Energy Transition 9781119629450, 1119629454, 9781786302113

Citation preview

Energy Transition

Series Editor Alain Dollet

Energy Transition

Bernard Lachal

First published 2019 in Great Britain and the United States by ISTE Ltd and John Wiley & Sons, Inc.

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms and licenses issued by the CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the undermentioned address: ISTE Ltd 27-37 St George’s Road London SW19 4EU UK

John Wiley & Sons, Inc. 111 River Street Hoboken, NJ 07030 USA

www.iste.co.uk

www.wiley.com

© ISTE Ltd 2019 The rights of Bernard Lachal to be identified as the author of this work have been asserted by him in accordance with the Copyright, Designs and Patents Act 1988. Library of Congress Control Number: 2019937032 British Library Cataloguing-in-Publication Data A CIP record for this book is available from the British Library ISBN 978-1-78630-211-3

Contents

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xi

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xv

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xvii

Part 1. The Context of Case Study Feedback (CSF)

. . . . . . . . . . . . . .

1

Chapter 1. Energy Transition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3

1.1. The global energy system and its evolution . . . . . . . . . . 1.2. The necessary transformation of the global energy system . 1.2.1. Fossil fuels: planned scarcity upstream and environmental problem downstream . . . . . . . . . . . . . . . 1.2.2. Nuclear energy: environmental and accessibility issues . 1.2.3. An overall inefficient system . . . . . . . . . . . . . . . 1.2.4. A productive and simple-energy vision . . . . . . . . . . 1.2.5. Energy transition . . . . . . . . . . . . . . . . . . . . . . 1.3. The three concordances . . . . . . . . . . . . . . . . . . . . . 1.3.1. Form concordance . . . . . . . . . . . . . . . . . . . . . 1.3.2. Place concordance . . . . . . . . . . . . . . . . . . . . . 1.3.3. Time concordance . . . . . . . . . . . . . . . . . . . . . 1.3.4. Economic, social and environmental constraints . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

3 5

. . . . . . . . . .

6 6 7 8 9 10 11 12 12 12

Chapter 2. Energy Systems and Technological Systems. . . . . . . . . . . .

15

2.1. Transformers and concordances . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1. Form converters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2. Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

16 17 17

vi

Energy Transition

2.1.3. Transport. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. From the transformer to the energy system . . . . . . . . . . . . . . . . . . . . 2.3. Effectiveness of resources and effectiveness of results . . . . . . . . . . . . . .

18 18 22

Chapter 3. The Innovation Process . . . . . . . . . . . . . . . . . . . . . . . . . .

27

3.1. A well-defined process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Limit of these curves in the context of energy systems . . . . . . . . . . . . . . 3.3. Operation and use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

27 33 36

Chapter 4. Case Study Feedback, the Basis of Learning by Using . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39

4.1. Innovation in energy systems . 4.2. Case study feedback . . . . . . 4.2.1. CSF classification test . . 4.2.2. CSF content . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

39 42 43 45

Part 2. CSF Tools: Operation and Envisaged Uses. . . . . . . . . . . . . . . .

47

Chapter 5. The Human Context . . . . . . . . . . . . . . . . . . . . . . . . . . . .

49

5.1. Why the human aspects? . . . . . . . . . . . . . . . . . 5.1.1. In vivo rather than in vitro . . . . . . . . . . . . . . . 5.1.2. The importance of objective information in the field of innovative energy systems . . . . . . . . . . . . . . 5.2. Who are the actors involved and how are they involved? 5.2.1. Actors involved in the innovation process . . . . . . 5.2.2. Actors related to the particular energy system . . . . 5.2.3. Actors involved in the implementation of CSF . . . . 5.3. How to take into account human aspects in CSF . . . . . 5.3.1. The perimeter . . . . . . . . . . . . . . . . . . . . . . 5.3.2. The objectives of the CSF . . . . . . . . . . . . . . . 5.3.3. The resources . . . . . . . . . . . . . . . . . . . . . . 5.3.4. The team’s experience . . . . . . . . . . . . . . . . . 5.3.5. The follow-up group . . . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . . . . . . . . . . . . . . . . . . . . . .

49 49

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . .

50 51 51 51 54 55 55 56 57 57 58

Chapter 6. The Energy Context and the Sankey Diagram . . . . . . . . . . .

59

6.1. A drawing is better than a long speech. . . . 6.2. Design, development and operation . . . . . 6.2.1. The importance of precise terminology . 6.2.2. Balance failure . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . . . . . . . . .

. . . .

. . . . . . . . . . .

. . . .

. . . . . . . . . . .

. . . .

. . . . . . . . . . .

. . . .

. . . . . . . . . . .

. . . .

. . . . . . . . . . .

. . . .

. . . . . . . . . . .

. . . .

. . . . . . . . . . .

. . . .

. . . . . . . . . . .

. . . .

. . . .

59 63 63 66

Contents

vii

6.2.3. To avoid having a chilling effect . . . . . . . . . . . . . . . . . . . . . . . 6.2.4. Shape: graphic rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3. Uses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

67 69 72

Chapter 7. From System to Experimental Concept . . . . . . . . . . . . . . . .

77

7.1. The importance and difficulties of a quantitative quality assessment . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2. From the energy system to be evaluated to the measurement concept . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.1. From objectives to a breakdown into subsystems and components . . . . . . . . . . . . . . . . . . . 7.2.2. Developing the measurement system . . . . . . . . . . . 7.2.3. Some properties of the sensors and their use . . . . . . . 7.2.4. Some remarks on the measurement of primary energies 7.3. Link to other phases of the evaluation . . . . . . . . . . . . .

. . . . . . . . . .

77

. . . . . . . . . .

78

. . . . .

. . . . .

80 84 91 93 96

Chapter 8. Data Observation and Global Indicators . . . . . . . . . . . . . . .

99

8.1. Observing and feeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2. Energy indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

99 101

Chapter 9. Input/Output and Signature Relationships: the Operation in Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

107

9.1. Convenient visualization of an expected relationship 9.2. Search for a global relationship . . . . . . . . . . . . . 9.3. Signatures as simple management tools . . . . . . . . 9.4. The signature as the basis for adjustment . . . . . . . 9.5. The signature as the basis for a standard . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . . .

. . . . .

. . . . .

. . . . . .

. . . . .

. . . . .

. . . . . .

. . . . .

. . . . .

. . . . . .

. . . . .

. . . . .

119

. . . . . .

. . . . .

. . . . .

Chapter 10. Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

108 111 114 115 116

. . . . . .

. . . . .

. . . . .

. . . . .

10.1. Why model? . . . . . . . . . . . . . . . . . . . . . . . . . 10.2. Analytical and systemic approaches . . . . . . . . . . . . . 10.3. Modeling and approximate knowledge . . . . . . . . . . . . 10.4. Modeling in the context of approximate knowledge of CSF 10.5. The steps of the modeling and the necessary validation . . 10.6. Some component modeling carried out in CSF . . . . . . . 10.6.1. Integrating dynamic aspects to check the proper functioning of a component . . . . . . . . . . . . . . . . . . . . 10.6.2. Developing a more explicit but simple model . . . . . 10.7. Simulation of energy systems . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . . .

. . . . .

. . . . . .

. . . . . .

119 121 123 124 126 128

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

128 132 135

viii

Energy Transition

Chapter 11. Conducting the Evaluation . . . . . . . . . . . . . . . . . . . . . . .

137

11.1. Publication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2. Summary of the CSF process . . . . . . . . . . . . . . . . . . . . . . . . . . .

137 140

Part 3. The Practice of CSF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

143

Chapter 12. Challenges of Innovation: Summer Overheating in an Administrative Building . . . . . . . . . . . . . . . . . . . . .

145

12.1. Background information . . . . . . . . . . . . . . . . . . . 12.2. Description of the building . . . . . . . . . . . . . . . . . 12.3. The measurement concept and initial findings . . . . . . . 12.4. Overheating indicators: strict application of the standard . 12.4.1. Proof of need according to standards . . . . . . . . . 12.4.2. Use of the standard by the design office when defining the concept . . . . . . . . . . . . . . . . . . . . . . . 12.4.3. Comparison with the real situation . . . . . . . . . . . 12.5. Building consensus . . . . . . . . . . . . . . . . . . . . . . 12.5.1. Is the indoor humidity in the offices too high? . . . 12.5.2. Is the ventilation through the windows as predicted? 12.5.3. Is the ventilation, even in accordance with predictions and properly used, sufficient? . . . . . . . . . . 12.5.4. Do occupants use night cooling as intended? . . . . 12.5.5. Is the false ceiling an inconvenience? . . . . . . . . 12.6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

145 147 147 149 150

. . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

151 152 153 153 153

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

155 156 156 157

Chapter 13. Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum. . . . . . . . . . . . . . . . . . .

159

13.1. The context of the study . . . . . . . . . . . . . . . 13.2. The Aymon CSF . . . . . . . . . . . . . . . . . . . 13.2.1. Measures and preliminary findings . . . . . . 13.2.2. System modeling . . . . . . . . . . . . . . . . 13.3. Return to Valère . . . . . . . . . . . . . . . . . . . 13.3.1. The building . . . . . . . . . . . . . . . . . . . 13.3.2. The building’s relationship with the weather . 13.3.3. The building’s relationship with the operation of the future museum . . . . . . . . . . . . . . . . . . 13.3.4. The building’s relationship with the technical installations . . . . . . . . . . . . . . . . . . 13.3.5. The resulting indoor climate . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

159 161 163 167 172 173 173

. . . . . . . . . . . . . . .

174

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

174 174

Contents

13.4. Modeling and scenarios: proposal of the concept based on the “Aymon system” . . . . . . . . . . . . . . . 13.4.1. Real in situ simulation of the new use . . . . . . 13.4.2. Virtual simulation of the new use . . . . . . . . 13.4.3. Results of scenarios and proposals . . . . . . . . 13.5. Implementation of the concept and commissioning by the Valais engineering school (now HES-SO Valais) 13.6. Conclusion . . . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

175 175 179 181

. . . . . . . . . . . . . . . . . . . . . . . . . . . .

182 186

Chapter 14. CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings . . . . . . . . . . . . . . . . . . . . . . . . .

187

14.1. Context: from the catalogue of solutions to real practice . . 14.2. Increased complexity of construction and systems techniques well-highlighted by the Sankey diagram . . . . . . . . 14.3. The importance of use and human aspects that are difficult to quantify . . . . . . . . . . . . . . . . . . . . . . . . . . 14.4. The problem of the “performance gap”: modeling to account for the difference in performance . . . . . . . . . . . . . 14.5. A surprising invariant in the functioning of the “building” system: the relevance of I/O relationships and signatures . . . . . 14.5.1. Modeling the thermal demand of buildings . . . . . . . 14.5.2. Investment for infrastructure development and reimbursement from the energy used . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

ix

. . . . . . . . . .

187

. . . . . . . . . .

190

. . . . . . . . . .

199

. . . . . . . . . .

203

. . . . . . . . . . . . . . . . . . . .

208 212

. . . . . . . . . .

212

Part 4. Towards Involved Research? . . . . . . . . . . . . . . . . . . . . . . . .

217

Chapter 15. CSF and Learning Through Use . . . . . . . . . . . . . . . . . . . .

219

15.1. Expertise or contested innovation . . . . . . . . . 15.2. Auditing or putting innovation into practice . . . 15.3. Feedback: in situ evaluation of the appropriation of an innovation . . . . . . . . . . . . . . . . . . . . . . 15.4. Big Data and CSF . . . . . . . . . . . . . . . . . 15.5. The different learning experiences . . . . . . . . 15.6. CSF and learning by use . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

220 221

. . . .

. . . .

223 224 225 230

Chapter 16. CSF, Energy Transition and Involved Research . . . . . . . . .

233

16.1. Current limitations and potential of CSF . . . . . . . . . . . . . . . . . . . . . 16.1.1. The impact of CSF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16.1.2. An evolution over time . . . . . . . . . . . . . . . . . . . . . . . . . . . .

233 233 234

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

x

Energy Transition

16.1.3. Supporting the trial-and-error approach . . . . . . . . . . 16.1.4. The exemplarity of the objects studied . . . . . . . . . . 16.1.5. Energy context and opportunism . . . . . . . . . . . . . . 16.2. Feedback and energy transition: towards involved research?

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

235 236 237 240

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

243

Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

249

Foreword

“Energy transition is certainly one of the most important challenges of our time, but it already started many years ago”. This quote from Bernard Lachal in his final lesson illustrates the added value of the studies his group has carried out on the reality of energy systems for more than 30 years. These studies are designed to evaluate innovative energy systems, but carried out in the traditional organization of construction and energy infrastructure, they allow all stakeholders to take a step forward and better understand the context in which their actions must take place. They also produce accurate data and analyses that lead to the optimization of the energy systems they have set up. This learning through use is essential in order to reproduce and improve the innovations needed to achieve energy transition. The University Centre for the Study of Energy Problems (CUEPE) of the University of Geneva was created in 1978 by Professors O. Guisan, F. Carlevaro and B. Giovannini, at the end of the first oil crisis, to initiate interdisciplinary research in the field of energy. In this context of concerns regarding the sustainability of energy supply, CUEPE quickly became interested in the potential for energy savings and renewable energies. It is worth noting the relevance of these pioneers’ vision, which has now become of primary importance, as concerns about energy resources have been replaced by the environmental effects of energy consumption, particularly the greenhouse effect. CUEPE disappeared in 2006, but a large part of the activities have continued within the new Energy Systems Group. Energy transition is therefore underway. Per capita consumption in Switzerland is declining for both electricity and fuels, with the exception of air transport.

xii

Energy Transition

However, this progress is not sufficient because the climate emergency requires us not only to think about a gradual reduction in the consumption of petroleum products, but also to imagine a solution without CO2 emissions, i.e. without fossil fuels, in the most immediate future. Technologies are already at a level that makes this image credible, but the political consensus formalized in Paris at the COP 21 is unfortunately not reflected in national public policies that would allow these technological advances to be implemented. Politicians in many countries consider energy transition primarily as an additional cost factor that would affect the competitiveness of companies in the context of international competition. However, energy transition is already a source of value creation, as demonstrated by the eco21 program of the Services industriels de Genève (SIG) (Industrial services of Geneva). In this energy efficiency program, launched in 2007, SIG invested 86 million francs in 10 years, more than half of which in direct financial incentives to consumers. They were then able to invest some 193 million francs in goods and services, mainly with local companies. And these consumers were able to reduce their energy bills by more than CHF 290 million, generating a net profit of CHF 140 million. The energy targets were exceeded, jobs were created and consumers spent less, making this a perfect example of value creation, which unfortunately could not be easily replicated in other cantons due to a lack of political involvement. This is the case for many other local initiatives, here as elsewhere. Unfortunately, these best practices have not been studied enough to understand how they have become successful, often overcoming many obstacles. The documentation of this learning through use would thus enable other actors to benefit from these innovations. This is why the analysis of experience feedback is essential and why this approach, initiated by the pioneers of CUEPE and developed by Bernard Lachal and his group, is so important. Let us take a concrete example: the 20-MW GLN lake deep-water network, which was commissioned in Geneva’s international organizations district in 2009. Five years of measurements and analyses carried out by the Energy Systems Group as part of a European project – and the subject of a doctoral thesis – have enabled SIG to improve energy and economic performance in a substantial way, making it possible to exceed the initial objectives of the project and making its replication possible. The GeniLac project was thus launched, targeting a territory more than 10 times larger than GLN’s.

Foreword

xiii

These 40 years of CUEPE’s experience, from 1978 to the present day, are offered to you by Bernard Lachal in this reference book, which will certainly convince you that research made by involved scholars is essential in the field of energy transition. Gilles GARAZI Energy Transition Director Services industriels de Genève April 2019 Marcel RUEGG Institutional Relations Director Services industriels de Genève April 2019

Preface

“I don’t think we can know everything simply through science. It is too accurate and too hard an instrument. The world has a thousand different ways in which it can be experienced in order to understand the sum of its parts… In other words, only the sailor knows the archipelago” [GIO 74]. While everyone is aware of the crucial importance of the development of new energy technologies, particularly those oriented towards renewable energies or the rational use of energy, the importance of their evaluation is only now beginning to be fully recognized. However, assessing the effective interest of these innovations is fundamental to enable them to be truly useful. However, a systematic analysis of methods for evaluating the performance after installation of the various nonconventional energy systems is still lacking. Our current practice and our permanent contacts with stakeholders in the field have also shown us that the way in which the energy efficiency of these new technologies is currently assessed suffers from this lack of a synthetic tool. This book, Energy Transition, therefore has two objectives. The first one is to provide researchers, engineers and anyone working in the energy sector with a summary of methods for evaluating energy systems, the result of several decades of work in this field. The book, based on examples from real cases, is intended to be both synthetic and concrete, presenting as exhaustive a view of the field as possible while at the same time providing a tool that can be easily used by the target audience. The second objective is to break the vicious cycle that still leaves in situ evaluation somewhat neglected today, because it is sometimes considered as a long, apparently expensive, difficult to value and low value work. By attempting to scientifically organize the experience gained over more than 30 years, Energy Transition hopes to convince the reader of the considerable usefulness of the approach, both economically and humanely.

xvi

Energy Transition

This book is organized into four parts. The first one provides a general overview to situate the context in which the types of CSF (case study feedback) that will interest us evolve. After a reminder of some concepts related to energy, its transformation and its consumption, it is necessary to clarify the concepts of systems (energy and technological), innovation, learning through use and finally feedback (CSF). The second part presents the relevant tools of CSF and sets some milestones for their use. In particular, it revisits the notion of measurement, presents different types of models for understanding a system in a quantitative way and also discusses the integration of human aspects. The third part illustrates the practice of evaluation by analyzing some real cases representative of various situations. It situates the use of the tools presented in the previous section in the CSF process. The fourth part is a reflection on the scientific nature of CSF. It is a question of asking how this approach is truly original, of presenting the particular type of knowledge it provides and of situating it in relation to other more recognized approaches such as Big Data. Neither is it fundamental research too far in advance of concrete problems, nor is it applied research too limited to its immediate objectives; feedback should be considered as “involved” research. Bernard LACHAL March 2019

Acknowledgments

This book is the result of more than three and a half decades of collaboration with a large number of players in the energy sector, within the stimulating framework of the University of Geneva. It is therefore impossible to try to thank all those who have, in one way or another, contributed to it. May those whose names I do not explicitly mention not resent me too much. I would like to express my gratitude in the first place to all the owners, project managers, engineers, architects, tenants and users of energy systems who have been scrutinized, the key players in energy transition, for their dynamism, patience and open-mindedness, and without whom REX, in vivo experiments, are simply not possible. Then I thank T. Seal and J. Faessler for the many discussions on the book and their constant support, as well as all the reviewers, especially my colleagues C. Ançay, M. Bonvin, V. Schroeter, J.-M. Zgraggen, J. Khoury and L. Quiquerez as well as S. Schiano for her sharp eye. I have special thoughts for O. Guisan, W. Weber and P. Hollmuller, colleagues responsible at one time or another for the “Energy Systems” group, P. Ineichen, A. Mermoud and E. Pampaloni as well as for all the many other colleagues and students with whom I shared moments of work with, often combined with friendship. Without funding from the Cantonal Office of Energy and the Federal Office of Energy, many REXs would not have been possible. Many thanks also to M. Ruegg, G. Garazi and their colleagues at Services industriels de Genève for the many fruitful exchanges as part of the partnership with the university and for their unfailing financial support – including for this book. Finally, I would like to express my sincere thanks to Catherine Rosselet, my partner and wife, for her continued support and my affectionate thoughts for her, for our children and for our jovial grandchildren.

Part 1

The Context of Case Study Feedback (CSF)

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

1 Energy Transition

The human problem has always been not to create energy, but to transform in a more or less rational way the energy resources available for use. Unlike other natural resources, the Earth is an open system in terms of energy: it receives a permanent and enormous flow of solar energy. This incidental solar radiation is intrinsically a good quality source since it comes from a 6,000 K thermal source; it could therefore be transformed into energy that can be used for our various uses with high efficiency. However, natural annual yields (photosynthesis) are generally well below 1%, and are at most 2.5% for the best plants, such as maize. At the biological level, human energy needs are covered exclusively by solar energy through photosynthesis – 2,500 kcal per day, or 10.5 MJ, which corresponds to an average power of about 120 W. The conversion efficiency of the human “machine”, despite being one of the highest in the animal kingdom, does not exceed 20%: a human therefore has relatively little power biologically and is constantly seeking additional energy (see Figure 1.1, the evolution of world energy consumption since 1800 [MAR 03]). 1.1. The global energy system and its evolution Each year, humanity consumes nearly 15 billion tons of1 oil equivalent, a quantity contained in a cube of about 2.5 km of ridge. This represents approximately 1.8 tons per inhabitant or 2,000 W of continuous power. The price of energy, which has remained relatively stable over the past few decades, although things are beginning to change, can be described as low since heating oil has the same price 1 The various energies are expressed in Gtoe or billions of tons of oil equivalent. One ton of oil equivalent (Toe) corresponds to the energy released by the perfect combustion of one ton of oil. 1 Toe = 42 GJ = 11.70 kWh.

4

Energy Transition

as bottled mineral water, which is a renewable, abundant and regional resource. The inhabitants of the countries of the North therefore very easily have all the necessary energy at their disposal and do not deprive themselves of what is superfluous. For citizens who are unfamiliar with the realities of energy problems, this may seem to indicate a very high abundance of energy, while nearly 85% of the resources used are not renewable (Figure 1.1).

Figure 1.1. Evolution of world energy consumption, according to [MAR 03]

This first observation must be put into perspective by the deep inequalities between the consumption of individuals on different continents. Thus, an average American will consume 8 tons of fuel oil per year compared to 0.3 tons for the citizens of some African or Asian countries. This is an average; we should not compare the energy consumption of the richest 5% of the world with that of the poorest 25%. An estimated 2 billion people live without electricity. The current trend in energy consumption is worrying: a headlong rush at a rate of about 2% per year of growth, i.e. a doubling of this consumption every 35 years and its multiplication by seven times every century. However, we must be careful not to extrapolate this observation too far into the future: in a finite world, growing exponentials also have an end!

Energy Transition

5

Table 1.1 shows the world energy balance in 2015. The figures come from the International Energy Agency and have been adapted to account for hydropower in the same way as nuclear power. Resources

Gtoe

%

Petroleum

4.38

30.3%

Coal

3.66

25.3%

Gas

3.21

22.2%

Fossils

11.21

77.8%

Nuclear power

0.59

4.1%

Hydro

0.91

6.3%

Other renewables

0.50

3.5%

Traditional biomass

1.20

8.3%

Renewable

2.62

18.1%

Total

14.45

100.0%

Table 1.1. World primary energy in 2015, according to [INT 16]

The energy sources are distributed as follows: – fossil fuels provide nearly 80% of the world’s energy (30.5% oil, 25.5% coal and 22% gas); – the nuclear sector (4%) only plays a modest role in global energy supply; – the renewable total is approaching one-fifth (18%), hydropower (6.5%) and especially other renewable energy sources (3.5%) are slowly but surely emerging, while traditional biomass (8%) is largely managed as a non-renewable resource (desertification problem). 1.2. The necessary transformation of the global energy system Several elements show that the current energy system is not sustainable in the long term and that it must evolve.

6

Energy Transition

1.2.1. Fossil fuels: planned scarcity upstream and environmental problem downstream Fossil fuels have exceptional qualities: low extraction prices, ease of exploitation, very easy storage, very easy transport for oil and gas (which does not prevent bad practices, which can be disastrous for the environment). They have major shortcomings (non-renewable resources, emission of various pollutants), but they have been and still are ideal energies for many countries for economic take-off. Their exhaustion will therefore pose problems that must be anticipated at all costs. On the available reserves, controversies are raging. For the pessimist, there are still enough fossil fuels to disturb the climate but never enough to satisfy all the desires of the inhabitants of this planet. For the optimist, and provided we also believe that we are collectively reasonable, there are plenty of them for basic needs and to develop a sustainable energy system, while limiting climate disruptions. The truth is probably in between. In addition to the problem of climate change, following the emission of greenhouse gases, a limitation of fossil fuel consumption can only be beneficial in view of other problems such as urban pollution, the geopolitical risks associated with the depletion of oil resources outside the Middle East or the economic consequences of high energy prices for developing countries.

1.2.2. Nuclear energy: environmental and accessibility issues With regard to uranium reserves, we must be very cautious about the figures for the following reasons [FIN 98]: – these are highly diluted deposits (< 1%), with poorly defined formation conditions; – uranium is a highly strategic raw material and reserve data is often considered a military secret; – many actors are inclined to underestimate these figures: those who are anti-nuclear in order to devalue the entire supply chain, and some pro-nuclear to promote other supply chains (breeder reactors that use 70 times more uranium than conventional reactors, thorium reactors or fusion).

Energy Transition

7

Nevertheless, with current technology, uranium resources are a definite limitation to a significant increase in the number of power plants. Several constraints weigh on the development of nuclear energy: – social acceptability. The specific nature of nuclear risks – very low probability but very high consequence accident risk, long-lived waste management risk spread over an intergenerational period, risk of military proliferation – makes collective preference formation difficult and scientific consensus impossible. However, these two conditions are necessary for a technology to develop; – economic constraints. These include the inadequacy of nuclear technology with the competitive organization of the electricity industries, competition from combined cycle gas turbines and financing constraints in emerging countries. 1.2.3. An overall inefficient system One-third of primary energy is degraded during successive transformations mainly due to electricity production via heat (two-thirds of the losses), the other major losses being the transformers’ own energy consumption and losses during transport and storage. All of these losses will end up as heat. Final energy is often grouped into three uses: mobility (about 30%), electricity (just under 20%) and heat (a good 50%). It should also be noted that the heat lost during the transformations is approximately equivalent to the amount of heat used.

Figure 1.2. From global primary energy to final energy, 2015, according to [INT 16]

8

Energy Transition

1.2.4. A productive and simple-energy vision Figure 1.3 shows the evolution since the industrial revolution of the distribution of primary energy consumed annually into three main types of resources: fossil, renewable and nuclear. In this ternary representation, each axis of the equilateral triangle corresponds to a type of energy and the position of the point projection on this axis indicates its contribution. In 2002, fossil fuels accounted for about 80%, renewable fuels 15% and nuclear energy 5%.

Figure 1.3. Historical evolution of the global distribution between fossil fuels, renewable and nuclear energy. For a color version of this figure, see:www.iste.co.uk/lachal/energy.zip

In the past, we have always had a strong predominance of energy over the others: from almost entirely renewable to almost entirely coal during the industrial revolution, joined by oil since World War II. In the 1970s, the heated debate was about which energy would dominate the upcoming energy scene: the nuclear newcomer or a return to solar energy? This productivist and mono-energetic vision

Energy Transition

9

has not been realized, far from it: today, the fossil has remained predominant, as in 1970, even though we know its limits and impacts well. One observation emerges from this image: a strong inertia of the system, due to the numerous and heavy energy infrastructures that have been developed. They have had a profound and lasting impact on the landscape, structured land use planning and the global economy. In addition, the dominant energies are still abundant, cheap and practical, and can be based on very well-proven sectors. A change in our energy vision is necessary: we must take the problem also on the consumption side to limit total primary consumption and accept that the transition to almost all renewable energy will take some time. Limiting the total amount of carbon emitted can be achieved by reducing the contribution of fossil fuels (substitution), reducing the total amount consumed (sobriety), or reducing losses between final energy and primary energy (efficiency). 1.2.5. Energy transition This transition sets the objective of a more rational use of energy, with the following three pillars: – efficiency: better transforming energy, i.e. the slow but stubborn reduction in energy intensity2, based on energy efficiency throughout the energy chain, from the resource to the end use; – substitution: decarbonizing and denuclearizing the energy system, which involves the gradual and rational substitution of fossil and fissile energies by renewable sources, based on the exceptional qualities of the former; – sobriety: to be more energetically sober, which certainly implies in the end a questioning of certain values on the basis of our domination – necessarily temporal – of nature such as individualism, excessive competition or the use of violence to settle conflicts. These three approaches are necessary, none is sufficient. Efficiency, because it is based on the existing; substitution because the existing is quantitatively important, easy to use and cheap; sobriety, because even as we reduce our consumption, we will have to produce energy. There is also a need to address the transfer of scarcity between energy and raw materials, as new technologies can consume large amounts of certain materials; this issue is the subject of much controversy [VID 18].

2 A quantity that measures the amount of energy required by a society to produce a unit of wealth. It is expressed in kWh/€, kWh/CHF or kWh/$.

10

Energy Transition

Like all human activities, the energy system has already suffered and will continue to suffer the repercussions of technological revolutions such as electronic, digital, material and nanomaterial revolutions, not to mention biotechnology. Some believe in a technological revolution that would solve problems related to energy consumption and production, for example hydrogen, others believe in a clear break in behavior following an energy crisis and more generally raw materials or the impacts of climate change. While such events may occur, they will only result from much more subtle fund movements, which should be encouraged above all and which are based on the slow development of innovations. This approach is related to the silent transformations [JUL 09], which involve: “[…] to get rid of the reactivity to events as well as to the jolts of the news in order to respond to changes, they are barely emerging, in order to prevent their danger, as long as it is not embryonic and easy to reduce, or to encourage their deployment over time, in the long term, when it turns to the common advantage. In other words, in both cases, to intervene discreetly upstream, at the level of conditions, to influence the situation in the desired direction; and not downstream, in the spectacular nature of the action and the urgency of the repair”. As far as energy is concerned, it is therefore a question of encouraging the long-term deployment of everything that has a common advantage. The term “common advantage” applies unambiguously to actions for the rational use of energy and the development of new renewable energies. It is the slow but fundamental reduction in energy intensity that must be targeted at all costs. The result is a double advantage: you gain in kWh/€ (efficiency) and you can still double the bet by sobriety (less €). Similarly, for new renewable energies, all the experience acquired by the pioneers must be disseminated in the community; not only the development of the transformers themselves but also the energy concepts that will make it possible to exploit them to the full and, in the long term, to generalize their use. We must develop a real “silent transformation”, radically changing the energy situation of the future. This transformation is now called the “energy transition”. 1.3. The three concordances Energy is a concept defined by physicists in the 18th Century as “what must be supplied or removed from a material system to transform or move it”.

Energy Transition

11

The physical constraints to which the transformation and development of energy resources are subject are linked to three concordances that must be ensured simultaneously: form, time and place. The purpose here is not to cover the whole concept of energy in physics, but to briefly explain its key elements. 1.3.1. Form concordance Energy exists in different forms which can be expressed under the first principle in a common unit (the joule in the international system), but are not equivalent: – heat; – mechanical energy; – electricity; – chemical energy (related to the binding energies between the nucleus and electrons); – nuclear energy (related to the binding energies of the nucleus particles); – energy electromagnetic radiation (such as light). From its “primary” state, an energy resource undergoes a series of transformations of its form in order to be finally used and provide the desired service. We generally distinguish between: – primary energy, linked to a source available in nature (Sun, water, wind, biomass, geothermal energy, oil, gas, coal, uranium); – secondary or intermediate energy, resulting from one or more transformations (petroleum products, hydrogen, electricity); – final energy, as delivered to the consumer (gasoline, firewood, heat distributed in the district heating, electricity at the socket); – useful energy, as used during the service (electricity for the operation of machines, movement for transport, light for lighting, heat for heating). Through its tools and machines, humans have developed over the course of history many transformers that have enabled us to satisfy our needs and desires: sailboats, water and wind mills, powder, steam engines, combustion engines, gas turbines, electric generators, photovoltaic cells, nuclear power plants, etc.

12

Energy Transition

Energy is governed by two main principles – known as “thermodynamics”: energy conservation (the energy of an isolated system is constant) and entropy increase (the quality of energy degrades spontaneously, any increase in the quality of part of the energy results in a greater degradation of the rest of the energy). The second principle imposes a hierarchy in transformations: heat is a degraded form of energy and its quality is directly related to its temperature. This principle indicates that heat spontaneously passes from hot to cold and imposes a maximum efficiency when heat is transformed into another form of energy; the maximum physically possible efficiency (called Carnot efficiency) is directly related to the temperature of the heat source. 1.3.2. Place concordance A second physical constraint must be solved: the energy must be available at the place of its use. Energy transport has always been a limiting factor until the industrial revolution, which allowed an efficient transport system to develop (oil all over the world, electricity on an entire continent, etc.) with the notable exception of heat, whose transport does not exceed some tens of kilometers in fact. Behind this constraint, there are huge infrastructures (oil and gas pipelines, electricity networks, gas networks, petrol stations, etc.), in terms of size and investments, which give the energy system a significant inertia. 1.3.3. Time concordance Energy must be available at the right time. For uses that depend on seasons such as heating, temporary energy storage seems to be the most appropriate response; it is difficult to imagine today’s energy demand being subject to the vagaries of the climate. Energy storage generally requires investment and causes additional energy losses. For some forms of energy such as electricity, there is not yet a large storage capacity. It should be noted that the time agreement can also be partly resolved by transport for spatial and temporal mutualization. 1.3.4. Economic, social and environmental constraints At the economic level, we will recall that it is not primary energy itself that has a cost, but the various stages of its exploitation, extraction, transformation, transport, storage and distribution. Thus, contrary to what is often heard, solar energy is no freer than coal, oil or uranium. The cost of a solar kWh produced to heat water includes the cost of extraction (solar collectors), transport (pumps, piping, and

Energy Transition

13

electricity) and storage (solar storage). It is also necessary to note the difference between costs and prices; for example, the cost of a barrel of oil (159 liters) is, on average, between about $10 and $20 while its selling price is currently five times higher; the difference between prices and costs being the oil rent. It currently amounts to more than $1.5 trillion per year, the equivalent of all the world’s military budgets. The social acceptability of an energy sector is difficult to identify. It is mainly related to society’s perception of the benefits and risks involved in the various technologies for processing, storage, transport and waste disposal. In people’s minds, there may even be a complete separation between benefits perceived as acceptable and problems perceived as unacceptable. For example, petrol may be considered by society as too expensive, and, at the same time, the climatic consequences of the combustion of fossil resources may be mostly perceived as the consequence of too cheap energy. Any use of energy will generate material flows, transformations and therefore environmental impacts, not to mention the energy itself, which always ends up degraded in the environment.

