84, 84t life-cycle boundary, 81
82 Lignocellulosic biomass, 49 Linear Fresnel collector, 18 Linear Fresnel Reflector (LFR), 22 Linear programming (LP), 205 Liquid desiccant cooling system, 30
31, 31f Liquid desiccation, 30
31 Liquified natural gas (LNG), 316 Lithium (Li), 245
247 LNG. See Liquified natural gas (LNG) Low concentrating photovoltaics (LCPV), 6

Low renewable hybrids, 148 Low-carbon building (LCB), 290
291 advancing, 305
307 renewable energy technologies building construction, 295 building material extraction and transportation, 292
294 building operation, 296
304 Low-carbon sustainable energy system, 311
312 Lower convective zone (LCZ), 32 LP. See Linear programming (LP)

M Macroeconomic trends towards (2050), 265
267, 266t Magnesium (Mg), 3
4 MALHs. See Methylammonium lead halides (MALHs) Mathematical model, 174
184 distributed energy resource management problem, 178
183 first-stage unit commitment model, 175
177 numerical results, 184
196 power system with DS1 and DS2, 175f second-stage economic dispatch model, 177
178 tighter formulations, 183
184 MCDM. See Multicriteria decision making (MCDM) MCDMA. See Multicriteria decision making analysis (MCDMA) MD. See Membrane distillation (MD) Mechanical vapour compression (MVC), 338 MED. See Multieffect distillation (MED) Medium-renewable hybrids, 148 Megawatts (MW), 386 Membrane desalination techniques, 339
340 Membrane distillation (MD), 57
59 Merit order effect (MOE), 529
533 Metaheuristic algorithm, 206
207 Metal chalcogenides, 13
14 Methane (CH4), 44
45, 235 producing bacteria, 47
48 Methanogens, 47
48 Methyl nutrition bacteria, 47
48 Methylammonium lead halides (MALHs), 12
13 Metric tons of coal equivalent (MTCE), 298
299 Microalgae, 49

Index MicroCCHP, 130 Microgeneration, 130 Microgrids, 132 Microorganisms, 45
48 metabolite pathway of C. acetobutylicum, 47f Microturbines (MTs), 128, 178
179 Mixed-integer linear programming (MILP), 174, 205 Modular advanced reactor (SMART), 352 MOE. See Merit order effect (MOE) Molten salts, 138 Monocrystalline silicon (Mono-Si), 297 Monosaccharides, 49 Monte Carlo simulation, 394
395 MOP. See Multiobjective programming (MOP) MSF. See Multistage flash (MSF) MSP. See Multistage stochastic programming (MSP) MTCE. See Metric tons of coal equivalent (MTCE) MTs. See Microturbines (MTs) Mufflers, 134 Multicriteria decision making (MCDM), 471, 492 Multicriteria decision making analysis (MCDMA), 409 Multicriteria integrated assessment, 84
85 hierarchy network structure of multifactor evaluation, 85f Multieffect distillation (MED), 337
338 Multifactor evaluation results of ORC, 112
117 exergy cost rate balances and auxiliary equations, 121t grading values of indicators for ORC, 116t indicators grading standards, 115t models of energy balance and exergy, 120t normalization of energetic, exergetic, economic and environmental performances, 114t results of energetic, exergetic, economic and environmental performances, 113t weights allocation of indicators, 117t Multiobjective optimization model, 206 Multiobjective programming (MOP), 205 Multistage flash (MSF), 337 desalination, 35, 35f, 338 techniques, 34 Multistage stochastic programming (MSP), 210

619

Municipal waste, 49
51 MVC. See Mechanical vapour compression (MVC) MW. See Megawatts (MW)

