A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040 3658421673, 9783658421670

Table of contents :
Preface
Table of Contents
List of Figures
List of Tables
List of Abbreviations
List of Symbols
Abstract
Kurzfassung
1 Introduction
2 Modelling Methods
2.1 Internal Combustion Engines
2.2 Electric Drive Systems
2.2.1 Permanent Magnet Synchronous Motor
2.2.2 Induction Motor
2.2.3 Power Electronics
2.3 Battery Cells
2.4 Fuel Cell System
2.4.1 Fuel Cell Stack
2.4.2 Air Supply
2.4.3 Fuel Supply
2.4.4 Electrical Functionalities and Control
2.4.5 Implementation in Simulation
2.5 Transmission
2.5.1 Automated gearboxes
2.5.2 Manual gearboxes
2.5.3 Differential gearboxes
2.5.4 Electronically Controlled Multi-Plate Clutch
3 Development in Powertrain Technology
3.1 Internal Combustion Engines
3.1.1 Gasoline Engine (High Efficiency Concept)
3.1.2 Gasoline Engine (Budget Optimized Concept)
3.1.3 Gasoline Engine (Range Extender Concept)
3.1.4 Natural Gas Engine
3.1.5 Diesel Engine
3.2 Electric Drive Systems
3.2.1 Permanent Magnet Synchronous Motor
3.2.2 Induction Motor
3.2.3 Power Electronics
3.3 Battery Systems
3.3.1 High Power Battery Cells
3.3.2 Medium Power Battery Cells
3.3.3 High Energy Battery Cells
3.4 Fuel Cell Systems
3.5 Transmissions
3.6 Tank and Charging Systems
4 Powertrain Simulation
4.1 Vehicle Models
4.2 Powertrain Component Models
4.3 Powertrain Design
4.4 Drive Cycles
4.5 Operating Strategies
4.5.1 Consumption Minimization Strategy
4.5.2 Equivalent Consumption Minimization Strategy for parallel hybrid powertrains
4.5.3 Equivalent Consumption Minimization Strategy for serial hybrid powertrains
4.5.4 Selective Equivalent Consumption Minimization Strategy for serial/parallel hybrid powertrains
4.5.5 Additional Control Strategy Parameters
5 Results of Powertrain Simulation
5.1 Sedan
5.2 Sport Utility Vehicle
5.3 Light-Duty Vehicle
5.4 Conclusion
6 Summary and Conclusion
Bibliography
Appendix
A1. Simulation Data Sheets
A2. Drive Cycle Data Sheets

Citation preview

Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart

Tobias Stoll

A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040

Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart Series Editors Michael Bargende, Stuttgart, Germany Hans-Christian Reuss, Stuttgart, Germany Jochen Wiedemann, Stuttgart, Germany

Das Institut für Fahrzeugtechnik Stuttgart (IFS) an der Universität Stuttgart erforscht, entwickelt, appliziert und erprobt, in enger Zusammenarbeit mit der Industrie, Elemente bzw. Technologien aus dem Bereich moderner Fahrzeugkonzepte. Das Institut gliedert sich in die drei Bereiche Kraftfahrwesen, Fahrzeugantriebe und Kraftfahrzeug-Mechatronik. Aufgabe dieser Bereiche ist die Ausarbeitung des Themengebietes im Prüfstandsbetrieb, in Theorie und Simulation. Schwerpunkte des Kraftfahrwesens sind hierbei die Aerodynamik, Akustik (NVH), Fahrdynamik und Fahrermodellierung, Leichtbau, Sicherheit, Kraftübertragung sowie Energie und Thermomanagement – auch in Verbindung mit hybriden und batterieelektrischen Fahrzeugkonzepten. Der Bereich Fahrzeugantriebe widmet sich den Themen Brennverfahrensentwicklung einschließlich Regelungs- und Steuerungskonzeptionen bei zugleich minimierten Emissionen, komplexe Abgasnachbehandlung, Aufladesysteme und -strategien, Hybridsysteme und Betriebsstrategien sowie mechanisch-akustischen Fragestellungen. Themen der Kraftfahrzeug-Mechatronik sind die Antriebsstrangregelung/Hybride, Elektromobilität, Bordnetz und Energiemanagement, Funktions- und Softwareentwicklung sowie Test und Diagnose. Die Erfüllung dieser Aufgaben wird prüfstandsseitig neben vielem anderen unterstützt durch 19 Motorenprüfstände, zwei Rollenprüfstände, einen 1:1-Fahrsimulator, einen Antriebsstrangprüfstand, einen Thermowindkanal sowie einen 1:1-Aeroakustikwindkanal. Die wissenschaftliche Reihe „Fahrzeugtechnik Universität Stuttgart“ präsentiert über die am Institut entstandenen Promotionen die hervorragenden Arbeitsergebnisse der Forschungstätigkeiten am IFS. Reihe herausgegeben von Prof. Dr.-Ing. Michael Bargende Lehrstuhl Fahrzeugantriebe Institut für Fahrzeugtechnik Stuttgart Universität Stuttgart Stuttgart, Deutschland Prof. Dr.-Ing. Hans-Christian Reuss Lehrstuhl Kraftfahrzeugmechatronik Institut für Fahrzeugtechnik Stuttgart Universität Stuttgart Stuttgart, Deutschland