2 Energy Systems and Technological Systems

The notion of a system is widely used; it has even given rise to a scientific approach, or even a discipline, the systemic one. This notion of a system is often associated with the adjective “complex”. Le Moigne [LEM 90] gives this definition, which is often repeated: “The concept of a System, understood as an intelligible and finalised tangle of interdependent actions, was quickly adopted to define complexity. Does it not serve to express the conjunction of two antagonistic perceptions: a phenomenon that is perceived in its unity, or coherence, or project (solar system, educational system, etc.) and in its internal interactions between active components of which it constitutes the resulting composition”. In this very broad definition, the term “energy system” will mean both: – everything that (from energy resources to services and processing, storage and transport infrastructures) enables society to develop (the global, European and Swiss energy system, etc.); – part of the assembly in interaction with other active components (e.g. a photovoltaic system will thus be an assembly for converting solar radiation into electrical energy, a building including occupants and interactions with the climate will also be considered as a system). Even though it is this second assertion that will be considered in detail in this book, we will always try to situate such subsets in their entirety. Le Moigne defined three main types of functions present in any complex system: the temporal,

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

16

Energy Transition

morphological and spatial transfer functions, which will make it possible to solve the three physical constraints to any energy use, which have already been mentioned. 2.1. Transformers and concordances An energy transformer is a component whose function is to achieve a match of quality, time or place: a certain quantity of energy is available but in a form that is not recoverable or does not correspond to the desired use, that must be moved or shifted in time to be able to be used. It makes it possible to develop a certain quantity of energy of different types (valued input) into a quantity of usable energy less than or equal to the sum of the inputs and of the desired type(s) (valued output) (see Figure 2.1).

Figure 2.1. Basic energy system with a single transformer

All energy transformers are based on a physical–chemical process that allows the desired transfer function. The most commonly used process at present is combustion: nearly 90% of the energy used undergoes this process at some point during its life cycle. Combustion is a chemical oxidation reaction between a fuel and an oxidizer (oxygen in the air), which has the advantage of producing energy (“exothermic reaction”). It concerns all uses of fossil fuels and biomass. This involves transforming the chemical energy obtained by photosynthesis (reduction reaction) into thermal energy at high temperature by an oxidation reaction. It is the first transformation controlled by humans (about 500,000 years ago). The fuels used are all from photosynthesis, present (biomass) or past (fossil) and are essentially composed of three atoms: carbon, hydrogen and oxygen. For fossil forms, known as hydrocarbons, oxygen has disappeared because it has been consumed over time.

Energy Systems and Technological Systems

17

From combustion, many transformers have been developed, from wood stoves to gas boilers, internal combustion engines and aircraft engines. It is not the intention of this book to further describe all these processes, as detailed presentations on them and their translation into transformers can already be found in publications such as [SAR 03]. 2.1.1. Form converters Form adaptation is done in a given place and time. It can concern not only the nature of energy but also an intrinsic quality of a type of energy such as its “potential”. For electricity, the potential is expressed in volts and it is possible to transform the voltage of an electric current, by lowering or increasing it, thanks to a well-named “transformer”. Other characteristics exist that can be modified, such as the frequency of an electric current (from 50 Hz to direct current or vice versa). The case of thermal energy should be mentioned. Temperature is a potential that defines – along with pressure – both the quantity and quality of energy contained in a given mass of matter; it also determines the state of matter and all its other properties. The physical processes involved form a vast field of knowledge, of thermodynamics, which is the basis of the industrial revolution. This discipline is one of the most active in the field of energy; there are very formalized treatments [BOR 84] and very practice-oriented descriptions [GIC 09]. Spontaneously, heat flows from a hot source to a cold source. A heat pump converts a certain amount of low quality heat (low temperature) into a higher amount of higher quality heat (higher temperature). The two principles of thermodynamics require, on the one hand, that the balance of energy must be supplied in “noble” form (electricity, movement), and on the other hand, that this quantity of noble energy is all the greater as the temperature difference between the cold source and the useful heat will be significant. The ratio between heat recoverable at high temperature and noble energy is called the coefficient of performance or COP, whose value is between 2 and 8, depending on the temperature levels and the intrinsic quality of the transformer. 2.1.2. Storage The function of storage is to accumulate energy available at a given time to return it when it can be recovered. The processes involved are very varied and concern all basic sciences (physics, chemistry, biology), while the components developed form an important economic

18

Energy Transition

activity that will continue to develop strongly. The very real environmental impacts are the subject of great controversy. The use of these components is of course to meet the time constraint, as well as the place constraint. For example, batteries for communication devices or electric mobility essentially respond to location constraints: a mobile phone loses much of its value if you have to be near a power socket to use it, not to mention the electric car. Concerning the quality of the energy removed from storage, if it is of the same type; its potential has generally decreased (temperature). 2.1.3. Transport The function of transport is to make energy available in a given place to be used in another place. The processes in place are also very varied and result in many heavy infrastructures such as transmission and distribution power lines, pipelines, oil and gas tankers, and district heating networks. Other important infrastructure is used such as ports, roads and airports. The lifting of the constraint of place that transport allows is at the root of the industrial revolution and the globalization we are currently experiencing. Developed economies have been able to feed on fossil fuels available in increasingly distant places. However, transport also plays a fundamental role in solving the time constraint by sharing it: energy available in a given place but in surplus can be used in another place and with a time lag corresponding to the transfer time of the network (from a few milliseconds for electricity to a few hours for heat). The spatial transfer of energy requires a certain amount of the same or other energy to overcome the resistance to movement existing in any network: internal electrical resistance, pressure drops for fluids, truck diesel to transport heating oil. In addition, an exchange with the environment inevitably takes place and results in heat losses. In total, the major networks lose about 10% of the energy transported. To conclude, there is no overlap between the three major physical constraints (form, time and place) and the three types of transformers (form converters, storage and transport). Thus, and contrary to widespread opinion, the massive development of fluctuating renewable energy involves not only its storage but also its transfer for the sharing of its value between users. 2.2. From the transformer to the energy system An energy system transforms an energy resource into recoverable energy according to a series of transformations (time transfer, location or quality) that solve

Energy Systems and Technological Systems

19

the three physical constraints previously seen. It is translated into a technical system composed of an arrangement of basic energy systems. For example, the solar photovoltaic system will transform solar radiation into usable electricity (230 V AC on a local grid). As shown in Figure 2.2, a first transformer adapts the quality of the incident energy from its electromagnetic nature to electrical energy contained in a direct current of a few volts. A second transformer adapts this direct current into an alternating current of 230 V, and perfectly in phase with that of the local network, which will distribute this electricity (concordance of location).

Figure 2.2. Simplified photovoltaic system

A second, more complete example concerns a solar thermal system that will transform incident solar energy into heat needed to heat the hot water used by the occupants of a single-family home. Figures 2.3 and 2.4 show this pathway from primary resource to final use; the heat produced, once used, ends up in the sewers as warm water. The latter can also be used as a heat pump heat sink either in decentralized form before the treatment plant or after the plant in a decentralized way.

Figure 2.3. Solar installation for domestic hot water (DHW) production

20

Energy Transition

Figure 2.4. Decomposition of a solar thermal system into subsystems

The essential elements are: – solar panels, which qualitatively transform radiation into heat; – the spatial transfer circuit of this heat produced to the rest of the system, including various other components such as an electric pump, piping, valves, electronic control and a heat transfer fluid; – a heat exchanger, which allows the transfer of this heat from the heat transfer fluid containing antifreeze to the domestic water part (form and location transfer);

Energy Systems and Technological Systems

21

– a storage, used to achieve the time match by accumulating the heat produced to release it in a timely manner; – an auxiliary heat power, to ensure time matching (in case of bad weather or strong extraction); in this example, an electrical resistance degrades the electricity into heat; – finally, the domestic hot water distribution circuit (location agreement), the energy required for the movement of the fluid being obtained by the pressure existing in the water network. This pressure is obtained from electrical energy. This system is very simple, but it has a “fractal” aspect that gives it a complex character. By zooming in, we can break down all the energy processes that take place into a series of systems that accommodate them, for example the operation of a selective black layer that absorbs solar radiation but reflects infrared far or heat transfer between the solar absorber and the heat transfer fluid. On the contrary, by zooming out, we can integrate the electric or heat pump sector, whose cold source is waste water, and we can also take a closer look at the use of domestic hot water. If this system is installed on the roof of a school and supplies the changing rooms of the gym, it will become even more complex, as shown in Figure 2.5. Several other technical systems will be integrated into the building and will be more or less linked. Several other inputs will be made at different levels: water, financial and human resources for all the technical installations as well as for the operation of the building and the services it can offer. In addition, the school is integrated into the educational system, itself into society and the democratic system that manages it.

Figure 2.5. Interweaving of different systems in a school

22

Energy Transition

This decomposition allows us to simplify a reality that we know is complex; complexity that reappears through the complementary energies necessary for the functioning of the system. For example, for a wood-burning system, diesel is needed to transfer the natural resource wood in the forest to delivered firewood, and electrical energy to extract and transport the wood-energy resource, during delivery and combustion (various conveyors, pumps, fan) and then at the valuation stage (circulation pump, regulation of the distribution system, etc.). This analysis is repeated by considering the output of a system as the input of a downstream system. For example, residual heat obtained after the production of electricity in a household waste incineration plant can in turn be transformed into a resource. The associated system could be a district heating system and its development, the connected buildings. This can be continued until the final performance. All of these interconnected chains in intense and moving interrelationships would form the physical energy system, in perpetual change. In this chain treatment, the systems often referred to as “production” and those referred to as “consumption” are treated in the same way: a bulb is a form converter (electricity to light) in the same way as a photovoltaic panel (reverse transformation) or a nuclear power plant (nuclear energy to electricity). 2.3. Effectiveness of resources and effectiveness of results The transformer itself has its own functioning which will vary both intrinsically, according to the quality and maturity of the technological system, and in a suffered way, according to the operating conditions imposed by the energy system of which it is a component, whether on the side of other inputs, outputs or other components (operation in use). Thus, the evaluation of a transformer must always be done according to these two points of view (see Figure 2.6).

Figure 2.6. Transformer and energy system

Energy Systems and Technological Systems

23

In other words, a poor transformer working in good conditions can perform better than an excellent transformer working in poor conditions. This link between operation and use is an important point that will be discussed in detail later. Energy efficiency refers to the operating state of a system for which energy consumption is minimized for the same service rendered. There are many other definitions that relate to the same idea. We consider the balance of a simple energy system with input, output and losses to the environment (Figure 2.7):

Figure 2.7. Assessment of a simple energy system

We then have quantitatively: /

1



/

with: – efficiency: the efficiency of the energy system; – input: the energy entering the system; – output: the recoverable energy; – losses: the energy leaving the system and that is not recovered. Efficiency can be measured in percentage terms if the input and output are energy expressed with the same unit (e.g. gas/heat) or with a more complex unit if the output is already measured in terms of performance; for example lumen/watt for a light source, the lumen measuring the light power. This definition can be extended by integrating all inputs (monetary, human, raw materials, etc.); we will therefore have a series of efficiencies such as the relationship between the output and each of the inputs. The gain in information on the quality of

24

Energy Transition

the system is reflected in a multiplication of performance indicators and raises the problem of their aggregation. This also exists in the case of the validation of several outputs; as such, the combined heat and power generation and the difficulty of qualifying the system solely on the basis of the overall efficiency, which does not take into account the difference in nature between electricity and heat; and in the case of the use of several inputs, as in the case of heat pumps, whose efficiency in the strict sense of the first principle is always very close to the value one, whatever the quality of the system may be. In this case, the COP is the performance indicator used, but it is always higher than one. In the case of sobriety, the output of an energy system is reduced by a voluntary reduction in the service required. If the energy system remains unchanged, sobriety does not spontaneously imply an increase in the efficiency of the energy system: the simultaneous decrease in input and output does not necessarily result in an increase in the ratio between these two quantities. The technical optimization of the energy system is not an end in itself but a means. It may be considered more rational to consume less energy with less performance even though the system becomes less efficient – but, in general, the two approaches (efficiency and sobriety) are not incompatible in practice. Substitution consists of completely or partially replacing one input with another. For example, one can substitute gas with solar energy to preheat part of the domestic hot water. Again, the technical optimization of the energy system is not an end in itself but a means. If it seems more optimal to substitute part or all of the input energy with energy of a different nature, even though the system becomes less efficient, the overall use of energy will be considered to have become more rational. However, there is a plethora of comparisons in the literature between the efficiency of a combustion boiler (close to 100%) and that of solar thermal panels (30–60% depending on the use), and these comparisons should be considered as not very rational. To judge the relevance of a substitution operation, the comparison of the two energy systems (without and with substitution) will require a value judgment: the value that is given to each energy input of different quality. Therefore, effectiveness should not be limited as a judgment criterion, and the distance between intention and outcome should be considered primarily. This is referred to as “effectiveness”, a concept used by the systemic: the relationship between intention and realization [LEM 90].

Energy Systems and Technological Systems

25

In conclusion, the transformer is the key component of the energy system, but it is also the result of a technological system that elaborates, builds and develops it. The technical system, of which it is the main component, aims to transform a certain quantity of energy of one or more given types into recoverable energy, at the lowest possible cost with the lowest possible environmental impact, both during its manufacture and during its operation. This component is generally accompanied for its operation by other components, themselves derived from a technological system. This ready-to-use technical system will itself be included in a larger energy system and so on, until it forms the energy system of a city, region, country, etc.

3 The Innovation Process

The various components of an energy system are derived from a technological system that designs, produces and distributes them. The appearance, evolution and continuous improvement of these components – in a word innovation – are the keys to success for energy transition. According to Alter [ALT 00]: “Innovation is always a story, that of a process. It makes it possible to transform a discovery, whether it concerns a technique, a product or a conception of social relations into new practices. But this process is not mechanical, not every discovery is transformed into innovation. A discovery may well remain in the state of invention”. 3.1. A well-defined process This slow collective process of moving from discovery to innovation has been and remains well-studied; many theories have been described, tested and refuted to explain both the why and the how of innovations. We will simply repeat the most “classic” representation of its deployment and that of a development according to an S-shaped logistics curve (Figure 3.1). The deployment along this “S” is classically separated into three steps: – the pioneering stage. This first step allows the discovery to be transformed into an invention, then into a prototype and a low-pass product. At the end of this stage, innovation works more or less correctly and is used by what are called “pioneers”, often composed of the “inventors” themselves and a handful of the convinced.

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

28

Energy Transition

This step is fundamental; it must allow technology to prove its worth, and the inevitable problems associated with the youth of innovation do not discourage these pioneers, who are facing teething problems. These “pioneers” are often described as deviations from the norm;

Figure 3.1. The innovation process, according to [ALT 10]

– deployment by appropriation and increasing adoption. At the “pioneering” stage, there is a phase of ownership by the actors involved in innovation. This passage is delicate because, from a marginal and often latent practice, innovation must “make its place”, convince people beyond the “believers”, persuade, percolate, diffuse and be adopted – in the terms commonly used in the literature to describe this stage. The adoption of innovation by all stakeholders is the key to the development of new technologies in the energy sector; – institutionalization or “standardization”. Innovation has become a standard practice, often becoming the rule or, in the field of energy, just one possible option among many others. Innovation continues to evolve, but at a slower pace. This institutionalization step should normally follow the appropriation step. It can be difficult to anticipate and formally institutionalize an innovation that is imperfectly accepted and adopted by the actors involved; otherwise, it could lead to what Alter calls a “dogmatic invention”. By “institutionalizing” a technology, we can think, for example, of a legal obligation to adopt it or a prohibition of other sectors that compete with it but that, for one reason or another, we want to eliminate. Examples include the obligation to use solar panels to heat part of the domestic hot

The Innovation Process

29

water (Barcelona, various Swiss cantons, etc.), the progressive ban on incandescent lamps at the European level or the obligation to provide certain energy diagnostics for buildings or, more broadly, parts of the country. On the contrary, the legal obligation in the field of energy is a powerful tool for developing and accelerating virtuous circles, but we must be aware of the difficulty of the exercise, and even the dangers of rejection. The denomination of these three steps may change depending on the authors. Thus, in the field of energy, Criqui et al. [CRI 00] have a more economist vision that is called: emergence, consolidation and competition. More detailed descriptions can also be found, up to five (Caron, cited by Alter [ALT 10]) distinct sequences! However, the number of steps is totally arbitrary since it is a continuous process. What can be learned from the above is that, without sufficient ownership by the actors involved, the invention does not lead to results and innovation is rarely practiced or rejected. Another confusing point for a physicist is the multiplicity of taxonomy in trying to classify innovations. As Martin [MAR 00] points out, “noting the heterogeneity of innovations, most authors who have written on the subject have tried to characterize them by multiplying the qualitative ones: minor, major, revolutionary (J. Schumpeter), incremental, radical, systemic, paradigmatic (C. Freeman); pseudo-innovation, improvement and fundamental (G. Mensch)”. Martin himself proposes to classify innovations in the field of energy technologies into three groups: – radical innovation: this involves either the use of a new resource for an existing valuation, or a new valuation from a resource already used, or both. The central point is therefore the transformer, which is necessarily of a new type. Examples include photovoltaic solar cells that convert solar electromagnetic radiation directly into electricity or light-emitting diodes (LEDs) that do the opposite. LEDs had been around for several decades but the light emitted was used for information: first as an indicator of device activation or level, later in traffic lights at road junctions. It is only very recently that “lighting” valuation has become possible. The result of these radical innovations is an increase in the variety of possibilities for energy supply; – the incremental innovation of the same sector: this only concerns the processor, which is gradually improving with its increasing adoption. This is not a linear causal relationship but a circular causal relationship: if the technology improves due to greater use, improved performance will also make this technology more attractive. This is a “virtuous circle” that must be initiated and maintained. It is not because

30

Energy Transition

a technology is effective that it is used, but because it is used that it will become effective; – major innovation in a sector: this is a qualitative leap in transformer technology, a fork that does not apply to the incremental. This may involve a new process or technology of part of the innovation. An example is the vacuum solar thermal collector, which increases the transformation efficiency at low sunlight or high temperature. Without fundamentally changing the use (the thermal requirements up to temperatures of about 120°C in our example), there is expansion since the operational improvements will positively influence the use environment and we can thus get closer to radical innovation. Similarly, thin-film solar cells are a major innovation in photovoltaic technology. It should be noted that radical innovation and major innovation are often confused. Be careful not to confuse invention with innovation, as defined here. Still according to Alter [ALT 10], four points separate them: – temporality: if the invention is a time-bound event, generally the starting point, innovation is the process that follows it: “It is the story of a state of permanent tension between the possibilities represented by the invention and the collective choices that are gradually drawn from it”; – the feeling of society: “Invention is generally considered as ‘good’. Innovation has nothing to do with this. It represents the way in which people affect, in a situation, a meaning to this good”; – the absence of a relationship between the intrinsic quality of an invention and the importance of its diffusion: “What allows innovation is therefore not the abstract potential represented by the novelty but the possibility of assigning it a use, taking into account the social system in which it operates”; – the relationship to efficiency: “If the technical invention generally refers to the idea of efficiency, innovation always partially rejects the potential represented by the invention”. The development trajectory of energy technologies reflects the long and continuous learning process that reduces costs and improves efficiency, reliability and safety. A great many studies have been devoted to this learning process. Empirically, it has been found that the quantity of inputs required to manufacture a unit of a product decreases as its cumulative total quantity increases. This relationship is called the “learning curve” because it is unanimously considered that this efficiency gain is due to the learning achieved during the repetition of the manufacturing process. The first study was conducted in 1929 and focused on the

The Innovation Process

31

decrease in the number of hours required to build aircraft cockpits in the United States. This method has been extended to all technologies; those concerning energy do not disregard it. The notion of a learning curve has been extended to the notion of an “experience curve” by considering all the factors necessary for manufacturing, aggregated by the cost. The term experience highlights greater complexity and therefore greater uncertainty about the mechanisms that cause prices to fall. The literature contains a very impressive number of both empirical and theoretical studies on this subject, which apply to a wide variety of technological systems. It is important to note here that time does not explicitly intervene; it is through the cumulative production that necessarily increases with it that time acts. In these curves, an indicator of the quality of the technology or technological system is linked to the cumulative quantity produced since the beginning of the technology. Quality can be represented by the transformer’s efficiency, its specific cost (€/kW), the quantity of CO2 produced during its construction, the number of hours required, etc. The cumulative quantity can be based, for example, on the number of transformers produced; in energy technology, the total nominal power of the transformers produced is often used (see Figure 3.2 for an example of electricity storage, where the experience indicator is based on the cumulative energy storage capacity). In general, the quality indicator improves by about 20% when the cumulative quantity doubles. The quality indicator is generally expressed in such a way that it decreases when the quality improves, and therefore its value decreases with the quantity produced. The nuclear sector is one of the few examples where the learning factor is negative (costs increase with the quantity produced) [GRU 14]. There are multiple uses for these curves. First, on a prospective basis, they make it possible to determine by extrapolation the cost of the component if a future production is assumed, or the production to be activated to obtain a given cost. Then, in complex technical-economic models, the simple relationship resulting from an experience curve makes it possible to determine quantity and price as a function of time. They are also used retrospectively, making it possible to analyze and understand the processes involved in improving production, mainly the different learning processes. Finally, it allows a summarizing comparison of different technologies, as shown in Figure 3.2.

Figure 3.2. Experience curve for electrical energy storage [SCH 17]. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

32 Energy Transition

The Innovation Process

33

3.2. Limit of these curves in the context of energy systems When we look at energy technologies, several remarks must be made about the construction of these curves, an exercise that is very difficult to carry out. For experience curves, the selling prices of the produced unit are most often considered rather than its cost, because it is a more easily obtained quantity than the cost of production, which is often considered very sensitive. The difference between a cost price and a selling price depends on the market in which the technology is evolving; the lessons that can be drawn from the experience curves are therefore more difficult to identify. They are sometimes used as a basis for comparing the price of energy obtained by different energy sources that use different primary resources. These curves characterize both the resource and its price evolution, as well as the evolution of the technologies themselves. Several perimeters can be considered: – What is the system involved? Is the technological system producing the component transformer? Is the technical system installed and ready to operate with its expected production? Is the energy system in operation with its actual recovered production? – On what scale is cumulative production achieved: producer, country, world, other geographical unit? The case of photovoltaics is a good illustration of these points. First, we can look at the evolution of the transformer itself, with the cost of the produced output unit in a given nominal condition as a performance indicator. In Figure 3.3, the learning curves of two globalized photovoltaic panel technologies are shown. The performance indicator is the selling price for each watt produced in the standard condition; they are expressed in USD and adjusted for average inflation. The cumulative production values are distinguished between the two technologies. The logarithmic scale hides the fact that crystalline silicon technology (c-Si) is more mature than the other (Cd/Te), since cumulative production is 10 times higher. The decrease in both curves is quite marked and indicates a price decrease of 22% when cumulative production doubles. This curve suggests that the second technology is inherently better economically because, for the same development, prices are about three times lower. The use of the cost-effectiveness indicator alone should not mask other important parameters directly related to physical transformers that can decisively influence the increasing adoption of the transformer: in the example of photovoltaics, the improvement of aesthetics through reduced thickness, lack of frame or panel coloring; a simplification of the mechanical and electrical assembly of the panels; a reduction in specific weight.

34

Energy Transition

Figure 3.3. Learning curve of two photovoltaic panel technologies [IRE 12]. Price indication 2017 by the author

If we restrict ourselves to the photovoltaic transformer, we remain at the level of the single component. It is at the foot of the building and still locked in the packing cases. A first step is to consider the technical system, i.e. all components and their installation. Compared to the previous scope, not only the other components, but also the design and installation of the system (called “BOS” or Balance of system in the specialized literature) were added. These BOS now represent nearly 60% of the total cost of a photovoltaic system: the cost per installed solar kW will therefore depend less and less on the cost of the panels themselves. The possible innovations at the BOS level are very important and concern technologies (not only those of the panels but also those of the quick connectors, inverters), standardization (voltage of the panels, quality level of the current supplied, etc.), organization (integration of the panels on the roof by the roofer, project management by specialized companies or the electrical companies themselves, etc.) or financing (third-party investors, contracting). The learning curve is well-known, but it is still very dependent on the location and weather, local practices, and the technical and organizational environment. This takes place until the possible globalization of practices if this technology lends itself to “turnkey” delivery, which is not so obvious. For example,

The Innovation Process

35

Figure 3.4 shows the average price of electricity sold in the United States and the cost of production of electricity from the photovoltaic sector, both of which are aggregated for photovoltaic electricity production. It shows an impending convergence1.

Figure 3.4. Learning curve of the cost of the photovoltaic electricity produced, with, in superposition, the evolution of the electricity price [NEM 06]

We remain here at the level of the physical system, with an electric generator ready to operate, but which has not yet produced any kWh! The costs presented for photovoltaics are expected, standardized costs (annual sunshine hours, temperatures, maintenance costs) of a well-monitored system that operates without problems and whose entire electricity produced is recovered. In the end, the only system that will interest us is the long-term energy system, which includes recovery. The operating environment is getting longer and longer, as the system itself needs to be monitored and maintained regularly; other systems connected to it can also transform and strongly influence the operation of an energy system (e.g. the degree of insulation of a building can lower the operating temperature of a heat pump and give the opportunity to increase its coefficient of performance, and in our case, it can be a new building that provides shade over part of the installation). The scope of use is also expanding significantly, on the one hand, towards the financial and economic aspects (it is necessary to pay for the investment and sell the energy produced in such a way that the operation is economically credible and the risks acceptable), and, on the other hand, towards the organizational aspects (it is necessary to take charge of the monitoring of the system itself and its interactions with the rest of the electricity system). Hidden costs such 1 The fact that the two experiment curves are only referred to photovoltaic production makes it difficult to use. The growth of solar electricity is indeed much higher than that of total production. In addition, a cost of production is compared to a retail price.

36

Energy Transition

as those related to security of supply and fluctuating production appear at this level, which should be taken into account in one way or another. In this field, organizational innovations are emerging, such as the concept of a virtual power plant, which consists of combining several complementary power generation systems and activating demand management, making it possible to ensure electricity production at all times with a strong renewable component (this is the case for wind turbines in northern Germany, photovoltaics in the South, a gas turbine between the two and electricity demand management). 3.3. Operation and use The case of the photovoltaic system presented above clearly shows the importance of the use and loss of efficiency of an invention during the innovation process, when we want to transform an invention into a proven and well-adopted technology. In his sociological approach to innovation, Flichy [FLI 03] emphasizes the importance of the distinction between “operating framework” and “usage framework”, which he defines as follows: “Function and use are two sides of the same coin. The reference framework can therefore be subdivided into two distinct but mutually articulated frameworks, the operating framework and the usage framework… The operating framework defines a set of knowledge and know-how that is mobilized or can be mobilized in the technical activity. This framework is not only that of the designers of a technical artifact, but also that of the builders, the repairers and also that of the users. They can mobilize this framework when they want to open the ‘black box’, tinker or modify the machine”. The usage framework is related to everything related to the use of the artifact in question. First of all, the designers, even though they are closely linked to the operating framework, have imagined a certain use, which will more or less coincide with the use made of it by users. At the level of these, two types of use exist: technical use and social use. In the first case, it is a matter of technically integrating the technology into a larger system, for example an engineer will incorporate solar thermal panels into a larger building heating system. In the second case, technology will have to be integrated into a broader social system, for example by making it mandatory by law, in order to achieve energy policy objectives, the production of at least 30% of the energy needed to heat domestic hot water with solar collectors.

The Innovation Process

37

As seen above, there are many actors involved and while some are more related to operation and others to use, it is important to keep in mind that there are no designers, engineers or repairers who operate, on the one hand, and users who use the technology in question, on the other hand. The links between operation and use are much more complex than often stated. If innovation consists of the continuous improvement of the technical, economic and environmental performance of energy transformers, then we cannot be satisfied with a binary logic for operation: in working order or in failure. Efficiency can vary almost continuously between optimal performance (that given by the manufacturer under good operating conditions and often considered as the transformer’s reference) and performance well below that which is expected. This variability in performance can come from several causes: – technical malfunction: failure, component maladjustment, wear, insufficient monitoring; – technical misuse: poorly chosen component, incorrect sizing, poor synergy with other system components; – poorly controlled social use: very often, there is too high an expectation of a technology, such as wanting to impose a high rate of renewable energy for domestic hot water production or the willingness of a homeowner to put solar panels in difficult resource conditions, as in the case of a poorly exposed or heavily shaded site. The malfunction of a technology due to inappropriate technical or social use does not make it possible to characterize its intrinsic effectiveness. Thus, if misused, the best energy transformer can give poor results without its own functioning being at stake: it is not the transformer that needs to be improved, but its use. The proper functioning of an energy transformer requires appropriate use, under penalty of disappointing results, which will most often be blamed on the transformer itself. In the innovation process, it is necessary to take into account the dual position of the transformer as seen in Chapter 2: technological innovation and the production of energy that is often referred to as “new”, even though it is electricity or heat. This is a definite difficulty, but it is also an opportunity, because we have two levers at our disposal to encourage the diffusion and adoption of these technologies: the attraction for technological innovation (technology push) and the social demand for new energies (market pull). It should never be forgotten that energy is not a service in itself, but is used by a technical system to provide a service (heat is used to heat bath water for washing or relaxing; electricity is used to lighting for reading).

4 Case Study Feedback, the Basis of Learning by Using

All of the innovations present in the energy system lead to a triple diversity: the multiplication of the actors involved, the diversity of possible energy pathways for a given use, and the varieties of technologies for each pathway. 4.1. Innovation in energy systems There are many actors involved in the development of innovation: – the producers of the transformer component, for example the fuel cell that converts chemical energy (hydrogen, methane, alcohol, etc.) into electrical and thermal energy; – the users of this component who will design and build the energy system that contains it, for example a design office that integrates a fuel cell into a district heating system and of which the electricity obtained is recovered by the electricity grid; – energy system managers, who must valorize the recovered energies, for example the district heating company that owns the installation and manages it; – the consumers of recovered energy, who can have a great influence on the possible valuation of this energy, and therefore on the functioning and technicaleconomic efficiency of the transformer; in our example, these will be the owners of buildings connected to district heating and their tenants. The first two groups of actors define the transformer market, and the last two groups determine the energy market. It is understood that the separation between the two systems is not complete, and that strong interactions between the two exist and must even be developed. These interactions include the strong relationships between

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

40

Energy Transition

the two central groups – the design office that develops the energy system from components and the owner who takes it on order, or the consequences that a misuse of a new transformer can have on the market if its efficiency or the price of the resulting energy are not met. The choice of options will therefore concern: the energy resources available for the desired use, the possible valuations for a given energy source and the type of transformers available on the market. It is thus possible to choose to heat part of the domestic hot water using the sun, then choose the type of transformer associated with it (type of sensors, brand, integrated or modular system, combined solar gas offer from the local gas distribution service or a brand of gas boiler or its heating company). These possibilities make it more difficult to make choices because of the great differences in the characteristics of the pathways. None of them have simple qualities, and the determining criteria can be contradictory. The information about them from which choices are made is very disparate. While mature sectors have a robust and reliable knowledge base, the energies they use are mainly fossil fuels or electricity; their prices are not very predictable and are perceived as structurally sound. Their brand image is bad from an environmental point of view. On the contrary, renewable sources are often well known and accessible – even though they are spontaneously variable – but transformation technologies are maturing and the information about them is considered unreliable, often from the circle of pioneers. Final energy is perceived as more expensive than conventional energy, despite uncertainties about the future price of traditional energy and the widespread view that it can only increase in the long term. Robust and transparent information on innovative energy systems is the sine qua non for resolving these tensions and developing their collective ownership and diffusion in society. Curiously, this problem of learning by collective appropriation and increasing adoption of all the actors involved in the functioning and use of an innovative energy system is little studied. Thus, Martin [MAR 00] recognizes the importance of this learning: “None of the innovation processes that have renewed energy technologies over the past 30 years have been completed. Neither the high level of performance achieved nor the indisputable competitiveness with existing technologies has slowed down the innovative activity of firms. Driven by fierce intra- and intertechnology competition, all are pursuing advances… In this process, learning by use is essential.” This same author highlights two important points. Unlike standardized massmarket products, the use of energy technology depends strongly on the local context, especially since it is close to use and linked to local resources that are often very

Case Study Feedback, the Basis of Learning by Using

41

different from one place to another. The importance of existing infrastructure may or may not favor the development of an innovative energy system: for example, the existence of a district heating network or a company capable of designing, building and operating such infrastructure. If we look at a complete energy supply chain, we have the diagram in Figure 4.1.