N Nano-filtration (NF), 339 National Renewable Energy Laboratory (NREL), 290
291 Nationally Determined Contributions (NDCs), 203
204, 555 Natural polymer, 56
57 NCZ. See Nonconvective zone (NCZ) NDCs. See Nationally Determined Contributions (NDCs) Needy InDS, 194f, 195 Negative-ideal solution (NIS), 471
472 Neon (Ne), 3
4 Net energy balance, 414 Net metering/net billing, 557
558 Net-zero energy building (NZEB), 290
291, 291f, 292f NF. See Nano-filtration (NF) n
i
p structure, 12 NIS. See Negative-ideal solution (NIS) Nitrogen, 58
59 Nitrous oxide (N2O), 235 NLP. See Nonlinear programming (NLP) Nonconcentrating solar collectors, 146 Nonconvective zone (NCZ), 32 Nondispatchable resources, optimal development of, 241
243 Nonfossil power generation, investment in, 261
264, 263t Nonlinear programming (NLP), 205 Nonrenewable energy sources, 204 NPPA. See Nuclear Power Plants Authority (NPPA) NREL. See National Renewable Energy Laboratory (NREL) Nuclear energy-driven desalination, 350
353 Nuclear Power Plants Authority (NPPA), 352 NZEB. See Net-zero energy building (NZEB)

O O&MC. See Operating and maintenance costs (O&MC) Ocean energy sources, 149 Ocean/wave energy-driven desalination, 357 Operating and maintenance costs (O&MC), 496
497 Operating hours (OH), 496
497

620

Index

OPF. See Optimal power flow (OPF) Optimal power flow (OPF), 173
174 ORC. See Organic Rankine cycle (ORC) ORC with internal heat exchanger (IHXORC), 69 Organic fluids, 68 Organic Rankine cycle (ORC), 68, 128 layouts of ORC systems, 108
112 optimum value for basic ORC with R1234yf, 109t optimum value for dual-pressure ORC with R1234yf, 111t optimum value for IHX-ORC with R1234yf, 110t life-cycle and carbon footprint analysis of ORC, 99
108 methods and models, 73
85 economic and exergoeconomic analysis, 76
81 heat exchanger model, 76 life-cycle environmental analysis, 81
84 multicriteria integrated assessment and decision-making, 84
85 thermodynamic and technical analysis, 73
76 results of multifactor evaluation, 112
117 system description and working fluid selection, 70
72, 71f conditions and input parameters, 72t properties of working fluids, 73t thermodynamic and economic results, 85
99 Osmotic desalination driven by solar energy, 36
37, 37f Ozone-depleting CFCs, 135

P P-GA-PSO. See Parallel hybrid genetic algorithm-particle swarm optimization algorithm (P-GA-PSO) PA. See Polyamide (PA) PACF. See Partial autocorrelation function (PACF) Packed bed systems, 137
138 Parabola-shaped reflector, 21 Parabolic trough collector (PTC), 152
153 Parallel hybrid genetic algorithm-particle swarm optimization algorithm (P-GAPSO), 206
207 Partial autocorrelation function (PACF), 411 Particle swarm optimization (PSO), 206 Passive system, 136

PBRs. See Photobioreactors (PBRs) PCA. See Principal component analysis (PCA) PCC. See Point of common coupling (PCC) PCMs. See Phase change materials (PCMs) PDF. See Probability distribution function (PDF) PDMS. See Polydimethylsiloxane (PDMS) PEC. See Photoelectrochemical cell (PEC) PEDOT:PSS. See Poly(3,4-ethylene dioxythiophene):polystyrene sulfonic acid (PEDOT:PSS) Peltier effect, 26
27 PEM. See Polymer electrolyte membrane electrolysis (PEM) Performance ratio (PR), 350 Perovskite solar cells (PSCs), 11
13 Pervaporation, 60 Petroleum-based fuels, 43 Phase change materials (PCMs), 136, 321 PHES. See Pumped-hydroelectric energy storage (PHES) Phosphorus (P), 8 Photobioreactors (PBRs), 392
393 Photoelectrochemical cell (PEC), 440 Photoelectrochemical water splitting, 440, 445
448 Photon, 6
7 Photovoltaic(s) (PV), 4, 260, 294, 336, 539
540 advancements in SPV technologies, 11
18 applications of solar cells, 18 generations, 173
174 modules, 445 production, 386 technologies, 5
18, 6f conversion efficiency, 9
11 FF, 9 power of solar cell, 9 SPV principles, 6
11 Photovoltaic
thermal technology (PV-T technology), 27, 296
297, 299
304 household biogas system, 303f PHWR. See Pressurized heavy water reactor (PHWR) p
i
n structure, 12 PIS. See Positive-ideal solution (PIS) Plug flow reactor, 52
54 Point of common coupling (PCC), 170
171 Policy implications, 277
278 Poly-Si. See Polycrystalline silicon (Poly-Si) Poly(1-trimethylsilyl-1-propyne) (PTMSP), 60
61