Prof. Dr.-Ing. Jochen Wiedemann Lehrstuhl Kraftfahrwesen Institut für Fahrzeugtechnik Stuttgart Universität Stuttgart Stuttgart, Deutschland

Tobias Stoll

A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040

Tobias Stoll IFS, Fakultät 7, Lehrstuhl für Fahrzeugantriebssysteme Universität Stuttgart Stuttgart, Germany Zugl.: Dissertation Universität Stuttgart, 2023 D93

ISSN 2567-0042 ISSN 2567-0352  (electronic) Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart ISBN 978-3-658-42168-7  (eBook) ISBN 978-3-658-42167-0 https://doi.org/10.1007/978-3-658-42168-7 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer Vieweg imprint is published by the registered company Springer Fachmedien Wiesbaden GmbH, part of Springer Nature. The registered company address is: Abraham-Lincoln-Str. 46, 65189 Wiesbaden, Germany

Preface This thesis was written during my work as a scientific employee at the Institute of Automotive Engineering Stuttgart (IFS) in the field of Automotive Powertrain Systems under the supervision of Prof. Dr.-Ing. Michael Bargende. My special thanks go to Prof. Dr.-Ing. Michael Bargende for supervising the thesis and giving me the opportunity to realise it according to my own visions. I would like to thank Univ.-Prof. Dipl.-Ing. Dr.techn. Helmut Eichlseder and Prof. Dr.-Ing. André Casal Kulzer for taking over the co-report and for reviewing the thesis. I would especially like to thank my colleagues, Hans-Jürgen Berner and Viktoria Kelich, for their guidance during the project, Feyyaz Negüs for the engine model provided and Sven Eberts for providing basic data for the simulation models. I would also like to thank the FVV (The Research Association for Combustion Engines eV) for financing the project. In addition, I want to thank Dr.-Ing. Thorsten Schnorbus for supervising the project on the part of the FVV and for the constant professional discourse. Last but not least I want to thank my family and friends for their support during the writing process and their will to submit constant feedback to this thesis.

Stuttgart

Tobias Stoll

Table of Contents Preface ............................................................................................. V List of Figures ............................................................................... XI List of Tables.............................................................................. XXI List of Abbreviations................................................................ XXIII List of Symbols ......................................................................... XXV Abstract ....................................................................................XXXI Kurzfassung........................................................................... XXXIX 1 Introduction ............................................................................. 1 2 Modelling Methods .................................................................. 5 2.1 Internal Combustion Engines ......................................................... 6 2.2 Electric Drive Systems ................................................................. 18 2.2.1

Permanent Magnet Synchronous Motor .......................... 18

2.2.2

Induction Motor ............................................................... 29

2.2.3

Power Electronics ............................................................ 38

2.3 Battery Cells ................................................................................. 46 2.4 Fuel Cell System .......................................................................... 51 2.4.1

Fuel Cell Stack ................................................................ 53

2.4.2

Air Supply ....................................................................... 55

2.4.3

Fuel Supply ...................................................................... 57

2.4.4

Electrical Functionalities and Control ............................. 59

2.4.5

Implementation in Simulation ......................................... 60

2.5 Transmission ................................................................................ 61 2.5.1

Automated gearboxes ...................................................... 61

2.5.2

Manual gearboxes ............................................................ 69

Table of Contents

VIII 2.5.3

Differential gearboxes ..................................................... 73

2.5.4

Electronically Controlled Multi-Plate Clutch .................. 76

3 Development in Powertrain Technology ............................. 79 3.1 Internal Combustion Engines ....................................................... 79 3.1.1

Gasoline Engine (High Efficiency Concept) ................... 80

3.1.2

Gasoline Engine (Budget Optimized Concept) ............... 82

3.1.3

Gasoline Engine (Range Extender Concept) ................... 84

3.1.4

Natural Gas Engine.......................................................... 85

3.1.5

Diesel Engine................................................................... 86