Figure 4.1. Complete range of low-energy lighting based on wind electricity

A “horizontal” part (Flichy’s operating framework) concerns the trainer themself and their own technological, economic, organizational and normative development. However, a transformer is never used for itself but for the secondary or final energy it delivers. There is therefore a second process (Flichy usage framework) that is intrinsically linked to the first. If the transformer itself has to impose itself in relation to its competitors, the energy it supplies is itself in competition with other pathways for a given use, often well established and mature. This diagram is repeated for the two steps of transformation of energy from wind to electricity and from electricity to light.

42

Energy Transition

This dual process is fundamental in the learning phase and involves two worlds: the one that operates the system and the one that uses the output. We can see the whole as a complex mechanism where the two movements must be coordinated: that of technological innovation which, through learning linked to its increasing adoption, will make the transformer more efficient, and that of use, since the transformer user will in turn have to value the energy produced and will do so all the more easily if the transformer is efficient. At this stage, the following hypothesis can be made: the development process of an innovation will be all the more complex if it is close to use. Indeed, if the transformer’s output is a secondary energy (electricity on a grid, for example), the actors involved in its recovery are limited in number and the constraints of use are absent. On the contrary, if the transformer is very close to use (solar system for DHW, passive system for heating or cooling buildings), the constraints of use will be decisive and the user must absolutely appropriate the innovation at least partially. This is the opposite of the previous case, where the use of electricity does not change according to its origin for a consumer coupled to a network. 4.2. Case study feedback Case study feedback (CSF) consists of comprehensive and in-depth evaluations of innovative energy systems in real-life situations, i.e. implemented within the framework of the traditional organization of construction and energy management. It should be noted that the innovations in question are generally related to technological aspects (mainly through the technological system and other components of the energy system), but they also concern financial, commercial and organizational aspects. These are the reasons why they are carried out in close collaboration with the various actors concerned (investors, prime contractors, designers, design offices, users, study funders, professional and academic circles concerned, public authorities, component manufacturers, etc.). The ultimate goal of this type of work, which is complementary to the research and development work carried out by other laboratories, is to create a knowledge base of practices and realities in the field of energy innovation, to stimulate it through a feedback process between academia and practitioners (incremental innovation process). In the innovation process, we are at the key stage of deployment by appropriation and increasing adoption of the actors involved. The aim is to complete the two well-known learning modes: learning by doing when manufacturing innovative elements and learning by searching during R&D activities of the products involved with learning by using, of which feedback is the key. To achieve the CSF of an energy system, a balance must be found between different factors: a manageable scope of study – which concerns the spatial,

Case Study Feedback, the Basis of Learning by Using

43

temporal and disciplinary dimensions; sufficiently ambitious objectives – negotiated between the CSF sponsors and the team that carries it out; and adequate resources in terms of finance, team and experience. It should be noted that the most common CSF performed is the one that systematically follows major technological accidents. These are systems that have had major failures and the purpose of these CSFs is primarily a safety issue. In the field of energy, we can mention the exemplary work done on the Fukushima accident [TRA 18]. This approach will not be discussed here. 4.2.1. CSF classification test The objectives of an evaluation can be very diverse. Nevertheless, we can try to classify CSF into five main categories: expertise, audit, feedback, benchmark and MetaCSF, the definitions of which are given in the following sections. As with any categorization of complex elements, this classification has its share of arbitrariness and some studies may fall into two categories. The proposed classification is based on the normative aspect of the objective, on the one hand, and on the measurement concept underlying CSF, on the other hand. A representative example is given for each type of CSF. These considerations will be discussed in detail later. 4.2.1.1. Expertise or the contested innovation This is the most prescriptive approach, as it involves providing a clear-cut opinion on an energy system. Often, its operation is a problem in a given application and it is a question of confirming this state of affairs, possibly explaining it and defining responsibilities. It is therefore a question of being very responsive and being able to identify the necessary skills in a timely manner. The duration is generally short (1–6 months). An example can be found in the reference [ZGR 05]. 4.2.1.2. Auditing or putting innovation into practice This is the most common form of evaluation. From a general point of view, the audit of an energy system is a process carried out by an expert pursuing one or more objectives using appropriate tools. The most developed is the audit energy, which consists of identifying and highlighting opportunities to increase the efficiency of the energy system. It is about building on the knowledge accumulated by the team. The duration is also generally short, in the order of 3–6 months, but the follow-up of the measures implemented can be extended over several years. An example can be found in the reference [MER 07].

44

Energy Transition

4.2.1.3. Case study feedback: evaluating in situ the appropriation of an innovation This is certainly the most attractive exercise because there is no normative character; it extends over a “long” period of time (often several years) and develops in a stimulating framework of risk taking and innovation. For example, several actors have committed themselves to the implementation of an innovative energy system but wish to be supported in this process by an ex-post evaluation, including technical aspects – such as measuring and estimating yields or modulating transformers in real conditions – as well as economic, organizational and social acceptability aspects. The “common good”, the desire to contribute to the building of a more sober and just society and the pleasure of being a pioneer animate most of the actors, and the demand for independent evaluation shows a willingness to go beyond the alibi stage. We are in the process of learning by trial and error, in observing the transformation of the energy system, which is essential but necessarily slow. Depending on the specific response time of the energy system to be evaluated, it should be between 6 months and 5 years. An example can be found in the reference [MER 14]. 4.2.1.4. The benchmark It is a comparison between different energy systems based on a few indicators. Their choice and the way they are calculated are in principle based on an initial CSF that provides a good knowledge of the energy system in question. Depending on the sample size and the complexity of obtaining the data for the indicators, the duration may vary from a few months to a few years. An example can be found in the reference [KHO 14]. 4.2.1.5. MetaCSF A special case of CSF may arise with already existing data (e.g. recording of system control data, data from a previous CSF). It makes it possible to extract a great deal of knowledge from an existing system with a lower investment or to multiply CSF with the same investment. In this case, the objectives of the CSF can only be adapted with the measures available. The benefits of not having to take measurements must be tempered by the constraints carried over to the next steps (analysis, modeling, etc.). Moreover, the system to be analyzed remains an object that is not very concrete, not very lively or unlike traditional CSF, where the human aspects are important. The ideal scenario is to have already followed an energy system of the same type and to complete the knowledge acquired through this type of approach. The team in charge must therefore have a good experience of CSF of similar systems. Depending on the complexity of the systems and the responsiveness of external stakeholders, it takes between a few months and two years. An example can be found in the reference [FAE 17].

Case Study Feedback, the Basis of Learning by Using

45

Table 4.1 summarizes the different CSFs. Objectives of the project

Format

Typical duration

Expertise The contested innovation

Highly standardized

Report generally confidential Conflictual

1 to 6 months

Auditing Innovation in practice

Standardized Appropriation of good practices

Report often public Application-oriented

3 to 6 months

CSF in situ Innovation in development

Knowledge Dissemination and appropriation of good practices

Publications Can be integrated into a training course

6 months to 5 years

Benchmark Inspected innovation

Comparison between different systems Dissemination and appropriation of good practices

Publication Can be integrated into a training course

3 months to 3 years

MetaCSF Shared innovation

Knowledge Dissemination and appropriation of good practices

Publication Can be integrated into a training course

3 months to 2 years

Table 4.1. Attempt to classify the different CSFs

4.2.2. CSF content The process of evaluating an energy system usually involves three phases: – measurement or data collection phase, mainly quantitative; – phase of analysis of these data, in order to understand the processes involved, to model the energy system or some of these components, to simulate it and possibly to carry out a sensitivity study of various parameters and operating conditions or configurations; – the evaluation phase itself, resulting from the synthesis of the various analyses and other considerations specific to the problem at hand. The tools required to carry out CSF, their operation and uses are described in detail in the chapters of Part 2.

Part 2

CSF Tools: Operation and Envisaged Uses

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

5 The Human Context

Human aspects are an integral part of any energy system: they are the people who decided to build it, designed it, manufactured it, put it into service and managed it for others who do not always use energy directly but via an appliance, home or a vehicle, operating thanks to this itself. This distinction between operation and use was discussed in Part 1, where the importance of CSF in improving practices was highlighted, as well as their beneficial effect on the deployment of innovation. 5.1. Why the human aspects? 5.1.1. In vivo rather than in vitro As Brisepierre mentions in his work on the place of human beings in the performance of high energy quality buildings [BRI 15], “the adoption of new technical solutions or organizations is not self-evident, it is confronted with systems of actors in place, who have a greater or lesser interest… The desired changes must be integrated into social dynamics, which do not always go in the same direction”. Moving from the laboratory (in vitro) to practical implementation (in vivo) is in fact an embrace of these aspects. The desire to put the actors back on a hypothetical straight path – that of the decision-maker, the designer, the builder, the politician – should not be part of the objectives of CSF. The integration of human behavior must make it possible to better understand the transition between planned and actual operation, to be able to judge the intrinsic functioning of the system and the effect of the use made of it on its mode of operation. In science, in vitro experimentation is commonly referred to as “manipulation”: care should be taken not to “manipulate” people in in vivo experimentation. For example, in the case of the observation of a failure or malfunction not detected by the operator, it is essential to allow some time to

50

Energy Transition

observe the responsiveness of the persons in charge of the system and then to inform them so that the system not only recovers correct operation but also understands the original management malfunction. 5.1.2. The importance of objective information in the field of innovative energy systems Stakeholders directly involved in an innovative energy system (ES) generally want to know whether the efforts made have been successful, whether it is possible to improve the functioning of the system and, in the case of good results, to be able to communicate on the basis of robust and credible information. However, some of these actors may have some reservations about a thorough evaluation of the system they have financed, designed, built or are responsible for. In addition, there is a more general need in society for objective information – or information judged as such – on the effectiveness of a solution, an invention, or on the opportunity for the large-scale dissemination of an innovation limited to the pioneer circle. The academic community can be considered as a well-placed actor to carry out these evaluations, since the issue of conflicts of interest is somewhat less relevant than elsewhere, and the search for objectivity should in theory prevail over any other consideration. To produce quality work, the scientist involved in CSF must be aware of two pitfalls that are the opposite of each other. The first pitfall can result from a personal commitment to certain environmental issues, a commitment that can lead to idealization. However, an overly positive vision (“it’s free, perfect for the environment and it works without any problem”) leads to the creation of technological bubbles leading to a big loss of credibility when they burst, often at the end of the pioneer-credulous stage. On the contrary, the second pitfall is that of criticism, which can result from an excessive desire to “appear credible” and from which results an overly negative vision (“it’s nice but it’s expensive and it doesn’t work”), which constitutes a powerful obstacle for actors willing to innovate both at the time of decision-making and at the time of financing, where these technologies are penalized by unfavorable financial conditions. Producing and disseminating credible and balanced information that is useful not only to actors close to an innovation but more generally to society as a whole is a difficult, collective and progressive task. It takes time to organize, carry out and disseminate the results of each CSF and the experience accumulated by the team is a good guarantee of quality and credibility. The latter is not decreed; it is built, fed and maintained.

The Human Context

51

5.2. Who are the actors involved and how are they involved? The actors involved in CSF can be divided into three categories: – those involved more generally in the different stages of innovation (discovery, invention, pioneer stage, increasing diffusion, adoption), of which the energy system is the pilot example or a more common example in the diffusion phase; – those involved in the various stages of development (decision, design, manufacturing) and use (commissioning, management and users) of a particular energy system; – those directly involved in the implementation of the CSF itself. 5.2.1. Actors involved in the innovation process The actors involved in the global innovation process are divided into several categories: – policies: elected officials, voters, offices, civil servants, standards, laws; – financial actors: banks, state innovation support offices, investors; – technical actors: design offices, manufacturers, assemblers, system or energy managers, professional associations; – citizens and users; – opinion formers: journalists, information centers, popular scientists. CSFs provide an opportunity to deliver scientifically established information and enable targeted and “objective” communication to be developed. Some actors important for the development of innovation may become disconnected from the reality on the ground. Integrating them into CSF or making this reality accessible to them can only be profitable. We are thinking here of the actors who prescribe standards, labels, regulations and laws, the content of “lifelong” training, public aid or the required rates of return. 5.2.2. Actors related to the particular energy system There are five sub-groups of actors corresponding to the five main stages of the construction of the system: – decision-makers, who may be an individual, a group of individuals in the form of a cooperative, a private company, a public entity such as the state or a municipality;

52

Energy Transition

– designers: architects, design offices, engineers, financiers; – builders, i.e. professionals in the fields of energy, civil engineering, construction, IT, etc.; – maintenance and energy managers; – the users of the service provided by the recovered energy. 5.2.2.1. Decision-related actors To embark on an innovative project, its decision-makers can be positively influenced by the knowledge acquired and information disseminated as a result of previous CSFs that have evaluated a nearby energy system or part of the planned system. It is therefore important that knowledge spreads in an understandable way to this circle, which is not always obvious to people from an academic background. An extremely promising approach is the development of the energy management profession in companies and institutions. As mediators between the community of energy professionals within or outside their organization and decision-makers, these managers can become an essential link in innovation in society. These actors should also be aware of the possibility of having CSF carried out, which can be co-financed either by other ES actors or by those related to innovation. The benefits to them are threefold. Feedback from experience makes it possible to verify the proper functioning of an innovative installation and to set up the necessary long-term monitoring in an efficient manner, both technically and financially. In addition, CSF often leads to improvements not only in energy efficiency but also in durability, comfort and overall profitability. Finally, the information collected is an excellent basis for communication work on the energy system, improving the image of the contracting authority. 5.2.2.2. Design-related actors Designing an innovative energy system leads to a multiplication of stakeholders in this phase because the structures that usually handle it do not yet have all the skills to carry out all the necessary studies. Thus, to design high energy performance buildings in France, Gournet and Beslay [GOU 15] note that even the design of a small administrative building with a high energy standard required the intervention of about ten stakeholders during the design phase. These authors speak of “upstream hypertrophy [which] refers to the complexity of producing an energy efficient building”. These additional experts often perform specific calculations such as the level of daylight in the various spaces, expected summer temperatures, etc. The labeled approaches further complicate the design work, since everything must be explicit and verifiable.

The Human Context

53

In this context, CSF can also make a significant contribution. Those previously carried out on similar objects have made it possible to document the catalogue of available solutions, to enrich it and to specify the right uses. They are often a source of inspiration and are the very basis of incremental innovation. A new CSF makes it possible to continue and reinforce this global approach, to verify ex post the choices made to design the system: technologies adopted, sizing, coupling between the different subsystems, etc. It also makes it possible to check ex post the tools used ex ante such as software, certain parameters or assumptions, and finally to check ex post the relevance of standards and labels and to develop them on a robust basis. The information collected is an excellent basis for communication work. 5.2.2.3. Manufacturing-related actors If building an innovative ES is not fundamentally different from building another object, two aspects must be taken into account. On the one hand, small imperfections, changes of components for another close but not strictly identical one, or incorrectly understood or misunderstood assembly instructions can significantly affect the energy performance of the system, even though they do not prevent it from operating. On the other hand, the increased complexity of the design often leads to a stronger interaction between the different stakeholders in this constructive phase and learning is often necessary to ensure good coordination between them. CSF makes it possible not only to check ex post the high quality of the work, the cost, the difficulties encountered during assembly, but also to specify their effects on performance. The knowledge base obtained contributes to the evolution of the professions. 5.2.2.4. Actors involved in the management of ESs In fact, the previous remarks concerning the actors involved in manufacturing can be repeated here. CSF will also answer the following two questions: – Is the information received from the beginning of the process sufficient to take control and manage the system in the long term? – How to reduce to an essential minimum the complexity of procedures, adjustments and ultimately costs? 5.2.2.5. Actors related to the use of ESs In the case where the studied energy system extends into the supply chain to the end user, the latter will interact strongly with it; the actual use of this ES will reveal an unexpected operation that may be far from optimal. This is particularly the case for buildings, where the responsibility for the performance gaps often observed is

54

Energy Transition

most often (always?) placed on the “bad” use of the inhabitants. Sometimes it seems that the ideal would be buildings without occupants, because they would behave “badly”. As the sociologist Brisepierre rightly points out in the case of new buildings with high energy quality: “In all cases, the inhabitants rarely behave as the designers predicted and adapt their practice.” [BRI 15] The answer given to this problem is generally a doubly prescriptive approach. On the one hand, it is necessary to support the inhabitants towards a standardized use and, on the other hand, the responsibility for this support is entrusted to building professionals as an additional task. CSF allows you to answer these few questions: – To what extent are these practices bad practices rather than a bad evaluation of these practices upstream, during the design phase? – How far should the uses be changed, for example the indoor temperature and ventilation of buildings? – How can practices be improved throughout the chain? – Who should change user behavior and how? – What use should be considered in the standards? On what basis should it be defined? 5.2.2.6. Relative impact of the five groups of actors This impact depends first of all on the degree of innovation in the system under study. If we are at the beginning of the innovation process (pioneers), the key players will be mainly designers and builders. If we place ourselves in the diffusion– appropriation phase, then all the actors will be involved. Then, the extent of the system in the global energy chain (from the resource to its use, see Part 1) and its position more or less close to the end use will give more or less weight to the end-user: the closer we are to primary energy, the less important the user will be because other cascading systems will intervene. Extreme examples are the first photovoltaic systems of a few tens of kW in the 1990s, coupled with electrical distribution networks of a hundred or a thousand times greater power, and the passive cooling system consisting of the opening of windows at night by the occupants of an administrative building when the outside air is cooler than the inside (see full description in Part 3). 5.2.3. Actors involved in the implementation of CSF Each CSF performed can be integrated into a specific training (master’s degree, doctorate, continuing education) and also forms the basis for an expert’s learning,

The Human Context

55

who can thus become increasingly experienced. Their attitude must be open and comprehensive: some of the previous actors are not familiar with the techniques or with energy issues. 5.3. How to take into account human aspects in CSF This section concerns the definition of the scope and objectives of the study and the necessary means, taking into account human aspects. This is indeed the key step that will not only define the essential characteristics of CSF but can also decide on its very existence. The challenge is to find a balance between a manageable scope of the study, sufficiently ambitious objectives and adequate resources (finance, team, knowledge and experience). 5.3.1. The perimeter Three dimensions must be considered: spatial, temporal and disciplinary. The spatial dimension covers the physical extent of the system. A first aspect to be defined is that of the system itself within the global energy chain. For example, for the study of a renewable and spontaneously variable electricity production system, is the scope limited to the system itself or does it include its impacts on the rest of the supply chain, either upstream (other systems necessary for security of supply given the fluctuating aspect of its production) or downstream (adaptation of electricity use downstream to limit its effects)? This will of course depend on the relative size of the system to be evaluated and the size of the injection network. If the latter is of a much larger size, the CSF itself will not be of much help, but the knowledge gained from this CSF can be integrated into a specific study on this aspect. If the size of the spontaneously fluctuating system and the size of the network to which it is coupled are comparable, then consideration should be given to extending the study spatially to other producers and users. In the context of energy systems linked to buildings, it is necessary to clearly specify the extent to which the building itself is concerned (technical system and energy supplied to the building, demand and quality of the exterior components, use of the inhabitants and its impacts, etc.). The second dimension of the perimeter to be defined is time. Is it only the stage of use of the energy system that is considered? Is the design and construction of the object included? Is this done ex post if these steps have already taken place? How long should it be evaluated? There is no single answer, but it can be pointed out that for systems related to seasonal climate variations (renewable production, buildings), the time unit is the year. In this case, since two units of time are a minimum to make comparisons and the few months necessary for preparation and synthesis should be

56

Energy Transition

taken into account, it appears that the minimum duration of the CSF is about 3 years. This is an important point since it conditions the response time of the CSF considered as feedback for learning. We are therefore on rather long time scales, which must be accepted and understood by the actors involved in innovation, especially the prescribers. In the case where seasonal aspects are absent or well known, the CSF may be of a shorter duration. The third dimension to be addressed is the disciplinary scope. At a minimum, an energy assessment is essential, which may include the analysis of the intrinsic and in-use functioning of the various subsystems, the modeling of these subsystems and the functional links between them. This section describes all the necessary tools. An economic evaluation is often carried out if we are in the diffusion–appropriation stage of innovation. In the early stages (discovery-pioneering prototypes), an economic evaluation may be discouraged as this very insignificant information at this stage of immaturity may take on an overestimated importance, for example by comparing it to mature systems. If an evaluation is still conducted, it should be accompanied by an estimate of the expected cost for a more mature system. Sociological evaluation can make a significant contribution to CSFs dealing with highly use-sensitive systems, such as high energy performance buildings or systems with a high level of spontaneously variable resources. The interest of including a new approach is to better understand what conditions behavior. Limiting oneself to an energy evaluation makes it possible to quantify the effect of uses, but does not make it possible to qualify or understand them. Several CSFs on pilot buildings in France have successfully integrated sociologists, which in the long term makes it possible to accelerate the implementation of good practices insofar as the motivations of the various actors are better understood [BIS 15]. This approach was undertaken for the study of an eco-neighborhood located in the suburbs of Geneva, which includes a sociological component in addition to the usual energy and economic components for a 5-year CSF (currently in progress). 5.3.2. The objectives of the CSF The objectives are closely linked to the definition of the scope; they must be clearly and explicitly stated. They are primarily intended for CSF sponsors and can be broken down into main objectives, sub-objectives and expected results. They concern both the evaluation of the system itself and the general development of energy technologies and their use. It is very rare that the in-depth study of an object, even a limited one, does not have a more general scope. Similarly, this step of defining the CSF is an opportunity to integrate as many actors as possible, which should therefore be well defined.

The Human Context

57

5.3.3. The resources The resources are of three kinds: financial, human and the knowledge–experience of the team. If the financial resources available are not sufficient, we must try to find other resources (material contribution, work or service of some partners, sharing of resources with other projects). A good project, already largely funded, has relatively easy access to the missing balance. Human resource needs are the problem of the required skills. For the energy aspects, it is necessary to know the general physics, as well as to have a sufficient level of mathematics to be able to choose in a sufficiently wide register the tool adapted to the problem. Theories describing the processes that significantly intervene in CSF will, at first sight and in a contradictory way, be all the more sophisticated if the subject of study is not very extensive. Thus, we can use the equations of heat and the imposing mathematical tools that go with it to study a simple geometry – for example, the diffusion of heat in a medium located around a cylinder subjected to cyclical temperature conditions [HOL 02]; we can no longer afford such mathematical luxury if we study the thermal losses of a lake water network, at temperatures that can fluctuate rapidly and measure several kilometers [VIQ 12]. It should be noted that very large systems – a district thermally coupled to a lake or river water system, the wood or agro-energy pathways as a whole – are also very well approached by researchers from the natural sciences who are ready to update their knowledge in physics and mathematics. The creation and maintenance of an interdisciplinary team is essential for this type of study. Economic evaluation is usually limited to simple cost-effectiveness calculations that do not require special knowledge of economics since they are taught in all engineering schools. 5.3.4. The team’s experience A major issue is to preserve the know-how accumulated during the “experience resulting from feedback”. The knowledge acquired during the CSF has been explained in the form of reports and various publications, but most of the practical and methodological knowledge to carry out CSF remains implicit and therefore disappears when a person leaves the team or with the files that are treated as recycled “material” during moves or compaction. The need to maintain a team therefore requires that a party stay long enough for all the implicit knowledge, know-how and interpersonal skills to be activated. In addition, students, trainees or transients will eventually leave the group to work most often in the same field of activity but in another organization. In this case, implicit

58

Energy Transition

knowledge will enrich general practice and recharge its batteries, as well as hybridize with other ways of doing things, of being, of thinking and in other intervention environments. The main problem is therefore the maintenance of a stable group; this requires sufficient ribbon resources to avoid accepting projects solely to ensure livelihoods. 5.3.5. The follow-up group The creation of a follow-up group for each CSF of a certain importance has given very good results. It is composed of the main actors involved in the CSF and has one to two dozen members. It meets regularly – typically twice a year – from the beginning, for the discussion of objectives and methodology, and until the end, for a critical reading of the report before its publication, through the long development of research, where the intermediate results will be presented and discussed, possibly redefining the objectives, reorienting part of the study, stopping the study of certain aspects or, on the contrary, deploying others, exchanging, encouraging and sometimes debating quite strongly on certain points. The continuous information provided saves actors directly involved in the evaluated energy system from waiting a few years before reading a report in the form of a sentence. The appropriation by the various actors of the CSF conclusions is more assured because they are better shared. The transparency of the process increases credibility, defuses potential conflicts and ultimately enriches the experience. While this results in additional work for researchers, with the pressure that these regular meetings and the expectations they generate, the outcome is really very positive in terms of improving the quality of the work, its usefulness and relevance. Another point worth mentioning is the learning for young researchers, which results from the management of these regular meetings.

6 The Energy Context and the Sankey Diagram

6.1. A drawing is better than a long speech The first known flow diagram was the work of Charles Joseph Minard in 1869: a figurative map of the successive losses of French soldiers in the Russian campaign in 1812–1813. It is very often considered to be one of the best statistical data charts ever created (see Figure 6.1). It represents, in a two-dimensional document, six pieces of information simultaneously: – the size of the Napoleonic army; – the distance traveled; – the temperature; – the latitude and longitude; – the direction of travel; – the location of the army relative to specific dates. The use of two colors (the advance in light brown, the return in black), the simplicity and the absence of any excess information give a great strength to this graph. To maintain the same level of information in a written text, it would probably require several pages of writing in which the reader would quickly lose the general vision. This diagram also has the advantage of creating in the person who first reads it an emotion (the size of the arrows which decrease drastically, indicating the dead and disappeared; the extremely cold temperatures, the suffering). Then, the information is there: by examining the document, we see the dramatic crossing of the Berezina, the parallel movements of troops; we imagine the logistical problems and the tribute paid by the inhabitants of the regions crossed. We want to know more about it.

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

60

Energy Transition

Figure 6.1. C. J. Minard’s diagram of Napoleon’s Russian campaign in the original French. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

The Energy Context and the Sankey Diagram

61

This type of diagram is called a “Sankey diagram”, in honor of the Irish engineer Matthew Henry Phineas Riall Sankey (1853–1926), who used it 30 years later to describe the energy efficiency of a steam engine. Widely used to track flows of energy, matter or money, it is an essential tool for evaluating energy systems: “Sankey diagrams are suggestive flow diagrams that help to quickly visualize the distribution and loss of material and energy in a process. The width of the lines used in the drawing of the diagram is proportional to the amount of material or energy”. [ALT 17] In 2012, the Agency for Renewable Energy and Energy Efficiency (AEE) and the Swiss Gas Industries Association (SGIA) published a document [AEE 12] on the possibility of storing surplus renewable electricity by first converting it into methane (SNG (substitute natural gas): renewable or synthetic renewable methane), a material that can be easily stored and transported, and then converting it back into electricity when the need arises. The aim is to balance electricity production and consumption in time and geographically and thus to allow the development of spontaneously variable renewable electricity sources, such as wind and photovoltaics. This process is called Power-to-Gas (P2G). In this document, it reads: “P2G (Power-to-Gas) uses 99% of the current used for electricity and about 1% for methanization1, with outputs distributed as follows: about 62% SNG (synthetic methane), 12% residual heat at high temperature and 27% residual heat at low temperature. Currently, well-designed CHP generation2 using waste heat can achieve energy efficiencies of 60% or more. However, the electricityelectricity efficiency is half as good as pumped storage or batteries. The conversion of gas into electricity must therefore comply with restrictive conditions, for example to balance the load (in the short term) or compensate for a lack (local). From an energy point of view, it makes much more sense to use the SNG: renewable as a fuel.”

1 These few lines contain an error in terminology: the process that transforms electricity into methane (CH4) is called methanation and not methanization. The first is purely chemical (transformation of hydrogen obtained by electrolysis of water into methane by reaction at high temperature and high pressure with carbon dioxide (CO2)); the second involves the use of micro-organisms to degrade plant matter in the absence of oxygen, and the gas obtained is biogas (mainly methane). 2 Combined Heat and Power generation: energy system producing electricity and recovering the waste heat produced.

62

Energy Transition

This text is unclear. For example, the sentence “well-designed CHP generation using waste heat can achieve energy efficiencies of 60%” contains several ambiguities. On the one hand, it is not known whether the residual heat thus named is that resulting from the methanation process (see the previous sentence of the text) which activates the CHP generations, thus allowing additional electricity production, or whether it applies to the recovery of heat waste produced by the CHP generation activated by the SNG. The first possibility can be excluded as it does not seem very sensible to produce electricity at a time when it is in excess. On the other hand, the term “energy efficiency” is not very precise. If it is the sum of the heat recovered and the electricity produced, it is a rather low efficiency. If it is an electrical efficiency, it is high and corresponds to the best current technologies (combined cycles or fuel cells). Similarly, claims about the possible uses of the SNG thus produced are questionable: once it is reintroduced into the gas sector, there is no reason to treat it physically differently from other methanes (natural gas, biogas); only labeling it as “renewable gas” (e.g. windgas if the electricity comes from wind turbines) could have an advantage for financing the operation. Thus, it is curious to read here that the gas industry can promote CHP generation for natural gas and discourage it for synthetic methane by advocating simple combustion! In Figure 6.2, the corresponding Sankey diagram has been drawn.

Figure 6.2. The Sankey diagram of Power-to-Gas-to-Power. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

The Energy Context and the Sankey Diagram

63

The diagram shows the two electrolysis and methanation processes with heat losses (recoverable or not), obtaining methane with 61% efficiency, to which are added the 12% recoverable heat. After storage (assumed perfect), and in a very symmetrical way, we find only 37% of the electricity, but at the chosen time, and about 12% of additional recoverable heat. With 100% wind power produced at a time when it was not needed, P2G was able to store and return 37% of this energy to the electricity system, plus 12% heat in the location and at the time of methanation (summer, high-wind weather) and 12% in the location and at the time of CHP generation (winter, low-wind weather). These explanations are superfluous as long as you are a little used to reading – or better still drawing – these diagrams. This avoids ambiguities and sterile discussions about use! 6.2. Design, development and operation Any Sankey diagram cannot escape the two principles of thermodynamics: – first principle: energy conservation; – second principle: energy can be spontaneously transformed into heat, heat can only be transformed into energy with a thermal machine and only if part of the incoming heat is sacrificed in a well at a lower temperature (see Chapter 1). The construction of the diagram is based on the description of the system to be evaluated, itself composed of different subsystems that interact. For its development it requires, on the one hand, a good understanding of the different subsystems and their interactions and, on the other hand, the evaluation of the input and output flows of each subsystem considered. It is both a valuable guide to developing the evaluation methodology and an important outcome of the study, and is useful at all stages of the evaluation. Its elaboration is based on the technical diagram of the installation: it is the graphic translation of the functioning of each subsystem considered and simultaneously of all the interactions taken into account. It is based on the systemic analysis seen above: resource, transformation and valorization. For each transformer/adapter considered, incoming energies are considered as “resources” and outgoing energies as recoverable energies. 6.2.1. The importance of precise terminology The bio-methanization system concerns the transformation of the biomass energy resource (chemical energy) contained in green household waste into biogas (also chemical energy). Several concepts must be distinguished according to the location in the energy chain.

64

Energy Transition

At the resource level, the recovered green waste involves upstream issues related to waste management, sorting and recycling. At the transformer level, the bio-methanization process is the result of both ancestral experience and multiple research projects still underway in many laboratories. The bio-methanization plant implements the process of the same name; it is the result of a complex technological system, where there is strong competition between manufacturers. This process requires other energy inputs such as electricity, fuel and heat. At the recovery level, the biogas produced is accompanied by the non-methanizable material waste from the process; the latter takes with it about half of the energy contained in the treated green waste. The methane contained in this biogas is the same molecule as that of natural gas. Its reinjection into the gas network requires a specific purification installation, with obviously additional inputs and waste. Finally, the designation “green gas” for the biogas produced allows better exploitation thanks to the label system. In the documents on the green waste treatment chain, there is considerable confusion between these terms, which are used indistinctly from their meaning, and sometimes used as synonyms to avoid repetition. To confuse an installation with a process is to deny all technological development and its difficulty; to not distinguish between a waste and a labeled energy is to deny the whole energy chain linked to it with its technical, financial and human aspects. Figure 6.3 shows the Sankey diagram of a standard methanation facility for a population pool of 100,000 inhabitants, each recycling an average of 100 kg of kitchen waste per year. In this diagram, the first energy flow connects the resource (organic waste) to a methanization unit; it is the main energy input. From there, an unwanted part (digestate and water) comes out, which is the inevitable result of the biochemical process, and the desired part: biogas. Heat losses are added. The valorization of biogas into methane requires a purification unit to extract various gases such as CO2, H2S, water vapor, etc. Symmetrically, a composting unit will allow material recovery in the form of compost, while further reducing biological energy by about half. The other half is heat related to the formation of carbon dioxide CO2 which is found in sensitive (compost is at about 60°C) and latent form (evaporation of water contained in the material being composted). This diagram does not include plant electricity (shredder, pumps, press) and any fuel used by machines used for material transfers (backhoe loaders, trucks, etc.). The waste resource is taken into account on site, but transport up to it is not counted, which should have been the case if we wanted to compare, for example, a centralized solution with a solution based on several smaller, probably less efficient but decentralized sites.