Index Poly(3,4-ethylene dioxythiophene):polystyrene sulfonic acid (PEDOT:PSS), 12 Polyamide (PA), 60
61 Polycrystalline silicon (Poly-Si), 297 Polydimethylsiloxane (PDMS), 60
61 Polyethylene, 59 Polygeneration, 130, 131f Polygeneration microgrids, 132 Polymer electrolyte membrane electrolysis (PEM), 379
380 Polypropylene, 59 Polysaccharides, 56
57 Polytetrafluoroethylene (PTFE), 59 Polyvinylidene fluoride (PVDF), 59 Portugal, energy-based power generation in, 588 dataset, 600t econometric model and data, 589
591 empirical results, 592
598 testing for cointegration and estimating parameters, 591
592 testing for unit roots and detecting outliers, 591 Positive-ideal solution (PIS), 471
472 Power (P), 496
497 operation strategies, 133 power-generation technology, 261 of solar cell, 9 structure, 270
271 Power Purchase Agreement (PPA), 251 PR. See Performance ratio (PR) Pressurized heavy water reactor (PHWR), 352 Pressurized water reactor (PWR), 352 Primal and dual decomposition, 210
211 Primary energy, 269
270 Prime-mover system classification by, 128
129 technology, 132 Principal component analysis (PCA), 409
410 Probability distribution function (PDF), 209
210 Proteins, 56
57 PSCs. See Perovskite solar cells (PSCs) PSO. See Particle swarm optimization (PSO) PTC. See Parabolic trough collector (PTC) PTFE. See Polytetrafluoroethylene (PTFE) PTMSP. See Poly(1-trimethylsilyl-1-propyne) (PTMSP) Public financing of renewable energy, 558 Pumped-hydroelectric energy storage (PHES), 245

621

Pumped-storage hydropower, 245 PV. See Photovoltaic(s) (PV) PV-T technology. See Photovoltaic
thermal technology (PV-T technology) PVDF. See Polyvinylidene fluoride (PVDF) PWR. See Pressurized water reactor (PWR)

Q Quantum dot sensitized solar cells (QDSCs), 16
17, 17f Quantum dots (QDs), 16 solar cells, 16
18

R R&D. See Research and development (R&D) Radial network, 178
179, 178f Random fuzzy variable, 219
220 Random variables (RVs), 175 RDF technology. See Refuse-derived fuel technology (RDF technology) RDGs. See Renewable distributed generations (RDGs) RE. See Renewable energy (RE) REC mechanism. See Renewable energy certificates mechanism (REC mechanism) Recuperative ORC, 68 Recuperators, 134 Reference scenario, 264
265, 268t Refuse-derived fuel technology (RDF technology), 142 Regenerative ORC, 68 Regenerators, 134, 136 Regional grain yield, 416 Regional Transmission Operators (RTOs), 519 Regression, Uncertainty and Sensitivity Tool (RUST), 394
395 Regulatory and policy barriers, 548 REmax scenario, 265 Renewable ammonia, 436
438 Renewable distributed generations (RDGs), 173 Renewable energy (RE), 127, 138
149, 203
204, 257
258, 335
336, 469
470, 539
540, 563
564. See also Solar energy barriers, 541
544 identification, 544
554 issues and opportunities in desalination, 357
360 biomass energy, 142
144

622

Index

Renewable energy (RE) (Continued) in carbon and air pollutant emissions reduction, 271
273 comparison with studies, 278
279, 278t consumption, 564f costs, decrease in, 248
251 evolution of LCOE on renewable sources, 250
251 desalination technology, 336
345, 358t development, 273
277, 567
570 economic impacts of renewable energy development, 273
277 in electricity production, 238t, 239 energy and desalination, 345
346 geothermal energy technologies, 141
142 hybrid energy systems, 139 impacts on energy system, 267
271 indicators, 565t integrated desalination, 347
357 geothermal energy-driven desalination, 354
356 nuclear energy-driven desalination, 350
353 ocean/wave energy-driven desalination, 357 solar desalination, 347
350 wind energy-driven desalination, 353
354 integrated trigeneration technologies, 149
154 investment, 481
485 for LCBs advancing LCB, 305
307 building and energy and environmental challenges, 289
290 building life-cycle systems and greenhouse gas emissions, 292 NZEB and LCB, 290
291 technologies for LCB, 292
304 limitations, 279
282 macroeconomic trends towards (2050), 265
267 measures to overcome barriers, 554
555 methods and scenarios, 259
265 data sources, 264 economic assessment of renewable energy, 260
261 IMED/CGE model, 259
260 investment in nonfossil power generation, 261
264 scenarios, 264
265 policy frameworks, 544
550

policy implications, 277
278 promotion measures, 555
557 renewable sources, 148
149, 150f resources, 171
172, 290, 435
436 sensitivity analysis, 279 solar energy, 144
148 subsidy in China, 575
577 mechanism of, 577
581 policies in European Union, 571
575 policies in United States, 570
571 policy implications, 580
581 systems, 132 targets, 554
555 wind energy, 139
141 Renewable Energy Act, 573 Renewable energy certificates mechanism (REC mechanism), 556
557 Renewable Energy Development Fund, 575
576 Renewable energy sources (RESs), 132, 469
470, 491
492. See also Intermittent renewable resources (IRRs) Renewable energy targets (RETs), 267 Renewable energy technologies (RETs), 545, 547 Renewable energy-based trigeneration systems, 154
158 challenges and barriers, 154
156 cogeneration and trigeneration, 127
134 cooling technologies, 135
136 heat-recovery units, 134
135 opportunities and prospects, 156
158 renewable energy, 138
149 renewable energy integrated trigeneration technologies, 149
154 TES, 136
138 Renewable heat sources, 321 Renewable hydrogen, 436 Renewable portfolio standard (RPS), 557, 566, 571 Renewable resource, 43 Renewable sources high penetration in power sector, 239
247 LCOE on, 250
251 Renewable-energy-driven district heating system district heating and cooling system, 313f integrated urban planning for, 324
328 technologies and system design for, 314
324

Index indicators and design principle for enhancement, 314
317 optimization for, 321
324 system design and key technologies of, 318
321 Research and development (R&D), 13, 565 Reserves, 522 RESs. See Renewable energy sources (RESs) RETs. See Renewable energy targets (RETs); Renewable energy technologies (RETs) Reverse auction, 556 Reverse osmosis (RO), 339 Rich InDS, 194f, 195 RO. See Reverse osmosis (RO); Robust optimization (RO) Robust optimization (RO), 204, 207
208, 212
214 RPS. See Renewable portfolio standard (RPS) RTOs. See Regional Transmission Operators (RTOs) RUST. See Regression, Uncertainty and Sensitivity Tool (RUST) RVs. See Random variables (RVs)