The Energy Context and the Sankey Diagram

65

Figure 6.3. Energy flow diagram for an installation of methanization treating kitchen waste from 100,000 inhabitants. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

Quantitatively, the energy flows here are expressed in GWh and each tonne of waste contains slightly more than 1,000 kWh of energy. We read that 25% of the incoming energy comes out as valuable material in agriculture (chemical energy for subsoil microorganisms). At the end of the process, the gross production of the installation is: – 5.3 GWh of methane (accounted for in energy) which can be labeled “green gas”, i.e. 44% of the waste input in energy; – nearly 4 GWh were released as heat into the environment or one-third of the input; – the remaining energy (almost a quarter) is found in the form of compost that can be recycled in agriculture. This material must absolutely be recovered, otherwise the direct incineration (without selective sorting) of this green waste would have been more interesting in terms of energy.

66

Energy Transition

This gives the general context; the box can then be opened as in the diagram in Figure 6.4, which is the result of the evaluation of a real system. The fractal aspect appears here because, again, we could open each box and zoom in repeatedly, up to the microscopic level. 6.2.2. Balance failure Energy conservation is a simple principle but actually poses several problems in transcribing the results of an evaluation into a Sankey diagram. In each subsystem, the energy balance must be strictly balanced. Two cases therefore arise. Either all quantities have been evaluated separately or it is rather unlikely that the energy conservation law will be respected. Simply add an additional output or input called “balance error” to the diagram. Far from devaluing the diagram, as long as this error remains compatible with the announced accuracy, it gives credibility to the work and gives a valuable indication of accuracy. Alternatively, in the second case, only part of the quantities has been evaluated and the balance corresponds to what has not been measured. For example, in a heat pump, it is common to only measure the electricity activating the compressor and one of two heat levels (at the entry or exit), and the second is then deducted by sum or difference. Implicitly, it is assumed that no heat loss occurs.

Figure 6.4. Flow diagram of a bio-methanization installation in Geneva [AEB 10]. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

The Energy Context and the Sankey Diagram

67

6.2.3. To avoid having a chilling effect The second principle of thermodynamics can also pose some problems, especially with systems whose performance is cold (“negative energy”), which must be treated carefully. Let us take the case of a refrigerator. The “trick” is to draw the diagram while keeping the energy as positive (see Figure 6.5):

Figure 6.5. Flow diagram of a refrigerator. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

The direction of the arrows has been reversed (from right to left), and you can of course flip the graph (mirror) to keep the direction “from left to right” (see Figure 6.6).

Figure 6.6. Flow diagram of a refrigerator, “returned” version of that in Figure 6.5. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

68

Energy Transition

This avoids the error often made (see Figure 6.7):

Figure 6.7. Incorrect flow diagram of a refrigerator. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

The latter diagram does not comply with the second principle: heat at low temperature is transformed into heat at high temperature and electricity, which corresponds to a decrease in entropy. And since we generally measure only two of the three quantities, we falsely deduce the third one to obey the principle of energy conservation. For example, the diagram in Figure 6.8 shows some of the energy flows of a large administrative building on the Place de Genève.

Figure 6.8. Sankey diagram of a large administrative building in Geneva. The air conditioning part (circled in red) is incorrect (see text). For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

The Energy Context and the Sankey Diagram

69

The “air conditioning” section (circled in red) shows the design error mentioned above: the heat extracted from the building is transformed into electricity and high temperature heat dissipated in the cooling towers. It should also be noted that the absence of installations in the diagram poses the problem of changing the type of energy and therefore the colors. Thus, the electricity for activating the cooling units becomes cold (in blue) before activating the cooling machine.

6.2.4. Shape: graphic rules The practical rules related to the shape of the graphs not only apply to the Sankey diagram, but also to all graphs and tables. To present data in a relevant way and based on Tufte’s book [TUF 01], four main principles can be stated: – the objective is to communicate complex ideas with clarity, precision and efficiency; – the graph must give the viewer (not the designer) the greatest number of ideas, in the shortest time, with the least amount of ink and in the smallest possible space; – it will almost always be composed of several variables; – it must tell the truth about the data. These principles are applied to the form as follows: – first of all, show the data; – maximize the ink used to represent the data in relation to the total ink used for the graph; – minimize redundant information; – don’t rush the work; a good graphic requires time. In this sense, the Minard diagram on the Russian campaign (Figure 6.1) is particularly successful. For the Sankey diagram, the following tips can be added: – avoid, as far as possible, cross-flows;

70

Energy Transition

– resources are located to the left of the graph for horizontal orientation (the most logical in a text), at the top in vertical orientation (the most practical in a diagram of a certain complexity); – the spatial order of energies should correspond as much as possible to their quality rather than their quantity: electricity at the top and low temperature heat at the bottom. The information on their respective quantity is already given by the width of the arrow, so there is no need to double this information; – remain simple in the illustration; everything that can be removed without impeding understanding must be removed; – colors help to ensure a quick and unambiguous understanding of the diagram as long as they are logical (red for heat, green for biomass, etc.) and stable (if the document has several diagrams, the color code must remain the same). In addition, it is desirable that the whole be aesthetic and sober (avoid rainbow effects); – the names of flows and processes must be correctly chosen: has an identifiable loss or balance defect or both been assessed? A single process is rarely evaluated alone; in general, one of the possible installations based on this process is evaluated: this must be clearly specified to avoid generalizing the results to all installations of the same type; – the values of the flows shown on the diagram must be easily apprehensible: 11.2 GWh rather than 11,222,365 kWh (or, a little more visibly, 1,120,000 kWh). The first number corresponds to an assumed accuracy of 1%, which is already excellent for this kind of exercise. It is up to the author of the diagrams, graphs and tables to round the values. If they do not, they explain without realizing the accuracy of the evaluation: in our example, the values expressed in kWh suggest that they are evaluated to the ten millionth. The diagram in Figure 6.9 is a counter-example; it was found on the Internet and is part of a presentation at an international conference on solar energy storage. Here, the width of the arrows does not correspond to the values, which is the basic principle of such a diagram. The image given is in perfect contradiction with the reality of the figures. Solar energy appears to be more important than that from gas, but it is half as low! Similarly, the energy flow towards the storage is especially overestimated by the width assigned to the arrow on the drawing, etc.

The Energy Context and the Sankey Diagram

Figure 6.9. Sankey diagram of a seasonal stock of solar energy, found on the Web. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

A corrected version is proposed in Figure 6.10.

Figure 6.10. Sankey diagram of the previous corrected figure. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

71

72

Energy Transition

Note that the first two transformers (gas boiler and solar collectors) are not treated as a subsystem due to the lack of information on incoming energies (quantity of gas and sunlight arriving on the collectors). They only appear as an indication of the transformer involved (i.e. they are excluded from the scope of the analysis). The poor performance of heat storage is clear (which has since been significantly improved by the installation of heat pumps), as well as the significant losses of district heating, which are equivalent to solar production. 6.3. Uses The Sankey diagram is used in all phases: in the definition of objectives and measurement concept, in the analysis of a system and to synthesize certain results. It is also used at all scales. As an illustration, heat pumps can be considered at three levels: – as devices available on the market, for example installed in a building near Geneva (Figure 6.11); – in a more complete energy system, such as the case of Riehen district heating, which is partly powered by geothermal energy (Figure 6.12); – on the scale of an energy policy, to show the role they could play in enabling the mass recovery of renewable energy (Figure 6.13). For the device (Figure 6.11), the measurements are accurate and the operation is detailed. The objective is to properly quantify the functioning of the heat pump in use. Figure 6.12 shows what happens when we look at the place of the heat pump in a larger energy system: it logically loses importance in the representation. In the prospective study, the objective of the diagram is to show the potential role of all energy sectors involved in heating buildings in the canton of Geneva. The diversity of heat pump use is clearly visible: in a centralized situation to supply remote heating or decentralized to supply individual buildings, as well as the diversity of cold sources. As the perimeter grows, the diagram becomes more complex but less precise at the same time.

The Energy Context and the Sankey Diagram

Figure 6.11. Sankey diagram of a gas pump [PAH 93]

73

Figure 6.12. Sankey diagram of the Riehen network for 2013 [FAE 17]. The contribution of the heat pump is circled in blue. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

74 Energy Transition

Figure 6.13. Sankey diagram of the heat market in the canton of Geneva in 2035, prospective study [QUI 17]. The contributions of heat pumps are circled in blue. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

The Energy Context and the Sankey Diagram 75

7 From System to Experimental Concept

To meet its objectives, the process of evaluating an energy system usually involves three phases: – a measurement or data collection phase, mainly quantitative; – a phase of analysis of these data points, in order to understand the processes involved, model the energy system or some of its components, simulate it and possibly carry out a study of its sensitivity to various parameters, operating conditions or configurations; – the evaluation phase itself, resulting from the summary of the various analyses and other considerations specific to the problem at hand. In this chapter, the measurement concept, the data collection phase itself and their links with the other phases of CSF will be discussed in turn. 7.1. The importance and difficulties of a quantitative quality assessment The more consistent the results of the analyses are, the better the final evaluation will be, based on relevant and good-quality measures. One of the first challenges is to extend the experimental physicist’s approach in a field where we are interested in physical processes that obey generally well-known laws, but in a context that is too vast to control all experimental conditions, measure and even define all state variables and model all processes. We are going

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

78

Energy Transition

right into what Bachelard called “approximated knowledge” [BAC 27]. In addition, the experimental conditions are not chosen, and so the unique and non-repeatable aspect of the experiment will be underlined. A second challenge is to include the human aspects that affect the functioning of the system to be measured through practices that are difficult to understand, such as ventilation by opening windows in a large building with several hundred inhabitants, the losses of which are an important part of energy consumption. The development of the measurement concept is a particularly tricky point, and no clear method can be given a priori. It is a balance between “too little” and “too much”, between a vision that is too “micro”, that hides interesting things in a jumble of details, and a vision that is too “macro”, that, by smoothing too much, is missing the most significant details. There is a real risk of getting lost in the avalanche of values that current technology allows. The storing of the experience acquired by the laboratory is the best guarantee for the sustainability of the know-how acquired, which otherwise remains implicit and benefits only a limited circle of actors. The difficulties of the exercise, which will be discussed in detail later, can already be outlined: the contradiction between what is desirable and what is possible (size, grip and electrical connections, inconvenience), representativeness of the points, limits of the system, numbers, costs and significant increase in work with the number of probes. A certain redundancy of measurements is nevertheless a guarantee of quality in case of problems with a sensor, and makes it possible to establish the accuracy of the measurements. 7.2. From the energy system to be evaluated to the measurement concept This physical evaluation part is the quantitative basis of CSF and can be done in a variety of ways: – existing data such as site weather data, energy bills, technical statements related to the management or maintenance of facilities and energy consumption forecasts are the most common and often simple to obtain. Together with the technical data, plans and other technical documents of the various devices and installations, they form an essential basis for any evaluation; – some on-site measurements, known as “spot measurements”, can be taken, on the basis of either spot measurements of condition (temperatures, humidity, pressure) and flow (flow rates, air velocity, sunshine) or quantities recorded by

From System to Experimental Concept

79

meter integrators by the people in charge of the surveillance system (heat, electricity, water volume). These data allow a first contact with the system and are often a source of valuable information; – autonomous recorders have developed considerably in recent years; they allow minimally invasive measurements, over a certain period of time and at a frequency as high as desired. In addition, a very high degree of flexibility is possible. A more detailed understanding of how the system works is then available, but a limitation appears if the number of measurement points is significant due to the non-synchronization of these measurements and the tedious work of grouping the data. In addition, some physical quantities have to be calculated by combining other quantities and it is sometimes difficult to determine them after the fact; – centralized instrumentation is the last possibility; it consists of a series of sensors connected to a central data-logger that allows not only centralized and coordinated reading but also instantaneous pre-treatment, such as the average and statistical fluctuation (turbulence) of measurements over a given interval, or various non-linear operations between quantities such as products or quantities from various data such as absolute humidity, which depends on well-known relationships between temperature and relative humidity. In addition, the values can be automatically repatriated by telephone. It should be noted that, little by little, the decentralized recording measuring devices of the previous point are approaching this solution thanks to wireless connections. Any evaluation will generally use several of these approaches in different ways depending on the more or less normative aspect of the CSF. Expertise1, which is very normative in nature since it must establish responsibilities in the event of a dysfunction, will often have simpler approaches but the choice of which partly meets non-rational criteria; measurement should be as minimally invasive as possible if it creates discomfort and increases tensions; on the contrary, one should not hesitate to deploy the measurement infrastructure if there is a demand for attention from occupants. Audits, the purpose of which is to diagnose a system and propose improvements for an unsatisfactory situation, may use the full range of approaches, depending on the size and complexity of the system and taking into account the time and resources available and the objectives. Finally, complete and long-term CSF on innovative systems generally relies on centralized instrumentation as the basis for data acquisition. It is almost always complemented by other approaches.

1 See Chapter 4 for a first presentation and Chapter 15 for a discussion.

80

Energy Transition

The measurement concept must answer the following questions: – What needs to be measured to meet the objectives of CSF? – How should it be done? – What are the quantities to be measured? Which ones should be deduced from these primary data? – Which probes should be used and where should they be placed? – How often are the data read, constructed and recorded? – How are the data collected, processed, verified and stored? – Who are the stakeholders? – What is the status of this data from the point of view of confidentiality, security and intellectual property, as well as the possibility of reusing them or transferring them to a third party? – What resources are needed (financial, human, time, knowledge)? In general, the answers are multiple, but they must be consistent, which requires a number of choices, including possibly adapting the initial objectives. The example mentioned in Chapter 2 of this book, which concerns a simple solar domestic hot water (DHW) system, illustrates the process of defining the measurement concept.

7.2.1. From objectives to a breakdown into subsystems and components The system in question is shown in Figure 7.1 and explained in Part 1. Such an energy system can be broken down into several elementary subsystems, each of which enables one of the adaptations (quality, time or place). As seen in Chapter 2, this decomposition is fractal in nature and the choice of scale depends essentially on the objectives of the study, the resources available and the knowledge already acquired about these systems (intrinsic functioning of the system components and the system’s functioning in use). As shown in Figure 7.2, five subsystems were selected.

From System to Experimental Concept

Figure 7.1. Simplified diagram of a solar preheating system for DHW

Figure 7.2. The five subsystems considered

81

82

Energy Transition

These are: – the solar subsystem, which extends to the internal heat exchanger of the stock; it therefore includes the circulation pump, the flow and return tubes and the regulation. Sufficient knowledge of these systems makes it possible to do without a detailed study of solar collectors. The measurement made will allow a good estimate of the effects of the piping and the quantity of heat transfer fluid contained in the circuit, and a spot measurement of the pump’s electricity consumption, which corresponds to a small percentage of the energy involved, is sufficient to take this into account. The observed performance will correspond to an energy system that is different from a simple solar collector as an object derived from a technological system. It will therefore not be possible to directly compare the results obtained with those given by the manufacturer because the evaluation carried out includes components other than solar collectors that can significantly influence the energies involved; – the internal heat exchanger; – the stock, which includes a solar zone in its lower part and an area devoted to the auxiliary; – the back up; – the hot water distribution circuit, in which only global demand will be considered. The energy from the pressure of the cold water supply system that allows the fluid to move is also not considered in the system (0.2 at 0.3 kWh per m3 of cold water for a distribution pressure of about 10 bars, which corresponds to a temperature increase of 0.3°C). Distribution losses between storage and withdrawal points will not be studied, nor will those of any recirculation that exists in large buildings to satisfy the time concordance (thermal transit time between storage and a remote withdrawal point). Another choice could be made, for example the detailed study of domestic hot water consumption, which could involve sociologists. This would greatly complicate the measurement system since it would be necessary to measure the consumption of each apartment, or even each withdrawal point. As a result of these choices, the thermal scheme (Figure 7.3) is simplified compared to that given in Chapter 2 (Figure 2.3). There are actually a very large number of ways to break down an energy system, even though it is very simple. This partition is a key step in the overall CSF process because it will strongly determine the potential for analysis and synthesis. While it is always possible to aggregate measurements or to not use them, it is often

From System to Experimental Concept

83

very difficult, if not impossible, to reconstruct unmeasured data if the need arises after the fact. A certain redundancy of measurements is therefore recommended, but while remaining reasonable, the measurement system must be assumed over the long term.

Figure 7.3. Simplified flow diagram corresponding to the chosen decomposition in five subsystems

84

Energy Transition

7.2.2. Developing the measurement system The different steps to move from the decomposition performed to the definition of the measurement system are the choice of the measurement points corresponding to the division of the performed system (type of sensor, location, etc.), the processing of the obtained information, the determination of the necessary resources and the adequacy with the available resources. 7.2.2.1. First step: defining the measuring points This involves defining the type of sensor and its location in order to determine the energy flows and state variables of the system. To do this, the process to be followed is shown in Figure 7.4.

Figure 7.4. Process required to measure energy

Direct measurement of energy is not always possible; it is very often necessary to use a basic formula to access its value from system state variables. Measuring these quantities therefore makes it possible to determine not only the value of the energy but also the state variables on which the operation of the system depends. For example, the heat delivered by a transformer to a fluid such as water can be measured directly by a heat meter; this device integrates the measurement of the fluid flow rate and the two required temperatures, then calculates the instantaneous power or the corresponding energy and finally can transmit this value. If temperatures and flow rates are useful quantities for understanding the system, they should be measured independently (using dedicated probes or the heat meter itself if possible).

From System to Experimental Concept

85

Once the quantities to be measured have been determined, the measuring probes for each of them must be selected and then positioned appropriately in the installation. Indeed, the location of the thermometers determines the limits of the subsystem thus defined. In the example of the DHW solar preheating system, measuring the solar heat supplied to the internal heat exchanger in the warehouse involves positioning the probes as close as possible to the exchanger. Placing the temperature probes at the terminals of the collector field gives access to the energy produced by the solar collectors, but without taking into account the various effects related to the transport of heat to the stock. This is a possible choice but, without additional probes, the solar heat actually transferred to the stock is not measured. The measurement at the terminals of the exchanger is also easier to carry out because it is closer geographically to the other probes. Information on the criteria for choosing the type of sensors will be given below. Some sensors impose conditions as to their location, for example devices measuring the flow rates of a fluid must be placed at a certain distance upstream and downstream without hydraulic disturbances, and some of them work better in a vertical rather than horizontal position, etc. The measurement system thus defined must be implemented on the physical system in such a way that it disrupts the energy system as little as possible. Temperature probes will thus produce heat exchanges with the environment and potentially disturb the system locally; flow measurement will cause a decrease in flow rate or an increase in pressure drops and therefore an increase in the electrical power of the circulator, etc. Installed in their place, the various sensors must be checked, read and recorded at regular intervals (time or frequency of measurement acquisition). On the one hand, the measurement rate must be fast enough not to lose information, and, on the other hand, there is no point in measuring faster than the response time of the various devices. We rarely wish to keep all the punctual measurements for the final storage of the averages or totals of the read values, and we can be satisfied with a less sustained rate (time or frequency of storage of the quantities), for example hourly. This measurement aggregation process will apply to the secondary quantities resulting from the point measurements. In the example, the probes and their locations are shown in the schematic diagram (Figure 7.5) and Table 7.1.

86

Energy Transition

Figure 7.5. Position of the measuring probes

Acronym

Type

Unit

Position

Text

Thermometer

°C

Roof, near sensors

Tsolout

Thermometer

°C

Stock exchanger inlet

Tsolin

Thermometer

°C

Exit from stock exchanger

Tstksol

Thermometer

°C

Internal stock, exchanger medium

Tefs

Thermometer

°C

Cold water inlet

Tdhw

Thermometer

°C

Stock removal from storage

Tloc

Thermometer

°C

In a boiler room

Dsol

Volumetric flowmeter

L

In a solar circuit, cold side

Ddhw

Volumetric flowmeter

L

In a DHW distribution circuit, cold side

Wapp

Power meter

W

In an electrical panel

Gsol

Solarimeter

W/m2

On the roof, in the collector plane

Table 7.1. List of probes

From System to Experimental Concept

87

From these quantities, secondary quantities are calculated at each measuring step (Table 7.2). Acronym secondary quantities

Type

Unit

Calculation method

Qsol

Heat

kWh/h

Dsol*(Tsolout-Tsolin)*Cpfluid

Qecs

Heat

kWh/h

Decs*(Tecs-Tefs)*Cpwater

Qapp

Heat

kWh/h

Wapp*Acquisition time

Table 7.2. Secondary quantities from Table 7.1. Cpfluid: specific heat of fluid (kWh/l.K), Cpwater: specific heat of water, acquisition time: time between two measurements

Last comment on this example: on the surface, there is redundancy for the energies passing through the storage, because the two energy flows entering the stock, as well as the energy leaving it, are all measured. In fact, storage losses and changes in the amount of heat stored will be removed from the incoming flows. If these values are integrated over a short time interval, the losses will be minimal but the variation in stored or destocked energy will be large compared to the energy flows through it; if they are integrated over a long time interval, it will be the opposite. The difference in the balance between incoming and outgoing energy should be attributed to stock losses, internal energy variation and measurement error. Only an appropriate analysis can specify its origin; at the measurement stage, it is important to remain cautious. A systematic definition of the acronym given to each variable is welcome; in our example, the first letter corresponds to the type of variable: – Q for a heat; – W for electricity or mechanical work; – T for a temperature; – D for a flow rate. Then, a first group of three to four letters is given to locate the quantity in the system via the subsystem to which it belongs, or the main one if this quantity is straddling two subsystems (in our example, the temperatures at the terminals of the exchanger will be classified in the solar system rather than in the exchanger system): “sol” for the solar subsystem and “aux” for the backup. Eventually, a second group of

88

Energy Transition

letters can be used to differentiate between the same two quantities belonging to the same subsystem for different reasons: – in the measurement of the two temperatures necessary to access the heat received or given by a thermal subsystem, the terms “in” and “out” have been used to distinguish between the inlet and outlet temperatures at its terminals. If these temperatures are associated with the accompanying system, the terms are reversed since the input of one system will be the output of the other and vice versa; this banality is often confusing and explanation in a table would be highly recommended; – the letters can be used to specify the exact function of an apparatus to which the measured quantity is linked if the subsystem contains several of them, for example to differentiate the measured electrical consumption of several pumps in the same circuit; – they can be used to specify the exact location of the magnitude, for example the lower part of the stock devoted to solar energy (Tstksol). The unit must be well chosen and specified as a reminder. Care must be taken to express the quantities in readable and meaningful forms: the basic unit must be adapted to enable easy reading and immediate perception of the order of magnitude (avoid too-small unit values that result in high numbers whose order of magnitude is difficult to distinguish, and symmetrically with too large unit values that induce many zeros after the decimal point). Thus, it is more sensible to store solar energy in the form of fluxes (W/m2), a meteorological quantity always between 0 and 1200, than in the form of system input energy, a quantity that will involve the surface, which will result in values expressed in units that are difficult to understand such as KJ/h or MWh/15minutes! The transition from flow to energy only involves the collection surface, which is a constant of the system, so there is no loss of information to consider it only after the measurement is taken. Unlike quantities such as temperatures or powers, a change in the storage time step of the measurements will modify the recorded values of the accumulated quantities. For example, a change from an hour to a quarter of an hour will divide the transcribed heat values by 4; making it explicit cannot do any harm. It is also possible to choose to normalize by the hour and obtain kWh/h, i.e. an average hourly power, as in the example above. The mass flow measurement (water, for example) can be done in two ways: either the measuring device provides information on the instantaneous flow rate or it alerts each time a certain volume has passed through it (one pulse per basic volume such as 1 liter, 100 liters, etc.). The integrated volume measurement has the advantage of not losing information, unlike the flow measurement, which does not provide information on what is happening between the two readings; however, care

From System to Experimental Concept

89

must be taken to adapt the basic volume so that at each reading, the transit of the fluid results in several pulses. A base volume that is too large can lead to an apparent absence of flow during some measurements and then to a large flow during the next measurement. The same problem may exist in electrical power measurements (reading of an instantaneous power or that of the energy consumed or produced since the last reading). 7.2.2.2. Processing of probe information All these quantities are either averaged (temperature, solar radiation) or integrated (heat, volumetric flow) during the data storage period (in our example, one hour). For secondary quantities, the calculation must be done for each measurement entry2. The specific heat of fluids is generally well known; the problem may be the exact composition of the fluid. In the illustrative solar system, a water–antifreeze mixture circulates, the composition of which can fluctuate according to leaks and fills. Regular measurement of composition by densimetry is recommended. Since it is difficult to instantly change this value in the acquisition program, the correction of the heat concerned can, if deemed necessary, be made after the fact. A delicate point to raise concerns the change of time which takes place every year at the end of October and the end of March. If the measurement system follows this time change, it results in a 23-hour and a 25-hour day and therefore requires special treatment of these two days. Maintaining the same schedule all year round is problematic in relation to human activities, which will be shifted either in winter if we keep the summer schedule or in summer if we keep the winter time. These differences in activity times will result in a shift in the use of the facilities, for example changes in ventilation flow rates according to office hours or meal times. Once the choice has been made, the problems are not over if other sources of recorded data are used, because time synchronization between files from different sources must be ensured. From experience, it is rare that the choice of time system made is clearly explained. In the same order of difficulty, a shift in the clocks governing data recording may occur. This was especially true at the beginning of the computer age, but regular verification of the accuracy of the clock is part of good practice, even though the clock is synchronized by satellite. The status of the data collected and stored in this way is an important legal aspect that should not be overlooked. First of all, even access to the data and information needed for CSF must be ensured from the outset, which may result in an obligation to provide information on the part of some partners. At the end, a large set of data could be created, which can retain a certain value for internal or external 2 Be careful that the product of the averages is not equal to the average of the products.

90

Energy Transition

use after CSF. Clarification of the status of the data is a good practice, preferably to be determined at the beginning of the exercise. This includes intellectual property rights over these data after the end of CSF (who owns them, who can use them, in what context, etc.), the confidentiality of the data thus collected and the possibility of publication. Similarly, the accessibility of the measurement sites and the need for insurance in case of damage must be considered at this stage. 7.2.2.3. The necessary resources The financial resources are of two kinds: investment for the equipment (purchase or amortization and installation) and costs incurred for the monitoring, maintenance and evolution of the measurement system over time. Setting up measurement points during the construction of the system to be studied allows for better integration and reduces costs; on the contrary, it can significantly extend the duration of CSF in the case of certain systems, such as buildings, because it will be necessary to wait until the building is occupied before starting the exercise. With regard to human resources, it is necessary to first ensure the necessary skills to define the concept, negotiate it with the other partners and set up the measurement system. It will then be necessary to organize the follow-up of this system, the repatriation and a first data processing. The work required varies depending on the importance of the system to be evaluated, between a few hours and one working day per week over the duration of the measurements, which can be counted in years. If CSF is carried out as part of a training course, this time may be extended due to the lack of experience of the person in charge – but whose salary is lower than that of an experienced collaborator – or due to the deepening of certain aspects that are not strictly necessary to achieve the objectives of CSF but are useful for the development of general knowledge. In any case, a significant CSF remains a team effort, where the person in charge of the work must be able to rely on a vast potential of skills and experience. In the case of MetaCSF (see section 4.2), the objectives of CSF can only be adapted to the measures available. The benefits of not having to take measurements must be tempered due to the constraints carried over to the next steps (analysis, modeling, etc.). The absence of a measurement concept makes the subject of study impractical and some knowledge is lacking, particularly the human context. On the contrary, this type of CSF is ideal when you have already followed a similar energy system, which you can then complete with new knowledge. An interesting example of this approach is the one performed by Faessler [FAE 17] on geothermal systems. Since there is no geothermal system in operation in the local context, tracking current pioneering installations based on existing data is a powerful accelerator of innovation. The local contexts in which these facilities have developed are given particular attention as learning must be adapted to local conditions: geology, energy and urban planning environment, legal, political and organizational setting, know-how, etc.

From System to Experimental Concept

91

The example used to illustrate the methodology to be applied to define the measurement concept remains simple; for a more extensive innovative system, the work will be more substantial but identical in substance. For example, a detailed description of the measurement concept can be found in [BRA 02, HOL 02b]. To conclude this point, it is good practice to clarify the concept through a document that postpones the planned procedures, which is often necessary in order to find or complete the financing. Changes, corrections or adaptations during the exercise are possible but they must be formulated as they occur, or otherwise there is a risk of losing valuable information. 7.2.3. Some properties of the sensors and their use 7.2.3.1. General properties Only the general properties of the sensors will be mentioned here; please refer to the many specialized books for more information, such as [GUY 99] for measurements on environmental energies or [GOP 90] for thermal sensors. A sensor (thermometer, solarimeter, flowmeter, etc.) is the product of a technological system resulting from the discovery of a measuring principle that is based on a physical–chemical process, an invention derived from this principle and the classical development from the invention as described in Chapter 3. The practice of CSF requires, on the one hand, a good knowledge of the developments of sensors and measuring devices to make relevant choices, and, on the other hand, a good knowledge of how these sensors work in order to make good use of them. In addition to their price and availability, the main characteristic of the sensors is their intrinsic accuracy and their adaptation to the specific use to be made of them. In this choice, the experience of the team and the network around them is essential. There are two main categories of causes of measurement error: measurement errors related to the intrinsic functioning of the instrument and errors induced by its use. It is not a question of making the most accurate measurements possible, because it has a cost in terms of resources, time and attention, but of building a measurement system that produces all the data over time in order to achieve the objectives of CSF. Errors related to the operation of the instruments can be made compatible with the desired precision requirement in two complementary ways. First, granting the intrinsic accuracy of the device given by the manufacturer is potentially necessary; for example, it is not necessary to measure with an accuracy that is two orders of magnitude higher than that required. It is much better to focus on other factors that limit accuracy. Then, a laboratory calibration of the most sensitive sensors may be recommended; either the knowledge of the sensor in question is not sufficiently established and its accuracy must be ensured ex ante, or the sensor itself inherently

92

Energy Transition

needs it, such as a solarimeter whose calibration constant varies with age. Measurement errors related to the use of sensors in an environment different from that of the laboratory are more difficult to understand. No systematic methodology can be given in all generalities. Common sense, use as close as possible to the manufacturer’s recommendations, the resources available and the team’s experience and know-how are the best guarantees of quality in the implementation of the measurement system. During the exercise, only regular monitoring allows the necessary reactivity to measurement drifts and imponderables, such as long-term and out-of-laboratory measurements. Random errors, which cancel each other out when we average a large number of measurements, are not really a problem in the sense that measurement is based on repetition. On the contrary, systematic errors are more embarrassing because they will persist. Some energies depend on differences in state (the heat gained by a fluid, for example, derives from the spatial and temporal variation in its temperature); if these errors vary only slightly from one probe to another and from one moment to another, then the effects of these systematic errors on the associated secondary quantities will be reduced accordingly. An example exists where these systematic errors can be anticipated: the measurement of the temperature of a fluid. 7.2.3.2. Measuring a temperature The basis of thermometry (the art of measuring temperature) is entirely in the expression: “A thermometer only measures its own temperature”. Unlike other measurements, the intrinsic accuracy of temperature measurement is usually limited by this principle. The main difficulty of thermometry lies in the coupling between the medium to be measured and the temperature sensor. The simple measurement of the temperature of the outside air can hardly be more precise than 0.1°C due to the difficulty of defining what and where to measure (effects of sun, IR radiation from the sky or other bodies). The main effects to be taken into account are the inertia of the sensor, which is characterized by a delay time and a response time; the thermal coupling between the medium to be measured, the thermometer and the environment; and the disturbance of the medium to be measured by the measurement system. Any change in temperature of the medium to be measured must be transmitted to the sensor as correctly as possible. Transmitting a temperature change necessarily means transmitting heat, so the transmission rate will be reduced – except in very rare cases such as measuring the temperature of a surface by its radiation and measuring this radiation by a photoelectric effect. The electrical filtering of the signal may also cause a delay.