S Saccharomyces cerevisiae, 45
46 Saline water, 335 SCCP. See Solar combined cooling and power (SCCP) SCWG. See Supercritical water gasification (SCWG) SDG. See Small Distributed Generation (SDG) SDGs. See Sustainable development goals (SDGs) Seawater desalination, 140 Second-order cone programming (SOCP), 179 Second-stage economic dispatch model, 177
178 SEG. See Stromeinspeisungsgesetz (SEG) SEIs. See Straw-energy industries (SEIs) Semiconductor, 7 Sensible heat storage, 136
137 Sensitivity analysis, 106
108, 107f, 108f, 194
195, 279, 280t, 281t on economic performance, 95
99, 99f Separate production (SP), 127, 129f Shannon’s entropy, 472 Silica gel, 30 Silicon (Si), 3
4, 8 Si-based semiconductor materials, 5
6

623

Silver (Ag), 17
18 Simulation-based mixed-integer optimization model, 205 Single flash steam power plant, 141 Single-stage continuous fermentation, 55
56 Small Distributed Generation (SDG), 520 SMART. See Modular advanced reactor (SMART) SMP. See Stochastic mathematical programming (SMP) SMR. See Steam methane reforming (SMR) Social and environmental barriers, 548 Socioeconomic benefits of straw, 421
424 SOCP. See Second-order cone programming (SOCP) Sodium chloride (NaC), 335 SOEC. See Solid oxide electrolysis (SOEC) Solar, 241
243, 516 cooking, 33, 34f desiccant cooling system, 30
31, 30f ejector cooling system, 31
32, 31f ground source heat pump system, 27
28 organic Rankine cycle, 36 photovoltaic desalination, 347
349 photovoltaic systems, 146 pond, 32
33, 32f, 136 solar-aided coal power plants, 148 sorption systems, 24
25 thermoelectric cooling system, 26
27, 27f Solar cells, 18 equivalent circuit of, 10f power of, 9 Solar collectors, 144
146 concentrating solar collectors, 146 nonconcentrating solar collectors, 146 Solar combined cooling and power (SCCP), 152
153 Solar cooling technologies, 23
32, 24f, 24t solar PV powered cooling system, 25
28 solar thermal powered cooling system, 28
32 Solar desalination, 34
38, 347
350 direct type desalination, 37
38 indirect type desalination, 34
37 solar photovoltaic desalination, 347
349 STD, 349
350 Solar energy, 144
148, 296. See also Renewable energy (RE) conversion technologies, 23 hybrid photovoltaic-thermal systems, 146 solar collectors, 144
146 solar photovoltaic systems, 146

624

Index

Solar energy (Continued) solar thermal applications, 146
147 solar-renewable hybrids, 147
148 system, 153
154 technologies PV technologies, 5
18 solar cooking, 33 solar cooling technologies, 23
32 solar desalination, 34
38 solar pond, 32
33 spectral properties of solar radiation outside earth’s atmosphere, 4f STC, 18
23 Solar PV (SPV), 5, 297
298, 539
540 advancements in SPV technologies CdTe, 13
14 copper indium gallium selenide, 14
15 DSSCs, 15
16 emerging PV technologies, 13 PSCs, 11
13 QD solar cells, 16
18 powered cooling system solar GSHP system, 27
28 solar thermoelectric cooling system, 26
27 SVCC system, 25
26 principles, 6
11 Solar thermal, 298
299, 299f applications, 146
147 energy, 31
32 storage, 138 powered cooling system, 28
32 solar desiccant cooling system, 30
31 solar ejector cooling system, 31
32 solar sorption cooling system, 29
30 Solar thermal collectors (STC), 18
23, 19t stationary collectors, 20
21 tracking concentrating collectors, 21
23 Solar thermal desalination (STD), 349
350 Solar vapour compression cooling system (SVCC system), 25
26, 26f Solar-based water electrolysis, inventory analysis of, 386 Solar-Brayton cycles, 148 Solar-renewable hybrids, 147
148 high-renewable hybrids, 147
148 low renewable hybrids, 148 medium-renewable hybrids, 148 Solid desiccant, 25 Solid materials, 136
137 Solid oxide electrolysis (SOEC), 379
380 Soluble organic matters, 47
48

Solventogenesis, 46, 55
56 Sorption technology, 135 SP. See Separate production (SP) Spain, RE subsidy policies in, 573
574 SPR. See Stochastic programming with recourse (SPR) SPV. See Solar PV (SPV) Starch, 49 wastewater, 51
52 Stationary collectors, 20
21, 20f STC. See Solar thermal collectors (STC) STD. See Solar thermal desalination (STD) Steam methane reforming (SMR), 439 Stochastic mathematical programming (SMP), 207
212 CCP, 208
210 SPR, 210
212 Stochastic programming formulation, 174 Stochastic programming with recourse (SPR), 208, 210
212 Storage concept in TES active system, 136 passive system, 136 mechanisms/types of TES, 136
137 chemical storage, 137 latent heat storage, 137 sensible heat storage, 136
137 Straw, 407
408 economic benefits, 429 energy, environmental and socioeconomic benefits of, 421
424 energy potential, 421 for energy production, 410
411 cost and profit, 411
413 environmental impacts, 413
414 factors affecting results, 425
428 parameters of energy conversion technologies, 419
421 quantity, 428 estimation, 416
419 selection of evaluation indicators, 414 Straw-energy industries (SEIs), 409 Straw
grain ratio, 408 Stromeinspeisungsgesetz (SEG), 573 Structural breaks, 588
589, 591
595 Subcritical ORC, 68, 71f Sulphur (S), 3
4 Sun, 3
4 Supercritical water gasification (SCWG), 376
377, 382
383 GHG footprints of, 398
399

Index inventory analysis of algae, 392
394 Supercritical/transcritical ORC, 68 Supply side optimization, 322
323 Support mechanisms, 555
557 Surface attachment, 56
57 Surplus and backup powers, 243
244 bivariate kernel density plot of stored energy, 245f Surplus electricity, 173
174 Sustainable development, 289 Sustainable development goals (SDGs), 333
334 SVCC system. See Solar vapour compression cooling system (SVCC system) Synthetic polymers, 56
57 System composition of renewable-energydriven district heating system, 318
319

T TCO. See Transparent conductive oxide (TCO) tCO2A. See Tons of CO2 avoided (tCO2A) TDM. See Thermal demand management (TDM) Technical barriers, 546
547 Technique for order preference by similarity to ideal solution method (TOPSIS method), 471
472 divergence measures-based fuzzy, 480
485 Telluride (Te), 13 Tellurium (Te), 5 Tendering schemes, 556 Terawatt power (TW power), 11
12 TES. See Thermal energy storage (TES) TGCs. See Tradable green certificates (TGCs) Thermal conversion processes, 142 Thermal demand management (TDM), 133 Thermal desalination techniques, 337
339 Thermal energy storage (TES), 126, 136
138 combined heat storage, 137 packed bed systems, 137
138 solar thermal energy storage, 138 storage concept, 136 storage mechanisms/types of TES, 136
137 Thermal gasification, 381
382 of biomass, 386
391 Thermal vapour compression (TVC), 338 Thermally activated cooling technologies, 127 Thermo-economic assessment, 151 Thermodynamic and economic results

625

effects of design parameters on economic performance, 89
92 on exergoeconomic performance, 92
95 on thermodynamic performance, 85
89, 86t sensitivity analysis on economic performance, 95
99 Thermodynamic and technical analysis, 73
76 flowchart of ORC system evaluation framework, 74f models of energy balance and exergy, 75t Thermoelectric effect, 27 Thin-film technology, 5 Tighter formulations, 183
184 Titanium dioxide (TiO2), 15
16 Tons of CO2 avoided (tCO2A), 496
497 TOPSIS method. See Technique for order preference by similarity to ideal solution method (TOPSIS method) Tracking concentrating collectors, 21
23, 22f Tradable green certificates (TGCs), 555
556 Transmission system (TS), 169
170 Transmission system operator (TSO), 172 Transparent conductive oxide (TCO), 13
14 Transport sector, 234 Transportation cost, 419
420 Trigeneration, 127
134 cooling applications in, 135
136 systems classification by applications, 128 by sequence of energy, 130 by size, 128 by type of prime-mover, 128
129 TS. See Transmission system (TS) TSO. See Transmission system operator (TSO) TSP. See Two-stage stochastic programming (TSP) Turbine technology, 140 TVC. See Thermal vapour compression (TVC) TW power. See Terawatt power (TW power) Two-stage continuous fermentation, 55
56 Two-stage stochastic programming (TSP), 210