From System to Experimental Concept

93

A more complex case is that of fluid measurements in pipes where a good thermalization of electrical cables is necessary to avoid leakage of heat or cold from outside the pipe. This thermalization consists of thermally coupling the electrical connection wires as much as possible to the medium to be measured and in minimizing the thermal conductivity of these wires (reduced diameter, high specific conductivity of the metal used). The counterpart is the parallel increase in the electrical resistance of the connections. Another difficulty is avoiding the self-heating of some sensors by the Joule effect, if they have to consume electrical energy to operate. This requires a limited power dissipated in the sensor, therefore a limitation of the signal to be read, and the best possible coupling between sensor and medium so that it can absorb the heat produced by the sensor (which must remain low since ultimately this heat is found in the medium to be measured). The literature [GOP 90] provides the necessary guidance for the most well-known cases, and it should be noted that the heat equation is easily solved manually in simple cases and is well suited for numerical simulation in more complex cases. It also happens that the measuring system changes the temperature to be measured, usually locally near the sensor. A classic example is given by measuring a contact temperature, where the sensor applied to the surface to be measured changes the local conditions. Measuring the temperature of a glazing in real conditions is only very difficult to achieve. The use of a contact probe will inevitably modify the superficial glass–air exchange, on the one hand, and solar absorption, on the other hand, which will result in a disturbed local surface temperature. Again, before performing any feats in determining the local temperature around the sensor, it is important to ensure that it is representative of the entire environment. 7.2.4. Some remarks on the measurement of primary energies Often, primary energy is involved in the energy system to be evaluated, in chemical form (biomass, energy carriers from oil, natural gas), radiative form (sun), mechanical form (wind) or thermal form via a temperature (geothermal, hydrothermal, air). The measurement of these energies is often more difficult than it seems at first sight. The easily accessible quantity for estimating the chemical energy contained in a gaseous fuel is its mass or volume, which is then multiplied by its higher heating value (HHV). There is also another heating value, called the lower heating value (LHV) because it does not take into account the energy contained in the water vapor resulting from combustion, which is recoverable if it is condensed. With the development of condensing boilers, the LHV does not measure all the energy resulting from the combustion process. For the most commonly used fuels, the HHV values are well known and tabulated. While the measurement of the volume of liquid

94

Energy Transition

fuel (oil in general) is not a problem, the situation is different for gas. The energy content of the natural gas delivered varies around an average value due to changes in composition, temperature and pressure. The average natural gas composition during the Solar City CSF [BRA 02] as given by the local distributor was: – methane, CH4, 94.610% vol; – ethane, C2H6, 3.185% vol; – propane, C3H8, 0.493% vol; – butane, C4H10, 0.049% vol; – heavy hydrocarbons, CnHm, 0.008% vol; – nitrogen, N2, 1.586% vol; – carbon dioxide, CO2, 0.003% vol. Under normal conditions (0°C, 1013.25 mbar) and in a dry state, the corresponding lower heat value (LHV) is 36.43 MJ/m³ and higher heat value (HHV) is 40.47 MJ/m³. If we take into account the average conditions of use (15°C, 20 mbar of overpressure compared to the atmospheric pressure of 970 mbar observed on average in Geneva, i.e. 990 mbar of distribution pressure), the density decreases according to the law of perfect gases and we arrive at the usual heat values: LHV 33.8 MJ/m³ and HHV 37.5 MJ/m³. Changes in composition are small. In general, atmospheric pressure is higher in winter than in summer, but daily fluctuations are greater than this seasonal trend. Thus, the minimum pressure was recorded in March 1998 (946 mbar) and the maximum in February (987 mbar). These variations will correspond to daily fluctuations in the gas density, and therefore in its heat value, of about ± 2%. The gas temperature will change at a slower rate. It is expected to roughly monitor the water temperature of the network, which varied very regularly in 1998 between 8°C in February and 18°C in August. This temperature trend introduces a ± 1.7% variation in heat value via density, with the maximum in winter and the minimum in summer. Taking into account the three parameters – composition, pressure and temperature – we can therefore expect fluctuations in heat value from one day to the next of around a few percent, without being able to know them precisely. For solid fuels, such as wood or biomass, the determination of energy content can be made difficult by the presence of large quantities of water, which not only does not burn, but also increases the gap between the HHV and the LHV since energy is required to evaporate it. It is therefore necessary to either have access to the mass of dry fuel material and apply the corresponding HHV, or have access to the LHVs of the fuel used based on its actual humidity, and consider the mass of the

From System to Experimental Concept

95

corresponding fuel. For example, Figure 7.6 shows the HHV and the LHV expressed per kg of wood delivered as a function of moisture.

Figure 7.6. Upper and lower heat values of wood according to its moisture content

Two effects are visible when the humidity level of the wood is increased. The most important is the decrease in the two heat values, due to the quantity of water which increases according to the moisture content of the wood delivered and therefore the consequent lower quantity of anhydrous (perfectly dry) wood per unit of mass delivered. The gap between the HHV and the LHV widens slightly because the water in the wood delivered must be evaporated and, if it is not re-condensed as assumed in the LHV calculation, this heat is lost. Thus, the evaluation of the chemical energy contained in wood requires a continuous evaluation of humidity (see, for example, [ARO 13, MER 15]). For inhomogeneous biomass or waste, in addition to the measurement of mass and moisture content, it is also necessary to determine the nature of this biomass and, if the heat value is poorly known, a measurement of this parameter must be carried out by calorimetry either by the team in charge or in a specialized laboratory.

96

Energy Transition

7.3. Link to other phases of the evaluation Great attention must be paid to the balance between the phases of measurement, exchange, analysis, synthesis and drafting, whether in terms of time spent or the skills or human and financial resources available. The difficulty of the full exercise cannot be repeated enough: on the one hand, acquiring a very large amount of data over a long period of time and, on the other hand, not having enough time to fully use them is just as frustrating as missing information when analyzing the measures, when it would have been easy to acquire them either by measuring or by exchanging them with the appropriate actors. The importance of taking into account the CSF context will be stressed again for several reasons: – in a very practical way, the study is not carried out in a laboratory, and this has various consequences that must be well understood: the aspects of security, accessibility, confidentiality or even susceptibility must be taken into account carefully. In addition, the energy system should be disturbed as little as possible, especially if the behavior of the actors has an influence on the functioning or efficiency of the system in question; – the lessons that can be learned from the measures will be all the more relevant and subject to generalization if the general context in which the system is to be evaluated is well understood and described; – this contextualization makes it possible, through the repetition of experiences, not only to accumulate specific knowledge but also to bring out broader knowledge; – with this in mind, the participation of actors related to the system in supporting the evaluation, including during the first phase of defining the measurement protocol, is a very positive point. The difference between a very standardized monitoring such as the IMPVP protocol of EVO [EVO 10] and a CSF must be highlighted: in the step of defining the measurement concept and even though the objective of CSF is defined in legal terms in the contract that links funders and the group performing the CSF, the measurement remains as the responsibility of the group performing it, whereas a protocol of the IMPVP type fixes this concept as well as the analysis step. The best way to ensure the relevance and quality of measurements (long-term and off-label) is regular monitoring, which alone allows the necessary reactivity to unforeseen circumstances. Beyond this direct monitoring of the measurement

From System to Experimental Concept

97

system, the following steps of CSF must be addressed at the beginning of the measurement. Actions such as graphically representing the data, observing them regularly to better understand the energy system and relating the different measured quantities, make it possible to check whether the measurement system is working and complete; other operations affect quality, such as carrying out analyses of the first calculable yields or developing preliminary indicators. If you wait several months before working on the measurements, the possibility of correcting or improving the measurement system can be greatly affected.

8 Data Observation and Global Indicators

After a certain time of measurement of an energy system, a large amount of data is available. They quantitatively reflect the functioning of the system in real conditions according to the point of view given by the measurement concept. The analysis of the operation and use of the system will be based on this data through a series of methods that will be discussed later. However, before making the brain function in all its analytical rationality, it is good to observe the system through the regular pulsation of the measured parameters: temperatures, material and energy flows evolve over time. This observation phase prepares the analysis phase, guides it and can inspire it. 8.1. Observing and feeling A series of curves representing the state or flow parameters as a function of time allow you to immerse yourself in the system and feel it. Initially, their regular observation makes it possible to quickly detect measurement errors such as outdoor climate values that do not correspond to the real climate, temperature values that are inconsistent with each other and inconsistent energies. Being an essential part at the beginning of the measurement step, this monitoring process remains necessary until the end of the data collection phase. It allows a first examination of the functioning of the system, a first approach to significant details through the research and observation of atypical conditions. For example, the endless repetition of photovoltaic electricity production modeled on the measurement of sunlight (Figure 8.1) quickly makes it clear that modeling can be mainly carried out with physical parameters, that a simplified model will give a good approximation and that it is possible to take the physical modeling quite far because repetitiveness means that the experimental conditions are close to those encountered in vitro (in the laboratory).

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

100

Energy Transition

Figure 8.1. Production of a photovoltaic system observed in Geneva, SIG installation, [SCH 95]

On the contrary, the observation of indoor office temperatures in a large administrative building in summer shows a chaotic aspect (Figure 8.2), which will require more caution in the analysis. Employees working in this uncooled building suffered from the summer heat. Very different temperatures were observed at the same time for spaces that were identical but used differently. On the other hand, in the same space used by the same group of people, the results were more consistent. In this type of case, the actors in charge of CSF can already sense, from the beginning of the measurement period, that the analysis phase will need to become considerably more complex in order to take these heterogeneities into account.

Data Observation and Global Indicators

101

Figure 8.2. Temperatures of nine offices on the west side of an administrative building without air conditioning. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

This example will be detailed in Chapter 12. The different use of opening windows to ventilate and cool the building largely explains these disparities. In practice, this observation and feeling phase can be carried out throughout the CSF, from the repatriation of measurements, through an automated graphical representation, to the description of the system’s operation in the final report through a dedicated paragraph. 8.2. Energy indicators The function of energy indicators is to transmit synthetic information about a system and is an important part of the results that CSF produces. They make visible elements that are not directly perceptible but are nevertheless necessary for describing how a system works and for evaluating it. They are not “given”, they must be constructed from the data from the measurements performed, which were partially designed for this purpose (see Chapter 7). The definition of an indicator is closely linked to the objective assigned. The interpretation of an indicator is more subjective than is often thought. The example of transport is very instructive. A common indicator to measure the energy efficiency of passenger transport is the energy consumption consumed by an

102

Energy Transition

individual to travel a certain distance. It is expressed in kWh/km.person. For example, if a person drives alone in a car that absorbs 6 liters of gasoline per 100 km, their specific consumption will be 0.6 kWh/km.person; if the car is occupied by three people, then each of their specific consumptions will be 0.2 kWh/km.person. This magnitude can vary widely depending on the mode of transportation, occupancy rate and vehicle type, driving style and trip characteristics. There are many websites on the Web comparing different modes of transport, in particular the car/aircraft comparison. Depending on the site, one or the other of these modes is more energy efficient; it is difficult to separate them. In other words, to get from one point to another, it is difficult to know which mode of transport between the two will consume the least energy. However, we can take the road or the plane for another reason than to go to a specific place: to escape, to get away from the place where one usually lives; this has become the case for a large part of the passengers on the planes. In this case, the key parameter is no longer the distance traveled but the time budget, associated with the financial budget. To escape on a long weekend, you will be willing to spend a few hours to go elsewhere; for a few weeks of rest, this acceptable travel time will be longer, for example one day. Expressed per hour of travel and no longer per kilometer traveled and because of its 10 times higher cruising speed, the energy consumption of the plane will be much higher than that of the car, even taking into account the various waiting times. This example shows the specificity of an indicator and the caution required for its use. The purpose of the indicator can be to characterize an existing object for itself; its value will therefore have an absolute meaning. It can also be used to compare two systems or two situations of the same object: before and after an event such as an intervention on the object or a change in operating conditions or use. In this case, the two absolute indices can be compared; it is also possible to monitor the evolution of a relative index, for example the relationship between a system variable and a reference value related to the system itself or to a standard. To define an indicator, there is always a tension between two desired qualities, which are robustness on the one hand and accessibility on the other. People with a good knowledge of energy systems will tend to look for the definition of the indicator that is as close as possible to reality, refine its elaboration and complete the data necessary for its construction. However, the ambition is to allow a wider circle of actors to understand than just energy experts. The tool is therefore used to characterize a complex system in a real situation and is intended for actors who are not always aware of the energy issue: the best can be the enemy of the good. In other words, a robust and accurate but not very explicit indicator will not achieve its appropriation objective, now will a more approximate but more suggestive indicator.

Data Observation and Global Indicators

103

In general, the use of standardized or commonly used indicators in CSF allows for a better dissemination of knowledge. However, this does not preclude critical reflection on them, nor the proposal of alternative indicators. An example of a standardized indicator is the family of energy indices for buildings that have been built over time with a slow evolution of the calculation and estimation process. This widely used family of indicators makes it possible to link the services provided (comfort temperature in winter inside a building, domestic hot water supply, specific uses of electricity) to the energy spent to provide these services. As far as possible, it should also include other important parameters such as climate or the quality of the energy entering the system. In Switzerland, it was defined by the Swiss Society of Engineers and Architects (SIA) through numerous recommendations and standards as early as 1982; it is now the basis for energy legislation in many countries. It is defined as the ratio between the annual energy required for a given service and an energy reference surface (ERS), itself carefully specified by a specific standard (and worth a little more than the net floor area). It is expressed in MJ/m2.year. The nature of the energy considered is varied: primary energy entering the system, energy used for the service which does not take into account the conversion efficiency; the same applies to the service: space heating at different temperatures, hot water for sanitary facilities, electricity for technical installations, lighting for common areas, occupants, etc. This indicator is thus broken down into many more or less aggregated sub-indices. The success of this index is explained by its qualities: simple, easily calculable from the preliminary design stage in a standard situation of use and easily accessible1 in real conditions. The difference between the real index, the standard index and the expected index is one of the important points of CSF carried out on buildings and its explanation often occupies a prominent place in the results. We will come back to this aspect in detail in Chapter 14, because even a standardized indicator remains difficult to use. Behind its simplicity, the thermal index hides many difficulties in its use. If the heating performance is very well defined by the energy reference surface, the same cannot be said for the overall hot water needs, which will depend directly on the number of occupants. The tension between robustness and universality is present here: the robustness of the index would require expressing the energy index for hot water by an average amount of heat per person per year. Since most of the energy used for heating was consumed at the time of publication of this indicator, accessibility prevailed and the hot water index was expressed per unit area, notwithstanding the occupancy rate of the dwellings. With this definition, the heat indices of social housing, including heating and DHW, are higher than those of non-subsidized but less densely populated housing, especially in new buildings 1 For Geneva, there is an important database of thermal indices, as they must be declared annually to the administration for each building.

104

Energy Transition

where the proportion devoted to heating domestic hot water becomes more significant than that allocated to space heating. This has nothing to do with statements that poor people are more wasteful because they are less educated, statements that can be heard here and there to explain this state of affairs. This is mainly due to the inadequate definition of the indicator, following the radical change represented by the extensive insulation of buildings. Concerning non-standardized or poorly standardized indicators, CSF is a very good opportunity to reinforce the knowledge specific to their development. A good illustration is the case of hotel energy efficiency indices which, with the development of eco-labels, have become an important point for the advantage that a green image or, more prosaically, the possibilities offered to avoid the CO2 tax can provide. In its manual Gestion de l’énergie dans l’hôtellerie, the association Hotellerie Suisse [HOT 10] “is aimed at all professionals in their sector who wish to reduce energy costs, plan new installations and are concerned about energy issues. The objective of this common thread is to make them aware of the existing savings potential and to provide them with updated work tools in order to carry out savings measures”. Thanks to a large amount of feedback over a period of 20 years and regular updates, it has been possible to define three energy efficiency indicators for an establishment. The first is the ratio of the annual cost of energy to turnover, expressed as a percentage. Very quickly calculated, it will have a very great significance for the managers and financiers of the institution: “Statistical evaluations show that the average Swiss hotel has an energy cost share of nearly 2.5% of turnover. If this percentage is significantly lower, we are dealing with reasonable energy management. If this share is higher, there are two possible scenarios: in the first case the building and facilities are obsolete (renovation of the building and facilities to be planned); in the second case the energy management leaves something to be desired. Thus, thanks to a detailed analysis, we can see that in establishments with or without a small restaurant, the share of energy costs in turnover is higher than that of hotels with a large restaurant or restaurants without accommodation”. The second indicator is the total energy consumed, weighted by its quality and related to turnover. The upper limit value indicative of poor energy management is 130 kWh/1,000 CHF.

Data Observation and Global Indicators

105

The third is based on the same weighted total energy in relation to a service performance index: in large establishments, particularly in urban hotels, the number of hot meals, visits to the indoor swimming pool, etc. are often recorded in addition to overnight stays. In this case, it is relevant and desirable to make a more precise calculation and to use the performance index defined as a weighted sum of services (overnight stays count as one, meals and pool entries as one-third, overnight stays and staff meals as one-sixth). A guideline value greater than 18 kWh/service unit indicates poor energy management for 4- or 5-star establishments (against a value greater than 14 for others). This indicator is directly intended for people in charge of energy and remains very understandable for the person in charge of the establishment. The energy index relative to the surface (see above) is of little use in assessing energy management, as the services provided are too important a factor to be neglected. The group of three indicators thus defined is at the same time robust, credible and well adapted to the different actors that revolve around this sector. Such an approach is rather rare; it would not be appropriate to carry out CSF or even a simple energy efficiency evaluation for a hotel without using these indicators, testing them and, if necessary, proposing improvements. This example shows the great variety of possibilities and the interest of a long-term approach, based on CSF to quickly and efficiently guide energy saving actions. A practical evaluation of an index is always made in a particular situation; if we want to compare it with others, it is desirable to correct certain deviations such as the specific climate, a use fluctuating from year to year, etc. In order to overcome these particularities, the values of the indicators are often reduced to reference conditions. This is called “standardization”. This is a difficult operation that requires a good knowledge of how the energy system works under different conditions of use and the ability to recalculate the index in the situation considered normal, which must also be specified. For standardized indicators, the standardization method is defined but often approximate in the sense that, if the system were actually measured under reference conditions, the value of the indicator would certainly be quite different from its standardized value. The compromise between robustness and simplicity already mentioned is also being implemented in the standardization process. CSFs, thanks to the extensive data and information collected, are opportunities to test, develop or improve standardization methods.

9 Input/Output and Signature Relationships: the Operation in Use

An input/output (I/O) relationship reflects the functioning in use of an energy component, subsystem or system. It describes the reaction of a given system in use through a relationship between a significant input and an output parameter. It is not a question of revealing a simple empirical relationship between two quantities linked to the same system, but of giving meaning to a set of measures. This is a key step in the analysis of energy systems, paving the way for its modeling. To reveal the relationship, a large number of points representing different conditions of the same system (or subsystem, or component) are chosen according to criteria that must be explained; these will form the basis of the scope of the validity of the relationship: we are looking for the regular rather than the singular. Very often, hourly or daily values are taken into account. An I/O relationship can perform several functions: the convenient visualization of an expected relationship, the search for a simple relationship between quantities coupled in a complex way, the base of simple energy monitoring tools, the foundation of standards. It is a tool of choice for the analysis of energy systems that methodologically lies between the use of indicators, which only give a value that characterizes the system in use, and the development of models, which aim to describe how the system works for different uses. I/O relationships can be extended to relationships where at least one of the parameters is not of an energy nature; in this case, we speak of “signatures”. As seen in Chapter 2, and in Figure 2.6, in addition to the energies entering and leaving the system, it is subjected to external stresses that condition its state, characterized

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

108

Energy Transition

by so-called state variables (Figure 9.1). For example, the input chosen to define a signature can be the outside temperature considered as the main external constraint on the building for heating a building. Similarly, the chosen output can be the summer indoor temperature of an uncooled building, considered as the main state variable defining the comfort of the occupants.

Figure 9.1. Schematic representation of an energy system

9.1. Convenient visualization of an expected relationship Starting from this simple diagram, it is possible to directly relate an output energy to an input energy (Figure 9.2). It is obviously necessary to choose energy quantities that make sense. For example, considering the sum of all incoming energies on the one side and the sum of all outgoing energies on the other side is only relevant as a test of compliance with the principle of energy conservation to estimate the accuracy of a series of measurements or the consistency of various components of an energy balance. The I/O relationship is very close to the concept of efficiency as described in Part 1, but not fully in line with it. The operation of a gas boiler will serve as an example. This condensing boiler was installed in the solar city of Plan-les-Ouates, which has been the subject of extensive feedback over a period of 5 years [BRA 02]. The energies considered are, in output, the heat produced by the boiler for one hour and, in input, the chemical energy of the gas, the latter being obtained on the basis of its average gross calorific value (37.5 MJ/m³, see Chapter 7). Only the hourly periods between June 1997 and June 1999 during which the circulator activating the fluid in the boiler operated for 60 minutes were retained. This represents 3,066 time points. Excluding points where

Input/Output and Signature Relationships: the Operation in Use

109

the boiler did not operate on a continuous basis, special operations with marked transient effects (boiler start-up or shutdown procedures, etc.) that may also disrupt measurements (sensor inertia) were avoided. The resulting point cloud shows a clear linear relationship, with fluctuations on both sides in a fairly symmetrical way (Figure 9.3, top curve). This linearity of the relationship is not always observed; a significant variation in efficiency with the variable power of the boiler would curve the relationship in one direction or the other.

Figure 9.2. Schematic representation of the input/output relationship

This curve shows that the boiler power varies between a few tens of kW and 225 kW. This variation is due to the possibility of modulating the burner power and, possibly, the actual burner running time – the fact that the circulator operated all the time does not imply that the same is true for the burner, information not available at the time of analysis. The linear regression line forced by the origin indicates a constant assumed hourly efficiency, regardless of the boiler output of 89.6%. Over this two year period, the average efficiency obtained by the ratio between the total heat produced by the boiler during these two years and the total HHV chemical energy of the gas is also 89.6%, indicating that the points excluded from the analysis have a close (weighted) average efficiency. The average efficiency achieved during the 1998–1999 season (91.1%) is higher than the previous year (88.3%), reflecting better use of the boiler – lower working temperature – rather than an improvement in its intrinsic operation.

110

Energy Transition

Figure 9.3. Input/output relationship of the gas boiler. Time points, June 1997–May 1999. Below, the variables “x” and “y” have been inverted

Input/Output and Signature Relationships: the Operation in Use

111

Efficiency fluctuations around an average value can have several physical causes, in addition to fluctuations in gas energy (composition, pressure, volume) around the average value considered and measurement errors. To do this, it is necessary to “open the box” and look at the operation of the “condensing gas boiler” component. The treatment of these points is beyond the scope of input/output relationships; it will be covered in Chapter 10 on modeling. NOTE.– The use of statistics. A widespread habit has emerged with the use of spreadsheets; their graphical utilities make it possible to visualize a relationship between quantities and to unearth linear regressions with a single click. It should be recalled that when estimating parameter A of a linear regression represented by the relationship y = A* x – as done above, the choice of variable x should be based on the best determined quantity1: fluctuations (random measurement errors, minimal variability of operating conditions perceptible, however, by their effect) are assumed to come only from the quantity y. This means that the two quantities considered in the I/O relationship are no longer of the same nature. The choice depends on which impacts the result. Thus, in our example, if we invert the two quantities, the relationship obviously remains unchanged (see Figure 9.3, bottom curve), but the linear regression forced by zero gives a value of 1.114 whose inverse is 0.898, which is different from the value previously obtained of 0.896. The gap may seem small, but it becomes significantly larger for a less well-established relationship with fewer points2. The equation of the I/O relationship thus calculated is therefore not a simple algebraic relationship that can be manipulated as desired, but the translation of an approximate knowledge, a statistical estimation of parameters based on a number of assumptions. 9.2. Search for a global relationship Sometimes the functioning of a complex system can be described in a simple way by an I/O-type relationship. This allows an easy representation of the functioning of this system, but it also shows that a modeling of the system under observation conditions must also lead to such a simple relationship; a more than exciting challenge for scientists! This is the case with solar thermal systems (see Chapter 7): if we trace the daily production of such a system according to the amount of sunlight irrigating the collectors, we generally obtain a straight line for very many types of solar collectors or recovery systems.

1 Obviously extensible to polynomial regression. 2 The same applies to a relationship with a constant term.

112

Energy Transition

The first I/O-type relationships on a daily basis were developed as part of the International Energy Agency’s (IEA) work in the 1980s on the potential of vacuum solar collectors developed simultaneously by several large companies (Philips, Corning, Sanyo, etc.) following the two oil crises of 1973 and 1978. This work was based on several CSFs of thermal solar systems with evacuated collectors, with a size of up to 1,000m2 and for industrial use or district heating supply. In these systems, the temperature to be delivered is high (> 80°C) and not very variable. Figure 9.4 shows the I/O relationship of Corning evacuated collectors coupled with the return portion of a district heating loop, located in the suburbs of Geneva.

Figure 9.4. I/O relationship of a solar thermal system with evacuated collectors. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

Input/Output and Signature Relationships: the Operation in Use

113

A linear relationship between daily solar production and the solar energy incident on the collectors exists, with a sunshine threshold below which no production is observed. It can be seen that, when the supply temperature of solar collectors increases, the threshold sunshine increases while the slope of the regression – which represents a net capture efficiency – decreases. This relationship is not surprising in itself, but its regularity is surprising in view of the many effects that can modify productivity, such as the daily distribution of sunshine, the variation in day length, the distribution between the direct and diffuse components of solar radiation, the average height of the sun, wind speed, etc. This relationship was also confirmed later, during work on solar domestic hot water (DHW) preheating systems installed on apartment buildings (see Figure 9.5 for the example of all apple trees in Geneva [ZGR 10]).

Figure 9.5. I/O ratio of the solar preheating system [ZGR 10]

The linear relationship remains true, but we can see that the threshold sunshine is very low: this system has provided heat for more than 350 days in the year. The average capture efficiency is well represented by the slope of the line and is about 50%. In these DHW preheating systems, the working temperature of the solar collectors increases with the productivity of the system due to the existence of a dedicated storage, whose temperature increases with solar production. Thus, unlike the previous case where the working temperature was determined by the use of the heat produced, it is here the output energy that determines it. This more complex situation is reflected in a contradictory way in a simpler relationship.

114

Energy Transition

These I/O relationships allow monitoring of the installations. For example, in Figure 9.5, two situations are shown where the operation of solar collectors was a problem. In the event of remaining snow on the solar collectors, the collection is nil or reduced and the output is below what is expected. In case of strong sunshine and low hot water extraction, the irrigation imbalance between the different collector rows can cause the least irrigated rows to exceed the boiling temperature and thus block the flow. They stagnate because the captured heat is no longer extracted; their temperature rises to a value where the heat losses balance the captured energy. These points are indicated in red and are located slightly below what is given by the I/O relationship. From this empirical relationship, a universal I/O curve model could be developed on a physical basis; all systems with thermal solar collectors can be represented by a single so-called “universal” I/O curve [GUI 90a, GUI 90b]. 9.3. Signatures as simple management tools This is an extension of the notion of I/O relationship insofar as at least one of the two quantities connected is not of an energy nature. In the case of a building’s energy signature, the Text outdoor temperature is considered to be the main constraint that determines the heating energy required to maintain the building at the indoor temperature Tint.

Figure 9.6. Schematic diagram of the energy signature of a building

There are obviously many other constraints that affect the amount of energy required to heat a building: sunshine, wind, number of people, electricity consumption, window management, indoor temperature, etc. However, the energy signature characterizes a building in a given situation: the condition of the building

Input/Output and Signature Relationships: the Operation in Use

115

and the technical system providing the heating, energy management by the professionals in charge and use of its occupants. Any change in the status, management or use will result in a change in this signature. It is therefore a very valuable and robust tool for building monitoring. For example, the energy signature of the Le Pommier complex [ZGR 10] is shown in Figure 9.7 in daily values for the 4 years following the commissioning of the buildings. Heating consumption refers to gas (HHV); it is expressed as an average power in relation to the SRE energy surface area of the assembly (W/m2). This unit allows comparison with other buildings, regardless of their size. The evolution of heating management is clearly visible. In the year of commissioning (2005), gas consumption was higher than in subsequent years for the same outdoor temperature. An improvement was still visible in 2006–2007, and then the consummation stabilizes. The functioning of this unit is thus very well described by its energy signature and its evolution.

Figure 9.7. Energy signature over 4 years of a building, daily values [ZGR 10]. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

9.4. The signature as the basis for adjustment The system has changed from the previous case because it is limited to the heating system of a building. The heating distribution temperature, which represents the main state variable of the heating system, is linked to the outside temperature as the main constraint (Figure 9.8). This so-called “heating curve” relationship is used in most building heating controllers. The distribution temperature also has an

116

Energy Transition

important influence on the possibility of developing renewable energies, directly or through heat pumps [DES 17, QUI 17].

Figure 9.8. Heating curve

9.5. The signature as the basis for a standard Finally, the signature is also used to define standards, for example for permissible summer comfort in buildings without air conditioning. The indoor temperatures during office hours are then linked to the maximum of the outdoor temperature of the corresponding day (Figures 9.9 and 9.10). An example of how to use this standard is provided in Chapter 12.

Figure 9.9. Summer comfort standard based on signature

Input/Output and Signature Relationships: the Operation in Use

Figure 9.10. Summer comfort standard for offices without air conditioning, SIA 382/3 standard [SIA 03]

117

10 Modeling

Modeling is, according to Le Moigne [LEM 90, p. 5]: “The action of intentional development and construction, by composition of symbols, of models likely to make intelligible a complex perceived phenomenon, and to amplify the reasoning of the actor projecting a deliberate intervention within the phenomenon; reasoning aiming in particular to anticipate the consequences of these possible action projects”. In this very general definition, intention is at the center of the modeling process. More than a simple tool for understanding, it is also a decision-making tool. 10.1. Why model? It is extremely rare that the modeling of a component, subsystem or even the energy system is the main objective of a CSF; we can even conduct a CSF without developing, testing or using a model. Modeling is a key tool for achieving the objectives of a CSF, an articulation between the components or energy system analysis step on which the models are based and the synthesis step that uses these models. Modeling is therefore an important tool in CSF and is fundamental for the team in charge, as each development or use of models is seasoned. The objective of modeling is varied, ranging from a simple description of how a system works in response to various constraints to a detailed description of the processes in progress in that system and their interaction, thus providing a thorough

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

120

Energy Transition

understanding of how it works. In other words, modeling can simply predict how a system will work as a whole or understand it in detail. It is useful both for the development of a component in the related technological system and for the development of the use of that component in the energy system. In a CSF, modeling can have several functions: – checking the proper functioning of a component or subsystem in relation to the manufacturer’s announcement, which is based on a standardized use of the component or subsystem: temperature level, sunlight, operating stability (static conditions), etc. These tests are carried out in the laboratory according to very precise standards. The operating conditions imposed are not always fulfilled in a real system, and it is therefore necessary to “renormalize” the observed operation to the operation provided for by the standard. This renormalization process most often involves a component operating model, which makes it possible to virtually reproduce its operation under standard conditions; – the operation of some components is simply defined by point efficiency values representing a limited number of conditions of use. For example, heat pump COPs are often given in table form with some standardized uses of cold source temperature and hot source temperature. In reality, the conditions vary and are rarely the same as those given in the tables. A more physical model allows an easier representation of the operation in non-standard use and gives a more understandable indication; – it happens that the multiple arrangement of a component in a subsystem changes its intrinsic functioning. Indeed, the subsystem thus composed can in turn disrupt the external constraints that disrupt its operation. We can talk about emerging property. For example, a series of unglazed solar collectors can act as a solar roof and provide multiple functions: protection against rain, thermal and sound insulation, energy capture. It will behave thermally in relation to the wind in a specific way depending on the arrangement. Indeed, the size of the roof, its orientation in relation to the prevailing winds and geometric details (dropouts, obstacles) will interact with the wind and thus modify the aeraulics near the collectors which greatly determines the convection exchanges. A difference in efficiency between the standardized efficiency of an element and that observed in situ during a CSF of the assembly may be due both to the element itself and to the effects of the multiplication arrangement. An adequate model should make it possible to switch from the overall operation of the roof to the operation of a unit solar collector as standardized and guaranteed by the manufacturer. An identical situation of emerging properties exists in a territory that is experiencing a strong development of shallow geothermal energy. At some point, the density of the

Modeling

121

geothermal probe fields will be such that they will influence each other. Currently, the standard model used to size each probe field considers it isolated and therefore does not take into account interactions between neighboring systems. A more complex model will have to emerge that can take this influence into account. CSFs are opportunities for the development and validation of such models; – in the event of a possible difference between the performance of a system resulting from CSF and that which is planned, determine the part that comes from the intrinsic operation of the components and that which comes from the conditions of their use by arranging them in the energy system monitored in the CSF type; – evaluate and improve existing component arrangements; – test arrangements other than the one measured (virtual experiment); – renormalize indicators, as already seen in Chapter 8. 10.2. Analytical and systemic approaches CSFs focus on energy systems that are arrangements of components, the latter being derived from its own technological system that has enabled their design, construction and development. It is therefore to be expected that there will be a difference between the models derived from technological systems to describe the functioning of these components and those used to evaluate the energy systems that are emerging from them by assembly. Systemic modeling is often opposed to analytical modeling: Le Moigne [LEM 90] compares it with Table 10.1, and it would be difficult and damaging to do without the analytical register to understand a complex energy system, just as it would be impossible to do without the systemic register. In practice, the two registers are used in a complementary way. In the technological system, the innovation process turns a discovery into an invention and then into a product that is increasingly used and can improve with the development of its market according to the experience curves (see Chapter 3). The process at the base of the component is generally well-known and well-controlled, as well as the object exploiting the process, which means that its operation can be modeled quite accurately. Sciences such as thermodynamics make it possible to quantitatively describe a whole set of coupled processes. For example, a heat pump can be very accurately modeled using its thermodynamic cycle, the corresponding fluid state variables (pressure, temperature) and the physical properties of these fluids. This will allow us to model with great precision the operation of such a component if we know all the necessary quantities.