U UAE. See United Arab Emirates (UAE) UASB. See Upflow anaerobic sludge blanket (UASB) UBF. See Upflow blanket filter (UBF) UC. See Unit commitment (UC)

626

Index

UCZ. See Upper convective zone (UCZ) UL. See Useful life (UL) Ultraviolet rays (UV rays), 3
4 Underground hydraulic digester, 52
54 UNFCCC. See United Nations Framework Convention on Climate Change (UNFCCC) Unfired units, 134 Unit commitment (UC), 172 United Arab Emirates (UAE), 353 United Nations Framework Convention on Climate Change (UNFCCC), 291 Upflow anaerobic sludge blanket (UASB), 51
52 Upflow blanket filter (UBF), 52
54 Upper convective zone (UCZ), 32 Urban renewal for promoting district heating, 326
328 Urban symbiosis, 324
326, 325f Useful life (UL), 496
497 UV rays. See Ultraviolet rays (UV rays)

V Value-added and employment, 274
277 Vapour compression (VC), 337 desalination, 36, 36f, 338 Vapour compression refrigeration (VCR), 25
26 Variable renewable energy (VRE), 282, 546
547 Variable renewable energy sources (VRES), 234, 243, 243t, 244f VAWT. See Vertical axis wind turbine (VAWT) VC. See Vapour compression (VC) VCR. See Vapour compression refrigeration (VCR) Vegetable oils, straight, 143 Vertical axis wind turbine (VAWT), 300
301, 301f Volatile organic compounds (VOCs), 235 VRE. See Variable renewable energy (VRE) VRES. See Variable renewable energy sources (VRES)

W WASP. See Wind Atlas Analysis and Application Program (WASP) Waste-to-energy technologies, 325
326

Water, 137, 238
239, 334 Water electrolysis, GHG footprints of, 395 Water
gas shift reaction, 382
383 Weighting scheme, 452 Wheat straw, 400 Whilst lithium bromide
water absorption systems, 136 Wind, 241
243, 516 Wind Atlas Analysis and Application Program (WASP), 139 Wind energy, 139
141 turbine technology, 140 wind energy-driven desalination, 353
354 Wind Gen, 185
186 Wind hybrid systems and applications seawater desalination, 140 wind
diesel system, 140 wind
photovoltaic-hydrogen system, 140 Wind power desalination, 140 development, 140
141 meteorology and wind modelling, 139
140 Wind power-based electrolysis, 439 and Haber
Bosch ammonia synthesis, 442
444 Wind-based water electrolysis inventory analysis of, 385
386 Wind-based water electrolysis, 399
400 Wind-solar-compressed air energy storage (WS-CAES), 154 Wind-thermal-hydropower-pumped storage system, 212 Wind
diesel system, 140 Wind
photovoltaic-hydrogen system, 140 Winery waste water, 51
52 Working fluid(s), 72, 73t CO2,eq emissions of working fluids, 83t costs per exergy unit for basic ORC with, 98f environmental evaluation of, 104
105 WS-CAES. See Wind-solar-compressed air energy storage (WS-CAES)

Z Zeolite, 30 Zero energy buildings (ZEBs), 139, 323
324 Zinc oxide (ZnO), 14
16 Zirconium dioxide (ZrO2), 379
380 Zymomomas mobilis, 45
46