122

Energy Transition

The two categories of modeling Concepts familiar to analytical modeling

Can be substituted

Concepts adapted to systematic modelling

Object

Project or process

Element

Active unit

Ensemble

System

Analysis

Design

Disjunction

Conjunction

(or cutting)

(or articulation)

Structure

Organization

Optimization

Balance

Control

Intelligence

Efficiency

Effectiveness

Application

Projection

Evidence

Pertinence

Causal explanation

Comprehension

Table 10.1. Analytical and systemic modeling, according to [LEM 90]. By changing the register, or the style, do we not create the conditions for a change of method?

For his modeling approach, Gicquel1 uses three steps (see [GIC 09, p. 3]): – a qualitative description of the transformer by decomposing it into components that are simple enough to be described mathematically according to the basic laws of thermodynamics; – a purely analytical approach that allows each component to be represented by a number of characteristic parameters, coupling variables and an appropriate set of equations; – a more systemic approach that makes it possible to model, in stationary condition, the whole transformer from the modeling of each part, the outputs of one being the inputs of the next and so on until the complete closure of the cycle.

1 We do not use exactly the same terminology as Gicquel with regard to components, transformers, subsystems and systems.

Modeling

123

This dual analytical–systemic approach can be used in the study of energy systems (ES) undergoing CSF. Indeed, and as already seen in Chapter 2, the system is broken down into subsystems or components, which will then be modeled using either an analytical, systemic or mixed approach. Then, all of these models will be coupled to describe how the ES works. Note that the analytical approach can, in some cases, lead to a numerical solution of the mathematical equations of a model. Similarly, a component can be approached by finite elements small enough to be modeled simply (e.g. linear relationships between parameters will be considered) and then numerically coupled to each other to restore the component. Several illustrations of these different approaches can be found in [HOL 02a]. 10.3. Modeling and approximate knowledge The concept best suited to the modeling of energy systems in use situations is most certainly that of Bachelard’s approximated knowledge: “To know is to describe in order to recover… A double necessity has emerged: you must be complete, but you must remain clear. We must make contact, an ever closer contact with reality, but the mind must be alert, familiar with its perspectives, confident of its points of reference” [BAC 27, p. 9]. “We only understand a mechanism well if we combine its pure and simple description with an examination of the harmony of means and purpose that leads to repeated judgments. A technology develops in the realm of ends” [BAC 27, p. 160]. The modeling step which Bachelard translates as “describing to recover” is very often at the heart of a CSF. However, given the spatial, temporal and intrinsic scope of the system to be modeled, the role of humans in its use and therefore its functioning and its real state is necessarily only known in an approximate way. It is therefore necessary to adapt the modeling approach to this unavoidable fact. The dual necessity of being complete and clear, taking into account the purpose of the system under study, and the inevitably close knowledge of its state can be summed up in this familiar motto: “As simple as possible and as complicated as necessary”. However, it remains to define what is possible and necessary and before that decide who defines it and on what criteria.

124

Energy Transition

Bachelard has admirably put the problem well: “Simplifying is sacrificing. It is the opposite movement of the explanation that does not fear prolixity. When we were measuring, we could neglect smaller order of magnitude terms because the measurement is a quantitative description and a small term cannot react mathematically to a larger order of magnitude term” [BAC 27, p. 101]. “Simplicity is not a state of things but a true state of mind. We do not believe because it is simple, it is simple because we believe” [BAC 27, p. 95]. “It is essentially a question of completely downgrading the detail, of taking away all its value as picturesque, all its strength before the occasion, of lowering it to a level where it cannot have any action on it seems. The detail thus leaves the order of magnitude where one acts as it left the order of knowledge where one measures” [BAC 27, p. 159]. Thus, simplifying means, above all, making the choice to separate from the details considered as insignificant, which must first be determined. It means focusing on the description of the regular, seeing the singular as a deviation that will eventually have to be integrated into the model if necessary. The previous steps of CSF which initially consist of observing and giving meaning to the accumulated measures (see Chapter 8) are key passages that allow this simplification, in particular the definition of the regular behavior of the system, as opposed to singular behaviors. The significant details also emerge from these phases of observation of the behavior of the ES, from the in-depth examination of the pulsations of its state parameters in parallel with those of the stresses acting on the system and those of the parameters internal to the system. It is a question of believing – and therefore of convincing oneself – that, despite the complexity of the system and the large number of parameters in action, the functioning of the system can be described in a simple way. An input/output relationship with an intelligible structure is the best testimony to simplicity and can provide a solid basis for belief. 10.4. Modeling in the context of approximate knowledge of CSF The modeling of an energy system performed as part of a CSF must take into account the specific context of the exercise, which differs significantly from that of modeling a component within its own technological system. The use of the models

Modeling

125

developed in the latter framework must therefore be carefully evaluated in the light of the contradiction between “being complete” and “remaining clear” and the fact that one has only an approximate knowledge of the system and its state. The energy system places its components in real-life situations2 over a long period of time. System management is more or less correctly carried out and efficient, deviations from certain adjustment parameters inevitably occur, the dynamic aspects are important and result from the continuous variation of operating conditions. Each CSF is unique, the human aspects are important and their impacts are incorporated into the functioning of the system itself. On the contrary, the test of a component is carried out under standardized and perfectly controlled conditions; the results must be reproducible and not dependent on experience. The human aspects are reduced to a minimum and must become insignificant details in the sense of Bachelard. Each component is in an arrangement with other components that will influence each other. Their functioning will therefore be in a “suffered” situation. Not all of these situations are desirable, either in cases where they do not achieve one of the objectives of the energy system (e.g. the indoor temperature of a building that is too high or too low) or in cases where they are notoriously inefficient (too many inputs observed compared to what might be expected). The modeling of the system may be different depending on whether we are interested in the regular operation or in its singular functioning. Broadly speaking, for an energy system that has reached a fairly developed stage of maturity, regular operation is the one that will require the most attention, since it is this situation that we want to evaluate and further improve. The singular operation must just be avoidable or correctable. In the case of an immature energy system, it may be difficult to make a clear separation between the regular and the singular. The definition of the details that significantly affect the system and must therefore be integrated into the model is made difficult by the fact that it has not been possible to collect all the necessary information for all the constraints, quantities and variables that could affect the system in one way or another. Choices had to be made at the time of the measurement concept, some of which are irreversible (see Chapter 7). A simple example is the opening of windows in a large building that has a significant impact on its energy consumption for heating: measuring the state of opening each of them is technically possible but unrealistic because of the necessary means (financial costs, work, data processing, etc.). And it is obviously out of the question to block the openings so that the object of study is

2 Although it may appear short compared to the component itself, an observation period of a few years, as practiced in CSF, is rare in in vitro research.

126

Energy Transition

compatible with a simple model! These sudden operating situations, observed imperfectly and therefore known in an approximate way, can be difficult to model. One possible answer is to focus on a limited number of aspects (a component, a particular functioning, a particular variable) and to carry out a series of so-called “spot” measurements, of short duration and which overlap a given time with that of CSF. This allows the information needed to perform the modeling to be collected on a more solid basis. To return to the example of the opening of windows by the occupants of a building, a direct observation is possible by regularly photographing the different façades during periods that are short but still considered representative. Observation and analysis can then be used to describe this phenomenon and enrich the building model (see Chapter 14). The models developed in CSF can be used for system design; the reverse is also true. A notable difference is that in a CSF, there is a detailed – albeit approximate – knowledge of the energy system built and in use. The objectives of modeling are therefore very different. In a CSF, it is a question of understanding how an existing system with an exemplary aspect works in use, and how it can be improved and evaluated. Based on this information, the aim is to disseminate this information and extend the adoption of good practices. In the design phase, the objective of modeling is to optimize the different possible arrangements between components, taking into account the multiple constraints that arise in any project. To make a choice, comparisons between several variants of a project are based primarily on a ranking between the predicted values of certain indicators. Ensuring an absolute value during this planning stage is a difficult but sometimes necessary exercise (see Chapter 12). In any case, knowledge of the models developed in CSF is important for the development of design tools. 10.5. The steps of the modeling and the necessary validation In general, the essential steps in the development of a component model or a system model can be given: – objective and context of the model; – choice between developing a model, adapting or using an existing one; – development of the model bases, mathematical or algorithmic development and translation into software; – determination of certain parameters and validation; – use or status of the model.

Modeling

127

The first step is to clearly define the objective of the model and its framework of use. Depending on whether one is interested in the intrinsic functioning of a component, its integration into a system or the overall functioning of the system, the approach will have to be different. In the first case, the analytical approach will most certainly be preferred, while in the third case, a systemic approach will most likely be chosen. Similarly, the maturity of the component in question or the system will influence the type of model. Finally, the measurement concept and the data collected will also have an influence through the finesse of the possible analysis upstream of the model and the possibilities of validation downstream. There are many advantages to using a model that already exists, that’s already well-developed, regularly used and has been validated on different systems. However, the spontaneous tendency of a modeler is to create “their” own model. Before embarking on such a development, it is always necessary to look at existing models and, if there is one that can handle the component or system in question, test it, possibly upgrade it in order to adapt it. Even though, in the end, its use does not prove judicious in the particular context of CSF, this approach will certainly have positive impacts in the necessary development of a new model. A model is based on a number of assumptions, which define its validity in use. These assumptions restrict the operating conditions of the system; they concern parameters that have not been classified as “details”, according to Bachelard. It is based on system input variables, the external constraints that apply to it and parameters internal to the system; it makes it possible to determine output variables or state variables most often, as well as input variables necessary for its operation. It is most often organized in software, the source of which can be protected or open. Environments such as Excel® or MathLab® allow for easy programming and sharing. A model must always be validated in the context of CSF. For an existing model, already well-validated in other evaluations and which is well-adapted to the problem addressed, validation sometimes involves determining certain parameters of the model whose value is not accessible by analysis but must be based on measurements. The comparison with an expected value can be used as a validation. In the case of an original model, care must be taken to ensure that the data used for the possible determination of parameters and those used for model validation are clearly separated. In any case, the model is a priori only valid under the conditions encountered by the system during the periods used for its validation. This also gives rise to the superiority of an existing “ready-to-wear” model in terms of robustness, even though it is objectively less well-adapted than a “tailor-made” model. The range of models for a given system is often extensive. On the one hand, we can find well-tried models, applicable to standardized systems and very easy to use;

128

Energy Transition

on the other hand, there are very open models where the choice of arrangements and characteristics of the components is very large, but they are more difficult to use. Another selection criterion is the status of the model. A commercial model (“ready-made”) has many advantages, as we have seen, but the knowledge acquired through its use does not advance its functioning, except in the case of an open model developed in a context of mutualization and sharing, such as the TRNSYS Software [KLE 12]. Finally, each CSF is an opportunity to develop and consolidate the experience of the team in charge, particularly through modeling work. Depending on the choices of approaches made, it can promote the appropriation and strengthening of new basic knowledge about processes or systems, mathematical or computer techniques, or practices in the use of existing software. In conclusion, the modeling step in a CSF is difficult to carry out, the experience acquired by the team is important, as is its ability to understand both the processes present in the components and their interactions. 10.6. Some component modeling carried out in CSF The various modeling examples are intended to illustrate the modeling stage. In this section, only the component models will be illustrated in order not to make the subject too cumbersome. Energy system modeling will be described in Chapters 12–14. 10.6.1. Integrating dynamic aspects to check the proper functioning of a component The objective of the modeling performed is to verify the proper functioning of a component according to what has been announced by the manufacturer. The consideration of operation under fluctuating conditions and its comparison with static standard conditions will be more particularly addressed. The example presented concerns the condensing boiler installed in the Solar City of Plan-les-Ouates, which was used as an example of an expected input/output relationship (Chapter 9). In section 9.1, efficiency fluctuations around an average value can have several physical causes, in addition to gas energy fluctuations (composition, pressure, volume) around the average value considered and measurement errors. To do this, it is necessary to “open the box” and look at the operation of the condensing gas boiler component. First, the lower the temperature of the water returning to the boiler, the better the recovery rate of condensation heat

Modeling

129

from the water vapor contained in the flue gases. Then, the energy losses are diffused through the boiler shell to the heating room and to the outside via the combustion gases. Many parameters will influence these losses. The most important thing is, for a well-controlled combustion, the temperature of the circulating water, which will determine the temperature levels in the boiler. Finally, the temperature variation of the boiler over the hour will influence the efficiency measurement since all the energy used to raise this temperature, or recover it if it has decreased, will respectively come from or be added to the energy produced by the gas. However, the effects of the water inlet temperature in the boiler and the variation in the water outlet temperature in the boiler over the hour can be demonstrated by plotting the hourly efficiency as a function of either of these parameters (Figure 10.1). The capacitive effect is clearly visible, as well as the increase in efficiency when the condensation temperature drops. A more precise quantification of these effects has been undertaken, but it has not been possible to highlight the three effects independently because the same parameters are involved; for example, a decrease in inlet temperature will both increase the quantity of condensed water vapor and decrease heat losses. We therefore considered the following equation to represent the boiler efficiency: =

−

∗

+

∗



[10.1]

where: – ETA: the hourly boiler efficiency, measured; – ETAbase: the static efficiency at 0°C, to be determined; – a (%/K): the average efficiency increase parameter with inlet temperature3 decrease, to be determined; – Ceff (kWh/K): the effective heat capacity, based on the variation in the outlet temperature of the water passing through the boiler, to be determined; – Tce (°C): the average hourly inlet temperature of the water passing through the boiler, measured; – Tcs60, Tcs0 (°C): the average boiler outlet temperature over the last 5 minutes of the hour in question and of the previous hour, measured; – Egas (kWh): the chemical energy of the natural gas entering the boiler, measured. 3 Largely related to the additional condensation, as well as taking into account the decrease in the temperature of the resulting fumes and the effect on heat losses between the boiler and the room.

130

Energy Transition

Figure 10.1. Effect on the efficiency of the inlet temperature (low) and the variation in the output temperature during the hour (high)

Modeling

131

A multi-linear adjustment according to the formula (10,1) of the 3,066 hourly efficiency points previously selected determined the three model parameters (Table 10.2):

Value 4

ETAbase -

a K-1

Ceff kWh/K

1.010

0.0028

0.286

±0.002

±0.0001

±0.005

Manufacturer, 40% Pmax

1.03

0.0028

0.284*

Manufacturer, 100% Pmax

1.00

0.0025

0.284*

Uncertainty

Table 10.2. Estimated model parameters [BRA 02] (* corresponds to the 245 liters of water contained in the boiler)

The static efficiency at 0°C ETAbase is greater than 1, but this is an extrapolated value because the return temperature is always greater than 25°C. A more physical value of 0.938 is obtained by considering this last temperature, which is also close to that of the boiler room. With parameter a, we see that we gain a little more than 1% in efficiency when the return temperature to the boiler drops by 4°C. The effective Ceff capacity corresponds to about 250 liters of water. These values are very close to those given by the manufacturer5, which confirms the technical seriousness of the firm in question. Model validation can also be performed by comparing the value of the three model parameters previously determined with the expected values. Static efficiency at 25°C (0.94) corresponds to the optimum combustion efficiency found in the literature [REC 95]. The estimation of the value of parameter a is a little more complicated. For stoichiometric combustion of methane under standard conditions, the flue gases will contain approximately two moles of water in vapor form in a total of 11 moles of flue gas, which corresponds to a partial water vapor pressure of 18,400 Pa (2/11 of 101,300 Pa) and a condensation temperature of 58°C for this water vapor. At 25°C, the saturated pressure of water vapor is 3,200 Pa; the proportion of condensed water vapor will be proportional to the relative decrease in its saturated pressure between 58°C and 25°C, or 77%. Taking into account a maximum possible gain by condensation of 10% – corresponding to the difference between HHV and LHV, the total gain obtained between 58°C and 25°C will be 7.7%, i.e. an average of 0.27% per degree of decrease, surprisingly close to the 0.0028% determined. 4 Defined from classic statistics, see, for example, the LINEST function on the Excel® spreadsheet. 5 For further details, see [BRA 02].

132

Energy Transition

As for the third parameter, such a perfect match is pleasing but always surprising. NOTE.– Integration of the dynamic aspect. In this model, the dynamic effects are taken into account using two hypotheses. On the one hand, the system is represented by a single temperature, the difference between the end and beginning of the period considered characterizing the effect of the inertia of the system. On the other hand, the calorific capacity taken into account to calculate the stored or destocked energy applies to this temperature. The art of the modeler is to choose the representative temperature so that the calorific capacity thus defined is close to the physical reality. This way of dealing with the inertia of a system makes it possible to consider non-consecutive values for the determination of the calorific capacity at the basis of the dynamic effects, with the enormous advantage of being able to select the operating conditions of the system: it is better to eliminate 10 “good points” than to keep one “bad” point – in a particular condition. We are placed in the conditions where the model can work. In a more traditional dynamic model, a long period of regular operation is required for validation. In addition, errors have repercussions from one point to another in the modeling because the final temperature of a time step is the initial temperature of the next time step, which makes validation difficult. An example of how to use such a simplified dynamic model is presented in Chapter 13. 10.6.2. Developing a more explicit but simple model The objective here is to develop a more explicit modeling of a component’s performance indicator than that provided by the manufacturer, who often only gives point values of these indicators. In the case treated, the use of the second law of thermodynamics through Carnot efficiency allows the characterization of a component by a single and very explicit parameter. This point is illustrated from a detailed CSF in [KHO 18], which concerns the use of a heat pump to recover heat from the air extracted from an existing residential building to preheat domestic hot water. The technical characteristics of the heat pump manufacturer have been shown in the graph at the top of Figure 10.2. A quadratic relationship between the COP and the temperature difference between the condenser and evaporator outputs is shown in the graph at the bottom of the same figure. There is simplification, but the two parameters obtained have no direct meaning.

Modeling

133

Figure 10.2. COP as a function of temperature according to the manufacturer’s test. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

At the top of Figure 10.3, it can be seen that the comparison of the COPs measured in situ in the CSF with those given by the manufacturer (via the quadratic model) shows a significant operating deficit (–25%). The bottom of the same figure shows the strong correlation between the COP measurements and the Carnot efficiency calculated with the inlet (cold source) and the outlet (recoverable heat) temperatures6 and reduced by a poor technical efficiency of 0.265.

6 On each side of the heat pump, the extreme temperature, i.e. the outlet temperature of the heat exchangers, was used.

134

Energy Transition

Figure 10.3. Comparison between the measured COPs and between the COPs given by the manufacturer (high) and the COPs of Carnot (low), for different temperature differences between heat produced and cold source. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

The advantage of this approach is that it only retains one parameter: technical efficiency compared to Carnot’s maximum efficiency. This parameter contains a lot of implicit information about the real thermodynamic cycle, the various imperfections and the temperature differences between different components inside the heat pump. The description of the operation of a heat pump is ideally simple

Modeling

135

since only one parameter is required, which also has a direct physical significance and whose value can be expected between 0.3 and 0.6, unlike the manufacturer’s data which has no physical significance. Hence, this approach is widely used in the modeling of energy systems containing heat pumps. On the contrary, it is a simple performance indicator, certainly not a useful model for improving the component within the technological system, in which the use of thermodynamics is inevitable [GIC 09]. CSFs make it possible to specify the operating values of installed devices, in order to monitor their evolution over time, types or manufacturers. 10.7. Simulation of energy systems The modeling of an energy system is done by coupling the models of its components. There are two cases: either this coupling can be carried out without establishing a circular logic (the input variable of one model depends on the output variable of another model which in turn depends on the output variable of the starting model), so the coupling is technically simple to carry out. If a circular logic is used, more sophisticated calculation techniques must be used, which are often integrated into existing software. A good example of this type of dynamic and systemic model is TrnSys, which solves coupling problems between component models [KLE 12]. It itself offers a library of component models, but accepts any other model. This is a very stimulating framework for CSF because it allows us to focus only on strictly necessary developments, which increases the robustness of the simulation while saving time. The use of an energy system model in CSF is varied and mainly concerns the understanding of the studied energy system, the study of the effect of alternative components within the system or different arrangements between components. In the last two cases, we can talk about virtual experiences. As already mentioned, the standardization of global performance indicators most often uses more or less simple simulation tools (see the case of buildings, Chapters 12 and 14).

11 Conducting the Evaluation

Several points are particularly important when implementing CSF and should be considered. 11.1. Publication CSF is always the subject of a written report, the form and status of which are discussed from the outset with the proxy. It should be planned to make it public, unless the normative aspect of the evaluation is important. First of all, it is customary to introduce the context in which the CSF was carried out: – Who is the proxy? – What are the objectives? – How does the object relate to technology and practices? – In what way is the object innovative or exemplary? – What are the reproducible aspects? – What are the local characteristics that influence the functioning or use of the object? For more information, see Chapters 5 and 6 on the human context and Sankey diagrams. A physical description of the system components is essential; a description of the context makes it possible to better locate the knowledge resulting from CSF, mainly its limits and possibility of generation. To encourage a wider audience than just the energy experts to read the study report, a ten-page synthesis of the most important conclusions, in a language that is as accessible as possible, is a proven solution.

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

138

Energy Transition

Particular attention must be paid to the drafting of the conclusion, which is often the first part read, the part that will form the initial impression, sometimes even the only one carefully read by some of the actors. It is therefore necessary to exclude any value judgment and to analyze the choices made by the designers, taking into account the time and conditions under which they were made: what information was available and how reliable was it, at what stage of the project were the choices made, by whom, etc.? For example, the conclusion of the CSF on the Solar City (see [BRA 02]) was very difficult to formulate because the project was initially highly publicized, as shown by the press clippings, then heavily criticized by the same press during the evaluation work (Figure 11.1).

Figure 11.1. French press clippings from July 17, 1995, Tribune de Genève, at the inauguration of the Solar City Complex (left) and 5 years later, on June 7, 2000 (right)

The conclusion of the report, written in 2002, was as follows: “The detailed study of the thermal performance of the Plan-les-Ouates Solar City has led to a number of conclusions, which we hope will be useful for the community. The main results are: – the thermal index of the Solar City is 246 MJ/m².year; – the solar coverage is worth 19.6%, at a price of 13 cents per kWh with subsidy and 20 cents per kWh in its absence.

Conducting the Evaluation

Without this study, many points would have remained in the shadows and the experience might finally have seemed unsatisfactory. In fact, the reality is much more positive. If the results may seem disappointing compared to what was expected, it is in fact the result of a confusion between the objective to be achieved at the time of project development and the actual situation. This confusion was fuelled by the contradiction between the innovative and pilot nature of these buildings and a certain amount of media hype at the beginning of the operation, which generated many expectations. Finally, a level of complexity that was too high for the technical installations led to difficult commissioning, malfunctions and major maintenance. The additional costs thus generated were not offset by the gains from energy savings and ultimately resulted in the image of expensive heating for tenants, and therefore misinterpreted as high energy consumption. The results obtained remain good compared to other experiments of this type, especially if we take into account that: – the beginning of the design is almost 10 years old and the construction is 5 years old; – it is a ‘life-size’ experiment, carried out in the traditional organization of construction; – high indoor temperatures: with standard conditions (20°C instead of 22.5°C measured), the thermal index is reduced to nearly 200 MJ/m² year instead of the 246 measured. There are many lessons to be learned from the Solar City: – a well-controlled exterior component remains the keystone of an energy-efficient building; – the energy concept must be simple and coherent; – the level of complexity of the technical installations must be manageable during the various stages: planning; execution and commissioning including adjustments and operation. Currently, energy prices are low relative to labour costs and it is important to properly assess the costs of operating complex facilities. Innovative pilot plants cannot work for sure and it is essential to monitor and evaluate them properly in order to benefit from the experience and improve what can be done.

139

140

Energy Transition

The experience of the Cité de Plan-les-Ouates can be described as positive. Although some weaknesses have been identified, the overall picture remains exemplary: – the concern for the common good at the origin of the project; – the risk taking by those who carried out the project and requested an independent evaluation; – the results obtained, even if they are a little below what could have been achieved with the investment made. The analysis carried out showed that an index as low as 160 MJ/m².an could be achieved in the traditional construction organisation, thanks to a well-managed exterior, simple but well-adapted technical installations and a significant contribution of renewable energy (1/3)”. 11.2. Summary of the CSF process Overall, CSF is a long-term process. Figure 11.2 shows the different steps and main interactions. First of all, it is necessary to contact the energy system to be evaluated and clearly define the objectives of the in situ evaluation that constitute the “CSF”. The human and energy aspects must be sufficiently well identified before the evaluation work itself begins. Chapters 5 and 6 are devoted to this approach. The system to be evaluated must then be measured, i.e. the objectives of the measurement must be defined, the system must be broken down into subsystems or components and the levels of detail at which they are to be assessed (Chapter 7). This first step should lead to a measurement concept and then to the installation of the resulting equipment. Then, we have the long and tedious data entry and collection, which must be organized and observed as closely as possible to reality. The observation phase is a key step in the process because it makes it possible, on the one hand, to react to any problem in the measurement system, to complete or adapt it, and, on the other hand, to prepare the analytical work. Similarly, the definition of general indicators and the calculation of a first estimate, even though it will appear very early, give valuable indications and make it possible to appropriate the system energetically. Chapter 8 is devoted to these two aspects. The subsequent analysis phase, which often takes place in parallel with the two previous phases, makes it possible to understand the system and its reactions to disturbances or conditions of use, and to clearly identify the interactions between components. This phase of understanding the operation may allow the measurement system to be adjusted in order to better investigate certain quantities.

Conducting the Evaluation

Figure 11.2. Diagram of the CSF process

141

142

Energy Transition

Input/output curves and signatures (Chapter 9) are the most commonly used methodological bases. This phase also explains how the system and its various subsystems and components work. Most of the time, the method used is the modeling of the individual components and the energy system they compose. Validation of any modeling is convenient in the context of CSF because there is a wealth of data on the functioning of the components and the system in general. See Chapter 10 for more information. By way of synthesis, the simulation phase makes it possible to predict the operation of the system subjected to another use or of a differently composed system. This virtual experiment phase has an important amplifying effect; it makes it possible to define the best uses or arrangements on a very robust basis since it is based on real practices evaluated in situ.

Part 3

The Practice of CSF

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

12 Challenges of Innovation: Summer Overheating in an Administrative Building

This study was carried out in the summer of 2006, and it concerns a situation of high summer temperature in the offices of a recently constructed administrative building with no air conditioning. The contracting authority had asked to establish whether such a situation existed, and if so what was the cause and therefore who was responsible. Any proposals were also solicited. This case illustrates to the point of caricature the link between use and functioning. 12.1. Background information From the beginning, occupants complained about high internal temperatures in summer; these complaints were relayed by site managers to the owner. The architects and the design office in charge of the energy concept had to explain these results. At Easter 2006, when the case was submitted, the building managers accused the designers of having delivered a building that did not meet the standards and state of the art. The designers considered that the occupants were not using the planned cooling systems properly, and therefore directly blamed the managers who had not been able to inform them. The situation of “overheating” was recognized by each party, but without being clearly identified and characterized. Everyone had made temperature measurements that supported their own claims.

146

Energy Transition

Faced with this latent conflict, the owner decided to have a study done to clarify the situation. After requesting an offer from two research laboratories, the Centre universitaire d’étude des problèmes de l’énergie was chosen to carry out this study. There were several constraints: – a time constraint, as summer was approaching and conclusions had to be delivered for the autumn, so only one summer was available. This might turn out to be cool, and thus it may not conveniently allow observation of “overheating”. In addition, the summer holidays limited the necessary skills available; – important issues for both building managers, who had to manage the building for several decades, and designers, who could face penalties and other problems if the building was found to be non-compliant; – complexity of defining summer discomfort, and of having relevant and authoritative indicators on this subject: the possibility existed of not being able to make a clear decision. It immediately became clear that the actors had to be strongly involved in the study and be closely involved in the work at all levels, from the investigative method to the conclusions. They were competent, knowledgeable in the field in question and very directly involved. Five meetings called “coordination sessions” were therefore planned between June 27, 2006 for the first findings and November 3, 2006 for our conclusions. During each of them, results were presented, the conclusions were clarified and the rest of the work was presented and discussed with the actors gathered. Great care has been taken to respond to questions, objections or proposals from all stakeholders. It was of the utmost importance that the content of the expertise and its conclusions be seen as legitimate by all parties, so that they would adhere to it, even though some conclusions were not in their favor. The search for points of convergence and agreement on findings or analyses was a constant concern; very few, if any, points were in the end problematic. A large part of the work was to build consensus about the real situation, as we will see later. The actors were easy to identify: building managers, designers, project owner representatives and occupants. The latter were not invited to the coordination sessions due to their number (about 100 offices are involved) and the difficulty of representation. There were constant exchanges with them during the many hours spent in the building during the summer.

Challenges of Innovation: Summer Overheating in an Administrative Building

147

12.2. Description of the building This six-storey metal and concrete building includes a conference room, offices and, on the top floor, a library. The orientation of the main facades is as follows: east/west. The typical floor plan includes a central service core and two corridors on either side of this core that serve offices facing east and west. Sun protection is provided on the south, east and west façades by fixed slats that also provide a railing for a small service walkway and by slat blinds on the outside of the windows. The conference room and library are air-conditioned, the corridors are cooled and the offices are simply naturally ventilated. This natural ventilation is provided by a 0.4 m × 2.2 m opening, which is sized to allow night cooling of three to four air changes per hour. This strategy should ensure summer comfort in the offices, which are naturally lit by large glazing (the entire width of the office) on an 80 cm sill.

12.3. The measurement concept and initial findings The first challenge was to characterize the overheating situation as objectively as possible. To this end, it was decided to have the most accurate and complete thermal image possible of the building, with four floors of non-air-conditioned offices each containing 25 offices, or 100 premises mostly oriented east or west. For this purpose, we selected 20 offices and placed recording thermometers (one point per 15 minutes) in them. The installation was carried out in the presence of the occupants, and the exact location of the measuring probe was chosen with their agreement each time. It was important to avoid adding any unnecessary tension to the situation. The results are presented in the graphs in Figure 12.1. During the assessment, many other specific measures – which sometimes took the form of real experiments – were carried out to answer the questions that appeared as the CSF progressed; they will be repeated later. Temperatures fluctuate differently depending on office orientations, occupancy and management of window opening. In general, office temperatures were quite high, especially during the hot month of July. We can clearly perceive the effect on the occupant by “zooming in” on a hot day, July 19, 2006 (Figure 12.2).

148

Energy Transition

Figure 12.1. Eastern office temperatures during a hot week, July 2006. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

Figure 12.2. Effect on the comfort of the occupant on a hot day. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

Challenges of Innovation: Summer Overheating in an Administrative Building

149

The results are also reported in Table 12.1. °C

Tmin

T 9h

T 12h

T 17h

Tmax

Max time

T ave

Outside measured on site

20.5

22.5

26.7

34

34.9

19h

27.1

4-505 Correct strategy

26

26.2

27.3

29.2

30.1

20h20

28.1

5-502 Unoccupied

28.1

28.2

28.4

29

29.4

22h

28.8

3-511 Inverted strategy

28.7

28.8

29.2

32.5

33.6

18h50

30

Table 12.1. Temperatures observed outside and in three offices, July 19, 2006

On this hot day (min: 20.5°C, max: 34.9°C, average: 27.1°C), a correct window management strategy (night opening, day closing) maintains the indoor temperature at 29.2°C at 5 pm, the maximum being postponed to 8:20 pm (30.1°C). On the other hand, a reverse strategy adds 3°C throughout the day, resulting in significant discomfort. It should be noted that an unoccupied office (i.e. without internal load but with openings constantly closed) remains very stable (1.3°C daily fluctuation) and at a higher average level than the occupied office, but well managed. This shows the good inertia of the building and its intrinsic quality in relation to summer overheating. It should be noted that the offices located to the north proved to be the hottest because they lacked sun protection, despite the morning irradiation present on this facade throughout the summer. This problem was corrected by an indoor sunscreen, a tested solution that proved to be effective. 12.4. Overheating indicators: strict application of the standard The overheating indicator used by the design office for the energy concept was directly derived from the standards of Swiss engineers and architects (known as “SIA standards”) in force at the time of the design and commissioning of the building (2005). Of the three standards that addressed issues related to air conditioning and ventilation installations, SIA 382/3 defines, among other things, the “proof of need” for air conditioning in a building [SIA 92]; it was therefore logically used to show that the building’s design made it possible to do without mechanical air conditioning.

150

Energy Transition

12.4.1. Proof of need according to standards The standard in question stipulates that proof of need: “[…] is provided by detailed calculation, when it can be shown by a recognized calculation method that without mechanical cooling the air temperature would reach unacceptable values. This is the case when the temperature exceeds the upper limit of the fluctuation range in Figure 12.3 for more than 30 Kelvin hours per year (KH/year). For the calculation of these hours, the following conditions are determined: – observation period from April 16, 1987 to October 15, 1987; – schedule of use of the service (office hours) according to section 6.1.2 of the SIA standard V382/2; – ambient air temperatures are related to the maximum daily outdoor air temperatures for the day in question; – no values occurring during hot days (max ext T > 30°C) are taken into consideration.”

Figure 12.3. Extract from SIA V382/3, definition of the summer comfort standard

Challenges of Innovation: Summer Overheating in an Administrative Building

151

Very precise conditions of use are given with the possibility of adaptation as required: “any possible deviation must be justified”. However, even anticipating realistic use to model the summer temperature of the building, it can only be a predicted use at this design stage. It is therefore a building operating standard, considered to pertain at this stage to the technological system and not yet entirely to the energy system – it will do this when it is handed over to managers and occupants. As seen in Part 2, Chapter 10 on modeling, the characterization of a component or system from the point of view of the technological system “is based on standardized use of the latter: temperature level, sunshine, etc. These tests are carried out in the laboratory according to very precise standards. The operating conditions imposed are not always fulfilled in a real system and it is therefore necessary to ‘renormalize’ the observed operation to the operation provided for by the standard. This re-normalization process usually involves a component operating model, which makes it possible to virtually reproduce its operation under standard conditions”. In the case of the summer comfort signature, the maximum outdoor temperature of the day is considered the main external stress that determines the most important system status variable: the indoor temperature. 12.4.2. Use of the standard by the design office when defining the concept The design office in charge of the energy concept carried out a series of simulations during the project phase, which strictly complied with the SIA 382/3 standard. The modeling is based on the well-known software TrnSys, in which conditions of use well adapted to the situation were considered, in particular opening the windows during the day and at night. Very concrete constructive solutions were chosen to encourage opening windows at night: narrow but high windows that can also remain open at night because they are protected from rain and intrusion, good solar protection to limit heat gains, high thermal inertia and an open false ceiling to access this thermal mass. Building instructions have been provided to managers to promote good practices. The calculation showed moderate overheating in the building, estimated at a maximum of 17 K.h for the west offices, i.e. below the 30 K.h limit that defines the proof of need. The building therefore did not require mechanical air conditioning.

152

Energy Transition

12.4.3. Comparison with the real situation To test whether the proof of need is indeed negative, the comparison between summer thermal conditions observed in situ and those predicted during design cannot be directly done via the exceedance indicator (K.h). Indeed, it is the strict application of the standard that makes designers liable and this standard says nothing about the actual functioning of the building in use. As already seen, the performance indicator must be renormalized to standard operating conditions. In our case, the greatest disturbance is the outdoor climate during the CSF (summer 2006), which is much warmer than that considered in the standard (30 days with a maximum temperature above 30°C against 15 in the standard, seven days with a maximum above 33°C against zero). The design office thus simply reran the dynamic simulation, with the only modification being the weather measured in situ by us. The comparison between the simulated and measured conditions is a validation test of the modeling carried out ex ante; it has also become a test for the application of the standard, insofar as it is based on the same modeling. This comparison is shown in Figure 12.4, for offices located on the east side during a typical summer week.

Figure 12.4. Comparison of simulation/measurements in offices on the east side of the building, hot week (top) and heat wave (bottom). For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

Challenges of Innovation: Summer Overheating in an Administrative Building

153

The concordance can be described as very good if we take into account the following two remarks. The effect of opening windows at night is clear: the simulation assumes a complete opening of the window, so its quality must be judged by considering offices whose occupants correctly manage the openings (e.g. office 4-522 between July 13 and 15). Daytime temperatures are overestimated by the simulation because the window was assumed to open during the day, which was not generally the case during the 2006 heat wave. In conclusion, observation of summer temperatures in 2006 confirms the absence of proof of need according to the standard in force, even though comfort is not ensured. We therefore had a real problem with this standard, which has since been modified by including, days of extreme heat on the one hand, and on the other hand, by adapting the performance criterion to less than 100 hours of excess but without accounting for its intensity. Unfortunately, continuing to use 1987 weather does not allow us to deal with the current hotter summers, as opposed to weather based on a hot, but not extreme, year. 12.5. Building consensus During this assessment, we tried to carefully listen to the various stakeholders and answer their questions as well as possible. Thus, one by one, we raised these different hypotheses to arrive at a conclusion that was as unanimous as possible. These assumptions are addressed in the following points. 12.5.1. Is the indoor humidity in the offices too high? This was an assumption raised by managers. The relative humidity was measured in two offices. For the comfort assessment, we used the Humidex1 method, developed by Canadians, which links the feeling of comfort (or discomfort) to indoor temperature and humidity. It appears that the two offices examined, seen as representative of the building, were “slightly uncomfortable” according to the terms of the method. Humidity is therefore not an important factor in this case. 12.5.2. Is the ventilation through the windows as predicted? Another hypothesis that was proposed by several stakeholders at the beginning of the study was that the ventilation through the windows was lower than the design office’s predictions. It is true that the size of the openings is not very generous

1 Available at: www.cchst.com/oshanswers/phys_agents/humidex.html.

154

Energy Transition

(40 cm wide by 2.2 m high) and is also obstructed in the lower part by the protective glass panels on the footbridge. To exclude this hypothesis, measurements of the ventilation rates in an office used as a conference room were taken during the night of July 12–13, 2006, during a hot period (outdoor temperature always above 20°C at night, having reached 30°C the previous afternoon). A second measurement was taken on the night of August 24–25 to sample conditions where there is a greater temperature difference between inside and outside. To do this, the velocity of fresh air entering through the lower part of the opening was measured, as well as that of the upper part where the air from the room is warmer than the outside. With the office door closed, there is no through ventilation and there is a neutral level (still air) towards the middle of the window. In view of the low speeds involved, an anemometer sensitive to cm/s was used. About 40 measurement points were used. This measurement was carried out with all the actors present at the coordination sessions (see Figure 12.5).

Figure 12.5. Ventilation rate measurement

Figure 12.6 shows our results as well as a comparison with the calculations of the design office. The rate varies with the temperature difference between inside and outside. We can see that the measurements are consistent and rather higher than the calculations. We also note that the partially open position of the blinds has little effect.

Challenges of Innovation: Summer Overheating in an Administrative Building

155

Figure 12.6. Ventilation rate measured as a function of the indoor/outdoor temperature difference, summer 2006

12.5.3. Is the ventilation, even in accordance with predictions and properly used, sufficient? This question could be asked in view of the still-high morning temperatures in the offices (25°C or more). Unfortunately, it was not possible to carry out a convincing in situ test due to the practical difficulty of effectively doubling the ventilation rate through the window. This study was carried out virtually by numerical simulation, an acceptable situation provided that the simulation process was properly validated. It was just a matter of virtually doubling the size of the window, all other things being equal, and comparing the two calculations. The results are shown in Figure 12.7. Doubling the ventilation would have saved 1–2°C in the morning, a difference that decreases to about 0.5°C in the late afternoon. Unfortunately, at this stage, it was not conceivable to double the number or size of windows. Another solution was digitally tested by the design office: the addition of a vertical outer blade to create an upstream overpressure and a downstream depression caused by local night winds, and thus increase the air speed. However, the results were not very convincing, and it was decided to abandon such a system.

156

Energy Transition

Figure 12.7. Calculated effects of double ventilation and absence of false ceilings. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

12.5.4. Do occupants use night cooling as intended? This was a hypothesis put forward by the building designers and the technical office, who wondered whether “the occupants were using their windows to ventilate at night”. A positive answer could be given for the majority of cases by three methods: direct viewing of the facade by photos regularly taken in August 2006, measurement of the temperature difference between the air above the false ceiling and the air below the window where outside air enters, in about 15 offices, also in August and by comparison between measurement and dynamic simulation. These methods converge to show that a significant number of occupants do use the window correctly. 12.5.5. Is the false ceiling an inconvenience? The last hypothesis provided by the building managers: false ceilings do not allow the thermal mass to play its full role, and their poor design explains the thermal behavior of the building.

Challenges of Innovation: Summer Overheating in an Administrative Building

157

Figure 12.8. Photo of the false ceiling

To exclude this hypothesis, it was decided to compare two identical offices, one having been stripped of its false ceilings. We measured these two spaces in detail, in order to clearly define these subtle effects during one week. No significant effects due to these false ceilings were found, and the results are in line with the calculations that were made to define the energy concept (see Figure 12.8). 12.6. Conclusions A meeting with the occupants and managers of the building was held in spring 2007 to present the results of the investigation and remind them of good practices. According to the owner contacted in early 2018, the current situation is no longer a problem. This CSF makes it possible to give a number of general remarks concerning studies with a pronounced normative character: – the work must be done in close collaboration with all the actors, who must first be well defined; – always try to answer the questions asked or, if they are poorly worded, clarify and rephrase them explicitly; – it is necessary to seek as many points of convergence as possible, as well as to reach consensus where possible; take note and report on irreducible differences or oppositions;

158

Energy Transition

– problems must be objectified; for example if office users complain of being too hot, find out whether this means objectively high temperature or humidity, whether it is due to a lack of ventilation, whether it is a desire for air conditioning or whether it is not ultimately an unpleasant atmosphere within the company and not the building. More generally, this example also illustrates the very general fact that an innovation is always more difficult to implement when use is more closely concerned. Here, operation and use are totally confused since the opening and closing of the windows are manual actions.

13 Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

Audit is the most well-known form of evaluation. From a general point of view, the audit of an energy system is a process carried out by an expert pursuing one or more objectives using appropriate tools. The most developed is the energy audit, which consists of detecting and highlighting opportunities to increase the efficiency of the energy system. It is about building on the knowledge accumulated by the team. The study presented here began in 1991 and is still ongoing at the time of writing. This case is in line with the logic of innovation with a deviation and feedback from previous experience, which has allowed us not only to develop knowledge, but also to prove the effectiveness of simple solutions. There was also a handover to a local actor, who was more competent in the implementation of the selected solutions than academics. Finally, this case is well documented and can be visited by everyone throughout the year. 13.1. The context of the study Valère Castle is a group of old buildings (13th Century) that dominates the city of Sion (Canton of Valais), part of which (800 m2) was transformed into a museum (opened in summer 2000) as part of the renovation of the complex (see Figure 13.1). This change of use has raised the problem of new indoor temperature and humidity conditions and how to control them. Until the renovation, the building was uninhabited and indoor climatic conditions were in balance, depending only on the external conditions and the physical characteristics of the building (very high inertia, no insulation and strong ventilation). The planned new use of Valère Castle, as a museum, requires very strict temperature and humidity conditions to be

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

160

Energy Transition

maintained. The study commissioned by the Canton of Valais Building Service focused on the possibility of reconciling requirements for the conservation of the museum’s collection objects with those for the conservation of the current building exterior components (light and reversible intervention). A local design office had already considered installing electric heating elements everywhere on the basis of a very simple calculation in steady state with estimated thermal parameters. The building’s energy manager wanted at all costs to reduce this oversized electric heating to limit the power demand.

Figure 13.1. Exterior photos of Valère (Vs, CH) and section of a building

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

161

Until now, the temperature and humidity of the building have been balanced according to external conditions. Maintaining a constant temperature and humidity (a priori 15°C, 50–60%) generates flows of energy and water vapor. The main risks of degradation of the external components are the possibility of condensation and water saturation. On the other hand, maintaining the indoor thermal conditions under consideration requires very consistent technical installations, which are not feasible in the building under consideration. So, it was a question of, from what is possible, defining the resulting indoor climate and checking its compatibility with the structure of the building and the operation of the museum. One point to be mentioned about this study: several teams have succeeded each other. For the design of the system, the University of Geneva worked until 1997, then from 1997 until 2002, the Valais Engineering School took charge of the implementation of this concept into a technical system from the dimensioning of the components to their adjustment, and finally the Canton of Valais Building Service supported the whole process and is still responsible for the proper functioning of the buildings in Valère. A previous CSF for a very simple cooling system, using fresh air from a deep cellar to cool offices in the attic, had a fundamental impact on the approach taken to solve the significant constraints in Valère. This CSF performed on the Aymon building will first be quickly presented, focusing on aspects that have had a significant impact on the definition of Valère’s energy concept. 13.2. The Aymon CSF Built during the economic recovery of the late 19th Century, the Aymon building was aligned with the location of the western wall of the city walls of Sion, which were demolished half a century earlier. At the time, the bishopric and the palace of government formed the front of the old town facing the historic Plaza de la Planta. The Aymon family used this building until the 1960s, when the Canton of Valais became the owner. Previously, the building expanded southwards with a commercial annex. The DIP (Department of Public Education) occupies the premises, which were by then dilapidated and uncomfortable. In 1983, the first renovation studies began. The work lasted two years, from 1988 to 1989. The building now houses the DIP and two shops on the rue de Lausanne. During studies, it was quickly decided to install offices under the attic, which immediately raised the problem of summer comfort given the excessively high temperatures, easily predictable under roof windows and under a roof that is very exposed to the south and west. In addition, this building has deep cellars, which are cool during the summer, odorless, radonfree and used for archival storage.

162

Energy Transition

The idea of taking this fresh air from the basements and blowing it into the overheated rooms in the attic was retained. A similar system designed for the attic room of the nearby Building Services Department had already been producing convincing results for some years (see Figure 13.2). Simple, because it consists of a fan in the cellar, a tube running through the floors and distribution ducts in the offices, the passive air-conditioning system (passive because it uses mainly natural phenomena) was installed without any problems. The dimensioning of the ducts and fan was based on rough assumptions: available space limits or maximum air velocity. Such a system, if operational, and related to a large number of buildings, can effectively slow down the increase in electrical power required for air conditioning in offices in summer. If for the Aymon building, we could simply adapt this system as best we could to its old walls, this would not be the case for buildings to be built that require cooling of the upper floors. For such systems to work properly, it is necessary to dimension them from the beginning for the service they have to provide. It is therefore essential to know what happens when a system is already built and to try to understand the various phenomena, in order to establish the rules of dimensioning to be respected when building future systems. To this end, contact was established between the Valais Canton Building Service and the CUEPE/Energy Group of the University of Geneva. A report was produced as a result of this collaboration [MEL 91] and will serve as the basis for the following. Information on this study can also be found in [LAC 95]. Although the measurement part was of short duration (the three months of summer), the CSF classically took place over a longer period: – April 1989: contact between the Canton of Valais and the University of Geneva; – summer 1989: measurement of office temperatures, the system is not in operation; – spring 1990: obtaining financial resources; – summer 1990: measurement; – autumn–winter 1990–1991: analysis of measurements, modeling and scenarios; – May 1991: report; – October 1991: drafting of a technical sheet describing the system installed in Aymon and its results, and proposing a simple rule for the size of such systems;

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

163

– summer 1995: commissioning of an identical system in the Cantonal Economat building (16th Century), based on the knowledge acquired by the Aymon CSF.

Figure 13.2. “Aymon” building: view of the building and diagram of the air/ground exchange ventilation system with the cellar

13.2.1. Measures and preliminary findings During the summer of 1989, 10 temperatures were measured using self-powered, wireless recording thermometers with excellent accuracy (±0.1°C) after individual correction (0–1°C). Four temperatures related to the transit of air through the cellar, while two offices were thermally characterized by three temperatures (low, medium and high). The 1990 campaign was more complex, and measures concerning sunshine, air flow, outdoor humidity and humidity in the cellar and office, electrical power and the positions of doors, windows and blinds were included. This data was recorded every minute, averaged and stored every hour. It was decided to launch a campaign of average complexity (29 measurement points): this is a compromise between the need to have detailed data available that is essential to understand this system and the desire to remain as simple as possible. Spot measurements completed the measurement concept and concerned the flow rate and electrical power of the fan for the different supply voltages, the pulsed flow rate in offices and the electrical power consumed in an office. These measurements revealed a weak point in the initial system: parasitic hot air from the elevator shaft was being sucked in by the fan. This was corrected in mid-August 1990 by connecting the fan inlet to the bottom of the cellar via a buried duct.

164

Energy Transition

The measurements confirmed the general impression of occupant satisfaction. Figure 13.3 shows the temperature in the office at three different locations: at 0.1 m from the floor; at working height, 0.9 m from the floor; and under the roof, or at 3.5 m from the floor during the hottest period (August 6–19). The fan was turned off for three days, the weekend of August 10–12. This figure highlights the homogenizing effect of ventilation temperatures, as well as the cooling effect in the short and medium terms. Note that occupants can choose the supply voltage of the fan located in the cellar through a button located in a common area of the cooled section. An intermediate voltage (160 V for a maximum of 380 V) was most often selected. This mastery of the use of the system by the occupants was very important for the success of the system thanks to the appropriation that the main stakeholders were able to develop. It also turned out that this setting was close to the energy optimum.

Figure 13.3. Three-level temperatures in an office from August 6 to 19, 1990, with corresponding weather data: outdoor temperature and sunshine G (bottom)

The temperatures classified from inside the office (Figure 13.4) are compared without the system (1989) and with the system (1990) during the same period from July 27 to September 9. Knowing that the weather was warmer and internal loads were higher in 1990 than in 1989, this figure shows the significant gain provided by the cooling system. Thus, over the 1,000 hours of this period, only about 30 exceeded 26°C (in fact, this was when the system was shut down at the weekend in mid-August for the improvement work already mentioned), whereas there were 560 hours before the system was installed.

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

165

Figure 13.4. Comparison of the classified temperatures of an office before (1989) and after (1990) the installation of the cooling system

Figure 13.5. Input/output relationship of the cooling of outside air passing through the cellar. Input: difference between outside temperatures and thermal mass; output: cooling of outside air during transit

Two important findings were made. The first concerns the cooling and dispersal of outside air as it passes through the deep cellar before being sucked in by the ventilation system. In reference to a temperature of the bulk of the cellar measured 50 cm below the clay floor of the cellar, an input/output diagram could be drawn (Figure 13.5). The input is considered the temperature difference between the

166

Energy Transition

outside air (Text) and the cellar mass (Tc4), and the output is the cooling of the outside air (Text-Tc1, Tc1 being the temperature of the air at the bottom of the cellar that is drawn in). Daily mean values are considered, with daily fluctuations in the suction temperature Tc1 being well below one degree (see Figure 13.6). The linear relationship obtained means that the air entering the ventilation system can be considered thermally as a mixture of 84% air at the temperature of the thermal mass and 16% air at outside temperature. In other words, day-to-day fluctuations in outdoor temperature were cushioned by 84%, eliminating heat waves and providing air at about 20°C throughout the summer, as the mass recharges with coolness in winter. The other finding was the importance of heating the air thus cooled by the ventilation system (see Figure 13.6). For example, air entering at 20°C is blown at 23°C into the offices. In the case of such an installation intended to cool rooms, this temperature increase must be minimal. By carrying out a qualitative analysis of the causes, we can mention not only the losses by parasitic infiltration, either at the entrance to the system or in the part of the ventilation ducts located upstream of the fan and therefore in depression, but also losses by heating in the fan and in the ventilation ducts (pressure drops) as well as losses by heat transfer in the duct network when they cross warmer spaces.

Figure 13.6. Air heating in the ventilation system

Simple constructive measures make it possible to limit this rise to less than one degree, greatly increasing the efficiency of such systems which work with low temperature differences: insulated ducts in hot parts and sufficient diameter, fan motor outside this duct.

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

167

13.2.2. System modeling The office could be modeled in a simple way with a dynamic model, called single-node because it assumes that the temperature of the space to be modeled is uniform (Figure 13.7).

Figure 13.7. Thermal diagram of the desktop model

In this model, the internal temperature of the Tint office is thermally coupled with the external temperature Text via the heat transfer coefficient Kb. All other contiguous spaces are assumed to be at the same temperature, and therefore, no heat exchange takes place. Two thermal energies heat the office atmosphere: solar gains Qs and internal gains (people, electricity from appliances, lighting) Qint. The cooling energy Qref, which will be negative and effective if the supply air temperature from the cellar is lower than the office environment, will offset the heat inputs. The balance of the energy will accumulate in the thermal mass of the space if it is positive, causing a heating ∆Tint or, on the contrary, a cooling ∆Tint if it is negative. The energy balance achieved during the time interval ∆t is written as follows: +

∗

+

−

∗(

−

) ∗ ∆ =

with: – Qref : the cooling energy in (Wh); – Gs: the solar radiation in the roof plane (Wh/m2); – Seff : the effective catchment area (m2); – Qint: hourly internal loads (Wh); – Kb: the heat transfer coefficient of the building (W/K); – Tint: the average indoor temperature of the office studied (°C);

∗ ∆

168

Energy Transition

– Text: the outdoor temperature (°C); – Cb: the storage capacity of the space (Wh/K); – ∆Tint: the temperature variation (°K); – ∆t: the time interval (h). From an ambient temperature Tint at hour h Tint(h), and by applying the balance described above, we can obtain the indoor temperature of the next hour Tint (h + 1) as (

(ℎ + 1) = ))/ −

(ℎ) + (

+

∗

+

−

∗

The energies here are expressed in Wh. The hourly cooling energy at time h + 1 is proportional to the hourly rate of forced air flow in the space represented as Flowrate and the average temperature difference between the forced air at time h + 1 and the indoor air in the office in the previous hour: (ℎ + 1) = 0.33 ∗

(ℎ + 1) ∗ (

(ℎ + 1) −

(ℎ))

Energy is expressed in (Wh), flow rate in (m3/h) and the proportionality factor 0.33 is the specific heat of the air in (Wh/m3.K). This is obviously an approximation because it would be necessary to consider the internal temperature of the space at time h + 1, a value that we want to evaluate. It can be obtained by successive iterations by calculating a first value of the indoor temperature from the previous hour, then a better estimate with the value obtained and starting the operation again until convergence. This is the operation carried out by the thermal software in so-called detailed buildings, but if the number of variables to be converged is large and the interrelationships between these variables are numerous, then these are long and complex operations, which considerably increase the workload of the simulation. Sometimes it is necessary. In our case, the high stability observed in the indoor climate when the system is active shows that we can trust the relevance of a simplified approach1. This “belief” in simplicity is the direct result of the knowledge created by CSF through system measurement and observation. Hence the importance of this observation phase where phenomena can be detected before processes are analyzed (see Part 2). The main advantage of this simplified approach is, on the one hand, that it is quickly programmable on a spreadsheet and, on the other hand, that it fress up time and attention for other points that would otherwise be neglected, such as here the significant and difficult-to-predict detail of parasitic

1 Detailed modelling of this room was carried out in [HOL 02a] with the ESP software, but without making any significant improvements to the knowledge already acquired.

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

169

infiltrations. Solar gains are obtained by considering an effective area Seff of capture of incident sunlight. This parameter can be estimated from the thermal and radiative properties of the elements of space, as well as determined from the measurements. The other two thermal parameters of the space represented as Kb and Cb are determined in a similar way. The internal loads of the office in question were measured. A detailed description can be found in [MEL 91]. The validation is shown in Figures 13.8 and 13.9, over a hot week in mid-August.

Figure 13.8. Simulation measurement comparison, period from August 6 to 19, 1990

Some comments can be made on Figure 13.8. The initial temperature of the simulation is the one measured at time t = 0. It can be seen that temperature peaks are less marked in the simulation than in reality when the system is shut down, which is explained by the fact that such a simple model is less relevant under these conditions. The results obtained when the system is in operation are fairly similar to reality, with the difference between the modeled and measured temperatures never exceeding 0.8°C. It should be recalled that the model parameters were based on measurements, but on a daily basis, the transition to an hourly basis highlighting the effect of inertia by diffusion, which is not a linear phenomenon over time. More meaningfully, because it can be directly used for dimensioning the installation, Figure 13.9 shows the same temperatures as the previous figure, ranked in descending order.

170

Energy Transition

Figure 13.9. Simulation measurement comparison, classified temperatures, period from August 6 to 19, 1990

Caution should therefore be exercised when comparing the simulation of the office temperature without a cooling system to measurements taken in 1989 before the system was installed. This is especially since the use of blinds and the opening of roof windows were necessarily different and adapted to a warmer indoor temperature. This probably explains why the simulated temperatures are slightly higher than those modeled (see Figures 13.4 and 13.9). In this validation step, it was not so much a question of testing the intrinsic value of the model but rather the relevance of its use in a given context. The knowledge acquired therefore concerns the context in which the modeling approach used works and makes it possible to better define the limits of its application. Thus, information on the boundary of a model at a node, during large temperature variations, will be taken into account for Valère and a more complex model will be retained, capable of better accounting for the dynamic response of very massive walls during powerful stresses. The complete system was thus modeled by chaining three coupled models dealing with the resource (basement to outdoor air heat exchange), the transfer of place between the cellar and the attic (heating along the route) and the application of the resource (the office thermal model) (see Figure 13.10).

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

171

Figure 13.10. Complete model of the Aymon system

Well designed, it is a self-regulating system. For example, an increase in the internal gains of the office will increase its temperature and in parallel the cooling power, which will produce regulatory feedback. Similarly, an increase in the temperature of the cold source will initially reduce the cooling power, thus increasing the temperature of the office, which in turn will increase the cooling power and stabilize at a somewhat higher value. This self-regulation is only possible because cold is produced from a resource that is sizeable in quantity but limited in quality: the supply air temperature in offices is not adjustable and is close to the desired comfort temperature. This is only possible thanks to the intrinsic quality of the protection against summer heat, mainly direct sunlight through the openings and diffused through the absorbent roof. A regulating effect also exists in traditional air conditioning but is much less active, because the supply air temperatures are much

172

Energy Transition

lower and are due to the fact that we work with a given cooling capacity. This is an important lesson for Valère, who will have to submit itself as much as possible to the conditions of this self-regulation, to thereby limit heat gains in summer. Two other points had positive implications for Valère. First, the detailed evaluation of this simple but little-known system had allowed for a good understanding. It had also confirmed that simplified modeling was possible because it is not so much the details of the functioning of each subsystem that must be well described as their interactions. The consideration of Valère’s very high spatial complexity (see plans and sections) could thus be avoided when modeling the system by considering only two typical spaces, with very different thermal responses to very different stresses, in favor of a deeper interaction between components, subsystems, parameters, system states and different factors influencing the indoor climate. It also demonstrated that it is necessary to be attentive to certain details that only experience can capture – as here the significant impact of uncontrolled parasitic infiltration. Finally, and in general, this CSF has created a relationship of mutual trust between the representatives of the Canton of Valais and the energy research group of the University of Geneva. The system’s performance proved to be very convincing; the analysis carried out was able not only to explain it but also to propose improvements that would further increase the system’s efficiency while increasing its robustness. The successful implementation of a third system optimized according to the lessons learned, which was carried out in the building of the Cantonal Economat, in parallel with the study of the energy concept of the Valère Museum, supported this excellent relationship. For a lower cost and lower power consumption, the summer comfort provided is excellent. The main improvement is a fresh air supply in the best possible place and a temperature increase during the transfer between cellar and roof, less than 1°C thanks to the thermal insulation of the ducts, good sealing of their parts in depression and the installation of the fan motor outside the duct. As a result, the flow rate could be reduced. 13.3. Return to Valère The study commissioned by the Canton of Valais Building Service therefore focused on the possibility of reconciling the requirements for the conservation of the museum’s collection objects with those for the conservation of the current building exterior components (light and reversible intervention) [LAC 93]. The situation analysis showed a set of very strong constraints that made the problem very difficult to solve without removing at least one of the constraints. Figure 13.11 shows the interaction between all the elements involved in the problem.

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

173

Figure 13.11. Interrelationships between the different factors involved in the problem

13.3.1. The building It is a listed building, worth visiting in and of itself. This situation requires caution, both in terms of the damage that the new use could cause and in terms of development that must respect these areas. These are two major constraints that will have a strong impact on the functioning of the future museum. It is thus not possible to modify the facades, externally and internally, or to impose technical installations that are too cumbersome. The renovation was therefore defined as “light and reversible”. The building which houses the museum is described in Figure 13.1. The surface area is 600 m2 on three levels with an additional 200 m2 in a neighboring building for the reception and cafeteria. It is characterized by thick walls, few openings and a complex room configuration, including many intermediate levels. Some rooms are wooded. Finally, only electricity is available on the site. 13.3.2. The building’s relationship with the weather The extremely massive building effectively absorbs daily variations in outdoor temperature, its response time being typically one week. For demands arising from the inside – those that will be introduced with the new layout – the situation is different, especially in wooded rooms, where the woodwork disconnects the indoor air from the thermal mass. As far as insulation is concerned, the only possibility considered is insulation of the attic floor, where there are large losses in winter and large thermal inputs in summer. Installing insulating interior plaster is also possible.

174

Energy Transition

13.3.3. The building’s relationship with the operation of the future museum Use as a museum involves several loads: – humidity released by visitors (about 50 g/hour per person, or 5 liters of water to be evacuated for 100 visitors remaining one hour); – thermal inputs from visitors, i.e. about 100 W per person; – thermal load related to the operation, i.e. basic lighting of 150 lux, representing 5 W/m2, switched on during the opening of the museum, and the presence of air-conditioned and highly illuminated display cases, the lighting being controlled by a presence sensor. These cases are a very significant load for summer, when a high indoor temperature increases the power released by the cooling unit, which will again increase the indoor temperature. This may lead to a divergent situation where high temperatures inside the building are maintained by this system. Particular attention must be paid to these cases. 13.3.4. The building’s relationship with the technical installations If heating and humidity do not pose serious improvement problems, the same cannot be said for equipping such a building with cooling systems and dehumidification. The impact of these technical installations would be too great, either in the form of a centralized installation, due to the installation of ducts for distributing treated air throughout the building, or in the form of decentralized installations, where thermal loads and condensates must be evacuated to the outside. They must therefore be excluded from use, as well as for reasons of energy saving and power peaks. Concerning aeration and ventilation, useful in winter to dehumidify and in summer to cool, the most suitable possibility is a general ventilation of the building: the fresh air comes from outside through the cellars and then sweeps through the whole building, from bottom to top. This ventilation must be modulated according to indoor and outdoor conditions. 13.3.5. The resulting indoor climate This will be determined by the interaction between all the factors described above and it is very unlikely that it can fall within the very narrow ranges desired a priori by the Museums Directorate (15°C ± 1°C and 50% ± 10% humidity all year round). Moreover, such a climate would be detrimental to the structure of the building through induced condensation, both in summer and winter. It will be subject to three constraints. First, it must be adapted to the structure of the building, i.e. for humidity

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

175

controlled at 50/60%, not to exceed the outside temperature by more than 15°C, to avoid any risk of internal condensation, especially behind the woodwork. Second, it must not harm the objects on display either, which requires small daily temperature variations and the relative humidity to be as stable as possible. Finally, the energy consumed and the peak power demand must be minimal. As we can see, this set of constraints makes the problem difficult to solve. It requires a dynamic simulation of indoor conditions that takes account of all these elements and their interactions. 13.4. Modeling and scenarios: proposal of the concept based on the “Aymon system” This building has many features in common with Aymon: a problematic change of use (attic/office; historical monument/museum), massive walls and deep and cool basements, even in summer, that are managed by the same people. It also has three major differences from Aymon: a desired indoor climate that is as stable as possible; stresses that affect this interior climate, which are significant, very variable, little known and directly applied inside the building; and an existing building but with a new use affecting the entire site. In order to be able to offer a suitable and robust energy concept, the experience acquired with Aymon was very valuable but had to be put into perspective. Thus, it was necessary to keep the simplicity of the concept and its modeling, to pay attention to details and to clearly identify the different interrelationships. Minimizing internal gains at source was a sine qua non condition. 13.4.1. Real in situ simulation of the new use In order to better validate the building model used to simulate the response of the indoor climate to the various stresses, two series of experiments were carried out, the first between December 3, 1992 and January 3, 1993 in Room F18, and the second between January 4 and March 3 in Room F21. The aim here was to take advantage of the fact that the building existed, that it would remain close to the level of its thermal properties and that it was empty. It was therefore possible to simulate in real time the thermal conditions of the future development to accompany its virtual simulation. This type of approach is very rarely used, despite the valuable indications it can provide. These experiments consist of intermittently heating with eight incandescent bulbs of 100 W each (from 9 am to 5 pm or from 5 pm to 5 am) and independently humidifying the room by boiling water at a rate varying between 150 and 400 g/h. During these experiments, the doors were carefully caulked to simulate the actual operation of the museum, where all the pieces will be subjected to the same stresses. Heating and humidification were switched on by clock, but with regular changes in conditions to cover the full range of possibilities.

176

Energy Transition

The measured values are as follows: indoor humidity at 2 m height, air temperature at three levels (ground, 1 m and 2 m), surface temperatures of the stone walls (thus, behind the woodwork if necessary), temperatures of the adjacent rooms, external conditions (temperature and humidity) and finally the electric power used for heating and humidification systems, the latter giving us access to the evaporation rate. The building’s custodians counted the total volume of evaporated water, which made it possible to check that the value of the rates was indeed accurate, within a limit of about 5%. All temperatures are measured with an accuracy of about 0.1°C.

Figure 13.12. Experimental device for measuring the thermal and water response of a space

Two rooms were chosen as representative: a room with completely wooded walls, thus cutting off access to the thermal mass of the thick walls, and a room with walls simply covered with plaster. The results are shown in Figures 13.13 and 13.14. In wooded room F18, the temperature increases by about 4°C after eight hours of heating with a power of 800 W, simulating the presence of six or seven visitors, and lighting, then it returns to its initial temperature in the same way, as soon as the heating is switched off. The neighboring rooms are affected these actions only very little. The same conclusions apply to the non-wooded room, with the important difference that the temperature change is limited to about 1.5°C, i.e. three times less. Woodwork is an obstacle to the diffusion of heat to the solid structure; it is moreover one of its roles to increase the comfort of rooms heated intermittently. For humidity, we note that wooded and painted materials are unable to absorb the large amount of humidity added to the air in the room. Greater stability of the relative humidity of the air is achieved in the non-wooded room F21 due to the presence of plaster. These experiments will not only be used to validate the building model, as described below, but have also very strongly shown the impact on the indoor climate of future visitor demands, resulting in unacceptable variations in conditions for the few wooded rooms present. It was therefore necessary either to remove the woodwork from these rooms or to limit the heat input to a minimum by restricting the museum interest in the content of these spaces to limit their visitor presence time.

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

177

First, a wooded room was heated with eight incandescent bulbs of 100 W each from 9 am to 5 pm. The temperature rises by about 4°C at the end of the heating, then drops back to its initial value during the night. Moisture, on the other hand, remains relatively stable thanks to the large wooden surface.

In the second step, the air was further humidified by simulating the presence of about seven people. Wood is no longer able to absorb water (350 g/hour) and indoor humidity reaches saturation (100%).

Figure 13.13. Responses of the indoor climate of the wooded room F18 to thermal and water disturbances simulating visitors

178

Energy Transition

First, a non-wooded room was heated with eight incandescent bulbs of 100 W each from 7 pm to 5 am. The temperature rises by about 1.5°C at the end of the heating, then drops back to its initial value during the night. The humidity remains very stable due to the large plaster surface.

In the second step, the air was further humidified by simulating the presence of about seven people. The plaster is no longer able to absorb water (350 g/hour) and indoor humidity is approaching saturation (100%).

Figure 13.14. Responses of the indoor climate of the non-wooded room F21 to thermal and water disturbances simulating visitors

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

179

13.4.2. Virtual simulation of the new use We have adopted Aymon’s CSF philosophy, both for the thermal concept to be applied to Valère and for the tools of analysis that were developed there. A chained model, as shown in Figure 13.10, was used with some notable differences. First, the space to be dealt with is the entire building, so there is no need for ducts: the air is swept from top to bottom of the building. The extraction of air from above does not produce heating by pressure drop; it is necessary to take account of probable infiltrations via leaks in the historic building thus put into depression. The modeling was done over a single space, although in reality, the warming accumulates as the same air passes from one space to another. Emphasis was placed on internal stresses, including moisture. 13.4.2.1. The building Aymon’s CSF clearly showed that a single-node model, where the whole thermal mass of the building is assumed to be isothermal, cannot correctly represent a massive space subjected to high internal stresses. Two solutions existed at the time: on the one hand, to slightly complicate the simplified model in order to preserve controlled simplicity in the approach to modeling the building, and thus to be able to concentrate on the many interactions between the building, thermo-hydraulic impacts, visitors, air-conditioned cases with heat rejection in the room itself, technical installations, etc. On the other hand, it was possible to use a very detailed dynamic building simulation program, which had the advantage of modeling the building well but with the disadvantage of severely limiting the possibilities of modeling other elements interacting with it. The SERIRES detailed simulation program, developed in the United States by the Solar Energy Research Institute in the 1980s, was used. Very efficient for heating buildings, it is unfortunately very inflexible when it comes to simulating unconventional internal loads. For example, it is impossible to satisfactorily simulate the presence of air-conditioned display cases. For greater flexibility, the LADY software, developed by the École nationale des travaux publics de Lyon, was used. It consists of representing the thermal characteristics of the building by two time constants, a short constant of around an hour that corresponds to the air, furniture and the first layers of the walls, and another much longer time constant (about 200 hours) that corresponds to the thermal mass of the walls. This program, based on

180

Energy Transition

measurements and SERIRES simulations, determines the parameters necessary for this simplified model. A mathematical representation of the building can then be used that is simple enough to work on in a spreadsheet, for example, without losing too much accuracy (0.2°C standard deviation). We therefore find ourselves in the situation of Aymon, where the modeling tool becomes simple to use, while being extremely open and flexible. Internal thermal stresses will all be measured in terms of power and only the total value is considered, which directly applies to the air in the room. All other parts are assumed to be subject to close stresses. The attic is supposed to be perfectly insulated. The moisture supply will also be mixed with the air in the room, and no regulating action of the materials has been considered, for two reasons. The first is, as previously seen, that materials cannot play an important regulatory role in the event of abundant moisture input, and the second is the high level of complexity introduced to take these effects into account. The indoor humidity variations calculated will therefore be slightly overestimated by the calculation. 13.4.2.2. Visitor attendance It is important to treat this problem correctly, as it directly affects the indoor climate through the humidity and heat released by visitors and the heat generated by the lighting of the display cases controlled by a presence sensor. Visitor attendance has a predictable component (there will be more visitors in July than in May, more visitors at weekends than during the week) and a random component (distribution of visits during the day and for the different rooms, dependence on weather or other activities in the city). To cover as many cases as possible, and especially to take into account peak days, the following assumptions have been taken into account: the total number of visitors for the year is chosen, for example 65,000; for each week of the year, a coefficient proportional to the number of visitors is set, which makes it possible to increase the number of visits during school holidays; a distribution between the different days of the week is set from 0 on Monday to a maximum on Sunday, and for each time step of the simulation (quarter of an hour), the number of groups of visitors who are actually present is chosen at random while respecting the total number of people who visited the museum on that day. The other hypotheses are: visitors stay one hour in the museum, there are 20 rooms to visit, the visit time is three minutes per room and visitors come in groups of two. 13.4.2.3. Internal loads In accordance with the specifications of the company in charge, the 150 lux are obtained with 5 W/m2 of electrical power. The lighting is switched on from opening to closing.

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

181

Very great care has been taken in simulating the display cases because they have a very important influence on the indoor climate, given the high levels of power involved since they are highly illuminated and air-conditioned at 15°C. The parameters that can be modified are the number of display cases in the room, the surface of the display cases, the average loss coefficient (in W/m2.K), the set temperature and the power of the cooling unit. 13.4.2.4. Technical installations Adiabatic humidification (without electricity to evaporate) vaporizes water vapor if the room humidity falls below a set point. This action cools the air in the room. No technical dehumidification process is planned. Heating starts as soon as the temperature is below the set point, which depends on the outdoor temperature as described above: 15°C for outdoor temperatures above 0°C, 10°C for temperatures below –5°C. Due to the building’s mechanical ventilation, air is extracted from above and can enter the building either directly from outside or through the cellars. In the latter case, the temperature of the incoming air is the average outdoor temperature of the previous day. The possibility of taking in air both from the cellar and directly from the outside has been provided for. The fan is controlled either on the basis of the absolute outside– inside humidity difference if the spaces are to be dehumidified or on the basis of the outside–inside temperature difference if the building is to be cooled. 13.4.3. Results of scenarios and proposals From this numerical simulation tool, the best energy concept was defined using different scenarios (ventilation rate, regulation, etc.). We present the one that was selected, namely a general ventilation of the building at a rate of 3 volumes/hour by aspiration from the attic by means of several fans. For the non-wooded room, the parameters selected for this scenario are: – number of visitors: 65,000 per year; – “standard” heating (800 W per room, set point 10/15°C for –5/0°C); – humidification of 150 g/h, set point at 50%;

182

Energy Transition

– ventilation of 3 volumes per hour, two from the cellar, one from outside; – ventilation instructions at 23°C and 60% indoor humidity; – two air-conditioned display cases per room, with quality interior lighting but limited to 150 W. The simulation results (Figure 13.15) show that during the heating period, the temperature remains between 10 and 15°C and the humidity is remarkably regular at 50%; during the non-heating period, and particularly during the critical summer period, the temperature remains between 20 and 25°C, similar conditions to those existing in the unoccupied state, and the humidity fluctuates between 50 and 70%, values that are quite reasonable. These results show that it is possible to make the new use compatible with the old structure with particular attention to new thermal loads. Finally, the effect of two parameters that could influence these results was studied: a doubling of the number of visitors and ventilation that does not use (or uses very little of) the cellar’s cold stock. These two factors do not have a significant influence on the results. Concerning wooded rooms, it appears, in comparison to the non-wooded room, that it would be wiser to not use display cases in the rare rooms of this type. This would eliminate any overheating problems, as thermal loads due to visitors would also be reduced by faster passage. 13.5. Implementation of the concept and commissioning by the Valais engineering school (now HES-SO Valais) The practical implementation of this concept was carried out by colleagues from the Valais engineering school, who took charge of this part. The manager, Michel Bonvin, had developed a simplified building simulation tool based on a two-node “bsol” model, similar to the one used. Driven by the same attraction to simplicity, he appropriated the energy concept and was able to transform it into a concrete object with unfailing determination. During the transformation, each room in the museum was equipped with an electric heater (from 700 W to 2,400 W), which was integrated into the furniture. Seven variable speed fans were placed in the attic, so as to ensure cross ventilation from bottom to top, thus taking advantage of the cool air accumulated in the cellars to cool the museum in summer. The control and regulation of heating and ventilation is ensured by a centralized technical management system.

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

Figure 13.15. Simulation results

183

184

Energy Transition

Many hours of surveying these historic sites were necessary to position the exhaust fans and design the air passages from one room to another. The result is convincing (Figure 13.16) and can be seen in situ during a visit to this museum.

Figure 13.16. Realization, Valère Museum, Sion. Top left: the ventilation principle; top right: exhaust fan; bottom left: room heating integrated into a docking station; bottom right: slots in the front of the stairs for air transfer from one room to another. Source: M. Bonvin. For a color version of this figure, see: www.iste.co.uk/lachal/ energy.zip

The climatic conditions obtained exactly correspond to the predictions, as can be seen in the measurements taken in the summer period in Figure 13.17, carried out by the HES-SO Valais in 2011, i.e. more than 10 years after the inauguration. The temperature of each room varies only slightly during the day (about 1.5°C, identical to what was observed during our experiment 20 years earlier). There is also a

Audits or Implementation of Knowledge: Transformation of Valère Castle to a Museum

185

stratification from below, where the atmosphere of these spaces does not exceed 23°C, to above, where the hottest spaces can reach and occasionally exceed 26°C. We can also note the very reactive operation of the fans. The relative humidity remains within the desired range (40–60%), except in some areas of the upper ground floor where it can fluctuate between 50 and 70%.

Figure 13.17. Interior climate of the Valère Museum, summer 2011. Top: temperatures and fan operation; bottom: relative humidity (source: M. Bovin). For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

186

Energy Transition

At the time of writing, the concept has expanded to two other buildings in Valère Castle, representing a total of about 40 rooms. Quality monitoring is always carried out by the Canton of Valais, which receives all indoor climate readings every month. The operation of these installations requires continuous monitoring with the possibility of rapid intervention (during the day), as a drift in humidity values may be harmful to the exposed parts. Turning a historic building into a museum is not an easy task, mainly because of the large internal production of heat (lighting and visitors) and humidity (visitors). However, it was possible to maintain acceptable indoor conditions (in terms of temperature and humidity) with minimum energy consumption (electrical power estimated before the concept was defined at 72 kW, and in practice was 45 kW) and without invasive technical installations (reduced investment and maintenance costs). 13.6. Conclusion This CSF highlights three points: – a long time scale has – necessarily – governed this important and original project, the objective of which is ultimately to expose local culture to the public. Energy, along with agriculture and tourism, plays a major role in the Valais landscape: such a project perfectly fits into this context; – the knowledge generated by the CSF is a powerful accelerator of innovation; learning through use is a guarantee of success; – the human aspects are a determining factor. The sustainability of the CSF team is fundamental, as is the importance of very good contact between the teams when transferring responsibility from one stage to another.

14 CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings

The thermal needs of buildings represent about 40% of energy consumption in Europe and are mainly based on fossil fuels. The importance of reducing the thermal consumption of buildings to achieve a successful energy transition is widely recognized. This is a long and complex process: long because it involves, to a large extent, the thermal renovation of existing buildings, an operation that requires several years between decision-making and implementation; complex because energy is only one aspect of these renovation or new construction operations. Financial, spatial planning, social and architectural aspects are also highly relevant. In addition, users have a definite influence on the energy consumption of these buildings, whether it is their occupants or their managers. CSF carried out over the past 30 years has proved to be very instructive and has contributed to the dissemination of good practices. 14.1. Context: from the catalogue of solutions to real practice While there is a wide range of solutions to reduce or even eliminate the use of fossil fuels for heating buildings, the transition to a successful implementation of these solutions over time is not easy: it is necessary to achieve correct operation, good efficiency, user satisfaction and acceptable economic costs simultaneously. It is therefore necessary to treat the implementation of energy transition in the built environment as an important technical and social innovation [REI 11]; it will necessarily follow the classic steps allowing for a gradual transition from prototypes developed by pioneers to labeled and, finally, standardized (and even trivialized) operations. To improve and accelerate this transition, feedback on innovative

Energy Transition, First Edition. Bernard Lachal. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

188

Energy Transition

“objects” built in the traditional construction system CSF is a key tool. This is certainly the most attractive option compared to the two approaches discussed above since there is no normative character; it extends over a “long” time period (often several years – see Chapter 4) and develops in a stimulating framework of risktaking and innovation. Since the mid 1980s, the Centre universitaire d’étude des problèmes de l’énergie1 has been working to conduct numerous CSFs on this type of operation. They concern both new builds and renovation, administrative and residential buildings, heating and air conditioning, external interventions, the technical system. Of the 17 objects studied between 1985 and 2017: – ten concerned residential buildings, six administrative complexes and one building for agricultural use; – by referring to the stages of innovation deployment as described in Part 1, they had a “pioneering” aspect marked at the beginning of the “CSF” activity, then the emphasis was placed on those entering a more experienced stage of “first adopters” to gradually focus interest on more mature interventions (dissemination and appropriation phase, or even standardization); – the median size of these objects is about 10,000 m2 (typically about a hundred dwellings), we are dealing with real operations and not one-off experiments; – the duration of their CSFs is several years, with the median being five years: these are indeed long-term studies, corresponding to the dynamics of these real estate operations and the specific response time of these systems, which is over a year; – the results of all these CSFs have been published in dedicated reports, in professional or scientific publications, and some of them are the practical basis for a thesis work and many student works at master’s level. There is little work of this kind in the scientific world, as noted by J. Khoury in his doctoral thesis [KHO 14]: “The literature review indicates that there are few existing publications providing feedback on renovated buildings. Studies on new operations are more numerous and highlight gaps of around 30 to 150% between planned and actual performance [4 references].” It is surprising that, despite its importance, in situ assessment is so neglected; here are some elements of answers that will be explored in more detail in the next section. First of all, CSF is a difficult, thankless, time-consuming and apparently 1 Later renamed the “Energy Systems” group.

CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings

189

expensive exercise, even though the investment for evaluation is minimal compared to the financial support provided by the community for the same innovative technologies. This exercise is little known, little valued and difficult to exploit. One of the aims of this book is to break this vicious cycle, to highlight its interest and to try to scientifically organize the acquired experience. In general, in relation to the tools presented in the previous section, the following remarks can be made. First of all, it is an exercise requiring a long-term commitment, the human aspects of which are of primary importance. First, the project is typically set up over a year, from the first contacts to the signing of contracts, which generally launch the CSF process. Then, this process itself generally lasts several years and, overall, taking publication and evaluation of the results into account, it can take up to nearly ten years, and, more often than not, takes at least five. The practice of setting up and leading a follow-up group for each project has become widespread since the CSF of PLO in 1998 [BRA 02], gradually increasing in importance, and no fewer than 20 people constituted the most recent follow-up groups, with semi-annual meetings. There are also, as a rule, multiple sources of financing, which has advantages in terms of the impact of CSF but complicates its management. The basic tool for this type of CSF is the Sankey diagram with its demonstrative power; it is present as a rule, and often illustrates the conclusions of the evaluation. It has gradually improved with the team’s experience, and the existence of dedicated software has allowed a qualitative leap forward. For taking measurements, the exclusive use of equipment developed by a laboratory which is based on current practices in experimental physics, which was the rule in the early 1980s, has since been replaced by the robust and reliable professional data-logger. In parallel to this, the integration of more frequent use of data from the control system has increased thanks to improvements in their quality. The extraordinary development of IT has not radically transformed this aspect, although it has made this essential part of CSF much more comfortable. The analysis, modeling, simulation and synthesis sections have been improved with accumulated experience, with each feedback contributing to the building. What has changed most, in fact, is the scope of the study – mainly the spatial scope since the networks supplying these buildings are increasingly taken into account – and the scope of the aspects studied. Thus, an economic evaluation is systematically carried out if one is in the diffusion–appropriation stage of innovation, just as sociological evaluation – which is what is being enacted in a CSF – can make a significant contribution when dealing with highly use-sensitive systems, such as high energy performance buildings or systems with a high level of spontaneously variable resources.

190

Energy Transition

In summary, it can be said that CSF operations have gradually become more professional over the past 30 years. Today, they tend to be increasingly interested in operations that broaden the diffusion of already well-identified solutions and therefore tend to naturally extend towards the finer analysis of usage. Four major findings could be made, which will be detailed later: – increased complexity of construction and technical systems, which is wellhighlighted by the Sankey diagram; – the importance of use and human aspects, which is not easy to quantify; – the problem known as the “performance gap” between expected normalized performance and ex post performance, which can be understood through modeling; – a surprising invariant in the thermal functioning of the “building” system, which demonstrates the relevance of input/output relationships. 14.2. Increased complexity of construction techniques well-highlighted by the Sankey diagram

and

systems

To reduce the thermal energy consumption from fossil fuels of a building – and excluding direct electric heating – two main levers can be used: either reducing demand or improving the energy production required (Figure 14.1).

Figure 14.1. Solutions to reduce a building’s thermal consumption

CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings

191

Concerning a decrease in demand, it may involve use of the building by the inhabitants: lowering the indoor temperature, better management of the windows and blinds and reducing the consumption of domestic hot water, or the operation of the building itself: improving the balance of the exterior by strengthening insulation or better capture of incident solar energy, controlling the ventilation rate, recovering calories released by stale air, reducing the consumption of domestic hot water through the use of suitable components or recovering calories released into wastewater. To improve the production of the energy thus reduced, it is possible to increase production efficiency (better boiler, flue gas condensation, lower supply temperatures for heating and domestic hot water systems) or to replace fossil energy with renewable energy. All of these solutions can be combined to achieve the objective of a building that is little or not at all dependent on fossil fuels. By complicating the “building” energy system, they generate difficulties such as the risk of summer overheating in highly insulated buildings (heat that enters easily but escapes with difficulty), physics problems such as condensation if ventilation is not properly provided, the gray energy of the materials and components involved and the operating electricity of the various integrated components (fans, circulators, heat pumps, computers, etc.). In addition, they require more careful design, implementation and operation, monitoring and use to avoid disappointing results, including economic results. This intrinsic complexity of low-energy buildings can be demonstrated very effectively with the Sankey diagram, as shown in the following example. This feedback concerns the energy renovation of a residential building dating from the 1960s, located at 40–42, avenue du Gros-Chêne in Onex (see Figure 14.2). The energy objective set was the Minergie label, i.e. an energy consumption for heating and domestic hot water of less than 80 kWh/m2/year after renovation. It was the first renovated complex of this size to be so labeled in Geneva [KHO 14, MER 12]. Some of the building’s characteristics are given below: – year of construction: 1963; – year of renovation: 2008; – energy reference area: 5,357 m2, 63 housing units, 134 inhabitants (2010).

192

Energy Transition

Figure 14.2. Renovated (left) and unrenovated (right) building [KHO 14]

The renovation mainly concerned the following elements: – exterior components: closing of loggia balconies over the entire width of the building and on both façades, replacement of single glazing with double glazing, insulation of the roof and floor as well as the gabled walls and ground floor; 
 – heat production: switch from oil to district heating, solar installation (100 m2 of collectors) for DHW production; 
 – technical installations: installation of a double-flow ventilation system with integrated heat pump, allowing the temperature of the forced air in the apartments to rise above 20°C. The feedback, carried out over two full years (2008–2010), covered both energy and economic aspects. It should be noted that Building 40–42 has a sister building located at 36–38, which has not been renovated. This was included in the monitoring and used as a reference (situation without renovation) for this study. 
 The energy analysis showed a decrease of nearly 300 MJ/m2.year in the IDC of the 40–42 building, from 667 MJ/m2.year before renovation to 392 after renovation (–40%). It should be noted, however, that the change in energy source (from oil to the CADIOM district heating network, based on the incineration of household

CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings

193

waste) accounts for a significant part of this decrease: building 36–38, which was not renovated, saw its IDC lose 70 MJ/m2.year following the switch to CADIOM alone and the fact that it no longer records losses from oil–heat conversion.

Figure 14.3. Sankey diagram of a renovated building (top) and an unrenovated building (bottom) [MER 12]. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

194

Energy Transition

A comparison between the Sankey diagrams of the renovated building (no. 40–42) and the unrenovated building (no. 36–38) clearly shows the complexity introduced by the Minergie-labeled renovation (Figure 14.3). First, the renovated building is now supplied with heat from three sources: heat waste from household waste treatment through the CADIOM network, solar energy from thermal solar collectors on the roof and electricity for the activation of the stale air heat pump and the various active components required for the air supply fan and solar system. On the demand side, improved glazing insulation and the creation of loggias on balconies have saved about 25% of heat loss by diffusion through the exterior. New components are emerging to substitute energy or recover heat losses from the building. Thus, out of the 96 MJ/m2.year of heat losses present in the extracted air before renovation, nearly 80% is recovered thanks to the double-flow ventilation system with heat pump. These systems are complex to install through renovation because it is necessary to install ducts in the building from the top floor to supply air from outside, which is preheated and then reheated in all living rooms. This complexity is only slightly visible in the Sankey diagram presented in this way. The following example illustrates the case where this diagram demonstrates the complexity of the installations better than the physical representation of the systems. This is the CSF carried out on the Solar City of PLO [BRA 02]. This building is a “pre-Minergie” pioneer since beginning in 1990, it set its consumption target as the limit given by the first version of the label (160 MJ/m2.year), while being designed about 10 years before its launch. Under the impetus of the project owner, the municipality of Plan-les-Ouates, the Solar City was conceived as a project that respects the environment and limits consumption of non-renewable energy. This objective could only be achieved by making the best use of renewable energies: 1,400 m2 of solar roofing, the presence of a Canadian well, heat recovery from stale air and optimization of free energy supplies. From the very beginning of the building’s design, it proved essential to bring together a multidisciplinary group capable of mastering all areas. This task was assigned to an interactive team, consisting of the project owner, the architect, the civil engineer, engineers and academic experts, specialists in eco-biology, geotechnical and acoustic engineering. The originality of this company and the novelty of the technologies required a detailed analysis of the overall thermal functioning of the whole system and a technical and economic evaluation of the real contributions of the various subsystems. To this end, the contracting authority, the Cantonal Office of Energy (OCEN) and the Federal Office of Energy (SFOE) commissioned the “Energy Systems” group to prepare a report on the performance of the various energy-saving installations that have been achieved. For the ventilation part, there are no less than five systems in series for the three distributions present in both of the boiler rooms, leading to excellent performance (see Figure 14.4). First, the presence of a system of underground pipes for

CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings

195

preheating fresh air in winter and cooling in summer, located under the collective garage; second, double flow ventilation (fresh air supply in rooms such as the living room and bedrooms, return of stale air in rooms such as the kitchen, bathroom and toilet, one with heat recovery on extracted air); then, a post-heating of the air at the outlet of the heat exchanger to ensure a temperature that will not disturb the occupants; and finally, distribution of fresh air behind the radiators, after passing through the slab. The extracted air is finally blown into the garage for final use.

Figure 14.4. Ventilation system, PLO

These five systems are largely redundant and, in addition to the over-investment they represent, they have made commissioning, monitoring and maintenance difficult in the long term, especially if six monoblocks are included in four different locations. Measurement was complete as it included, for each of the three distributions of one of the boiler rooms measured in detail, a supply air velocity and an exhaust air velocity, the Canadian well outlet temperatures, the recovery exchanger outlet temperatures, the post-heating outlet temperatures (supply side), the return room temperatures (exhaust side) at the heat exchanger inlet and the heat

196

Energy Transition

dissipated in the post-heating batteries (water side). Seasonal variations in the main temperatures of the air as it passes through the ventilation system are shown in Figure 14.5.

Figure 14.5. Average daily air temperatures at different points of the ventilation system of the dwellings. For a color version of this figure, see: www.iste.co.uk/lachal/energy.zip

The estimation of energies requires knowledge of the air flows, which have been carefully measured [BRA 02] and which require the following remarks. Part of the dwellings (3–4–5) are properly ventilated, with an average controlled air exchange rate of 0.46 volumes per hour. The apartments are slightly over-pressurized (more pulsed flow than extracted), which limits the infiltration of parasitic cold air. The controlled ventilation of the other part of the dwellings (6–7) is minimal (0.33 volume per hour in average value for all dwellings), but apparently sufficient insofar as we are not aware of any complaints on the subject. It should be noted that the supply air flow rate is low (half of the extraction), which has resulted in two unfortunate consequences: putting the apartments into depression and consequent parasitic air intakes, as well as the absence of any energy recovery via the Canadian well and the heat exchanger for half of the air renewal. It is interesting to compare the two situations (Sankey diagram, Figures 14.6 and 14.7). In properly ventilated dwellings, air enters the Canadian well from outside at an average temperature of 6.6°C and exits at 10.5°C, corresponding to a gain of 100,000 MJ (24.7 MJ/m2), the heat recovery unit on the extracted air yields an additional 137,400 MJ (33.9 MJ/m2), with air at 15.7°C. The post-heating (two-thirds) and the fan (one-third) raise the air temperature to 18.1°C through 69,500 MJ (17 MJ/m2); the fan alone would have raised the air temperature to 16.6°C, a sufficient supply air temperature not to be uncomfortable. The average temperature of the dwellings being 22.9°C, the remaining heat (118,250 MJ or

CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings

197

29.2 MJ/m2) is provided by the radiators, partly supplied by boiler room 1 and partly by boiler room 2. The system extracts most of the forced air (87%) and directs it to the heat recovery unit. Finally, the air, with a temperature of 16.8°C, ends its cycle in the garage, where it finally releases the remaining 232,000 MJ. About half of this energy will be taken up by the Canadian well, the other half being used to temper this space.

Figure 14.6. Sankey diagram of energy flows of ventilation air, buildings 3–4–5

For dwellings with a lack of ventilation, the energy diagram shown in Figure 14.7 is complex in appearance, corresponding to the complexity of the systems and indicates the difficulty we had in drawing a coherent analysis. It is similar to the previous scheme (ventilation of dwellings 3–4–5), but doubled by mirror symmetry in the post-heating energy flow. Compared to the previous analysis of the ventilation flows of dwellings 3–4–5, it should be noted that only half of the

198

Energy Transition

ventilation air in the dwellings 6–7 is treated by the systems integrated into the supply air, including the recuperators. This corresponds to a shortfall of 33,600 MJ (14 MJ/m2), which is largely recovered by the low ventilation rate. The ventilation system is certainly the least satisfactory part of the whole. One of the reasons for these dysfunctions is the excessive complexity of the ventilation concept. This example has not been reproduced: this pioneering group, including the CSF that accompanied it, has played its role perfectly.

Figure 14.7. Energy flow diagram of ventilation air, buildings 6–7

CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings

199

14.3. The importance of use and human aspects that are difficult to quantify The example presented here is a residential building built in 1963 and renovated in 1990 [LAC 92]. Between 1960 and 1970, the population of the canton of Geneva almost doubled, creating a strong need for apartments and leading to the creation of several satellite cities. Onex 1 consists of 750 apartments built in 1962 with prefabricated concrete elements. Built in a very short period of time, these nine-storey buildings are poorly insulated, their main facades are highly glazed and include many thermal bridges. After 25 years, these buildings are in poor condition and need to be renovated. One of them, comprising 126 homes that do not cross over and are oriented east–west, has been the subject of feedback. The insulation of the exterior components has been greatly improved and the windows replaced. Balconies have been added for both north- and south-facing apartments. The building is connected to a district heating system powered by heat from the household waste incineration plant; it includes two heat distribution systems to distinguish the demands of the northern and southern zones. Thermostatic valves were installed on the radiators. The building was measured two years before and two years after its renovation. Consumption after renovation has been reduced by 40%. Five data points were measured: heat demand for each of the two zones (north and south), outside temperature, solar irradiation and inside temperature of an unoccupied dwelling in the south zone. The data is read 20 times an hour by a standard data collection system and stored on an hourly basis. Only the part concerning the quantification of the thermal effects of occupant use on solar gains will be developed here. At the same time, in order to study this aspect, two series of direct observations were made by photographing the south facade, using a programmable camera placed on the roof of the neighboring building. The shooting frequency was two hours over two one-week periods in November and March. These shots provided a better understanding of how occupants used window openings and exterior blinds (76 observations for 212 windows and 160 residents). Figures 14.8 and 14.9 detail the method. For each window, we estimate the proportion of its surface area blocked by the external blind Rs (in 10% steps), left open (0, 10%, 25%, 33%, etc.) and covered with internal curtains Rr (in 10% steps). All this information is reported on a spreadsheet, which makes it possible to calculate for each window a transmission coefficient Tfint: = (1 −

) ∗ (1 − 0.4 ∗

)

[14.1]

where 0.4 is the average reflection coefficient of curtains, measured in the laboratory.

200

Energy Transition

Figure 14.8. Photograph of the facade

Figure 14.9. Analysis of the photography by spreadsheet

By taking the transmission factor of single glazing (0.8) into account, the total surface area of the glazing composing the south façade, the loss of shade from nearby trees, flowerpots, barriers (15%), we can obtain for each photograph of the façade the corresponding solar collection surface. The contribution of the staircase is

CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings

201

difficult to estimate. An overall transmission of 60% was considered with a utility factor of one-third to account for the fact that this space is not heated, resulting in a contribution to the constant effective catchment area of 46 m2. The total absolute value of the catchment is estimated with an accuracy of about 20%, but it is very sensitive to any change in use by the occupants. The values obtained for the effective collection area and the proportion of open windows both vary according to the interior temperature of the unoccupied apartment (see Figure 14.10).

Figure 14.10. Variation of the effective collection area (top) and the rate of open windows (bottom) as a function of temperature inside an unoccupied apartment

In a residential building, there is no direct effect of the level of sunlight on the closing of blinds, but an indirect one via the indoor temperature, which rises with

202

Energy Transition

solar gains and the outdoor temperature. To respond to this discomfort, the current occupants have no choice but to close the external blinds and/or open the windows. It should be noted that approximately 30% of blinds are closed in non-sunny weather. The habitat of the occupants in a comfortable situation – when the indoor temperature is below 23°C – is to open 2% of the window area. Similarly, interior curtains are an important factor in reducing solar gains. Due to the use of photography in the visible range, it was not possible to quantify night-time opening of the windows. In absolute terms, the effective catchment area is about 300 m2, or about 3% of the energy reference area. Using the same approach as that used to model the interior temperature of the Aymon building (see Chapter 13 and the detailed description in [LAC 92]), this value was confirmed (280 m2), as was the decrease in the effective collection area and/or the opening of windows with the increase in the interior temperature. This method was used in 2016 in a CSF on a building renovated under the Minergie-P label (see [KHO 18]), with comparable results. In his thesis on district heating networks, Quiquerez [QUI 17] estimated this solar collector area on a larger scale (3 million m2, or 15% of Geneva’s residential park), by linking the heat demand of a large district heating network to sunlight. In order to avoid the influence of external temperature and morning and evening restart peaks, the average hourly amounts of power drawn down between 10 am and 3 pm for an external temperature of 7°C (+/1°C) have been related to overall solar radiation on a vertical plane determined by averaging the measured values on the north, south, west and east vertical planes (Figure 14.11). The fall in demand is significant since good sunshine saves 20 MW of heat, or 20% of the power generally used at this temperature and time of day.

Figure 14.11. Relationship between CAD request and sunlight [QUI 17]

CSF to Evaluate and Improve the Appropriation of Innovation: the Case of Buildings

203

Taking the total heated area of 3 million m2 into account, the estimated 38,000 m2 represents about 1.2% of the heated area. Considering only the points with low sunlight (150

Equivalent surface area (m )

150,000

300,000

750,000

>1,500,000

Number of people

3,000

6,000

15,000

>30,000

2

Geothermal energy (GWh)

15

26

36

44

Number of hours

3,000

5,100

7,200

>8,000

Contribution

100%

86%

48%