3.2 Electric Drive Systems ................................................................. 88 3.2.1

Permanent Magnet Synchronous Motor .......................... 89

3.2.2

Induction Motor ............................................................... 90

3.2.3

Power Electronics ............................................................ 91

3.3 Battery Systems ............................................................................ 92 3.3.1

High Power Battery Cells ................................................ 93

3.3.2

Medium Power Battery Cells .......................................... 94

3.3.3

High Energy Battery Cells............................................... 95

3.4 Fuel Cell Systems ......................................................................... 96 3.5 Transmissions ............................................................................... 98 3.6 Tank and Charging Systems ....................................................... 100

4 Powertrain Simulation ........................................................ 103 4.1 Vehicle Models........................................................................... 105 4.2 Powertrain Component Models .................................................. 105 4.3 Powertrain Design ...................................................................... 110 4.4 Drive Cycles ............................................................................... 129 4.5 Operating Strategies ................................................................... 135 4.5.1

Consumption Minimization Strategy ............................ 136

Table of Contents

IX

4.5.2

Equivalent Consumption Minimization Strategy for parallel hybrid powertrains ............................................ 137

4.5.3

Equivalent Consumption Minimization Strategy for serial hybrid powertrains ............................................... 139

4.5.4

Selective Equivalent Consumption Minimization Strategy for serial/parallel hybrid powertrains .............. 140

4.5.5

Additional Control Strategy Parameters ........................ 140

5 Results of Powertrain Simulation ...................................... 143 5.1 Sedan .......................................................................................... 143 5.2 Sport Utility Vehicle .................................................................. 153 5.3 Light-Duty Vehicle..................................................................... 168 5.4 Conclusion .................................................................................. 178

6 Summary and Conclusion ................................................... 183 Bibliography .................................................................................187 Appendix ......................................................................................195 A1. Simulation Data Sheets...............................................................195 A2. Drive Cycle Data Sheets.............................................................199

List of Figures Figure 1.1: 

CO2-emissions of newly registered passenger cars: Historical trend line [5], EU-27 reduction targets [2] [3] [4]. ...............................................................................2

Figure 2.1: 

Model development and simulation process. ...................5

Figure 2.2: 

Factors for scaling of friction mean effective pressure over the cylinder displacement (top), stroke-to-bore ratio (mid) and the number of cylinders (bottom) in the GT-Power models. .................10

Figure 2.3:

Scaling functions of the characteristic map model for an example Otto engine for the influence of change in engine displacement on the specific injected fuel mass (top), for the change in engine displacement on the friction mean effective pressure (mid) and for the change in engine temperature on the friction mean effective pressure (bottom). ...............14

Figure 2.4: 

Heat transfer coefficient for the outer surface of the catalytic converters against the environment, dependent on vehicle speed and installation position of the catalytic converts, implemented for the static characteristic model of the internal combustion engines. ...........................................................................17

Figure 2.5: 

Heat transfer coefficient for the inner surface of the catalytic converters against the exhaust gas, dependent on the rotational speed of the engine, implemented for the static characteristic model of the internal combustion engines. ....................................17

Figure 2.6:

Equivalent circuit model of the permanent magnet synchronous motor in the F2 coordinate system, with d-axis (top) and q-axis (bottom). ............................20

Figure 2.7: 

Schematic sketch of a permanent magnet synchronous motor with one pole pair (ɘ ʹൌ ɘ‡…ŠǤ), showing important motor parameters in the

XII

List of Figures F2 rotor fixed coordinate system in steady state motor operation (ɘ ʹൌɘͳሻǤ ..........................................21

Figure 2.8: 

Signal flow chart of torque controlled permanent magnet synchronous motor (PMSM) electric drive system with field-oriented control. .................................24

Figure 2.9:

Schematic display of the different control areas of the PMSM motor map (1. quadrant)...............................25

Figure 2.10:

Efficiency characteristic of the synthetic reference permanent magnet synchronous motor with 85 kW rated power. ....................................................................26

Figure 2.11:

Scaling function for the characteristic efficiency map model as a function of the nominal power of the permanent magnet synchronous motor (IE5efficiency class), referred to the 85 kW reference motor. .............................................................................28

Figure 2.12: 

Equivalent circuit model of the induction motor in the K coordinate system. ................................................31

Figure 2.13:

Schematic torque/slip characteristic of an induction motor. .............................................................................31

Figure 2.14:

Efficiency characteristic of the synthetic reference induction motor with 150 kW rated power.....................35

Figure 2.15:

Scaling function for the characteristic efficiency map model as a function of the nominal power of the induction motor (IE3-efficiency class), referred to the 150 kW reference motor. ......................................37

Figure 2.16:

Schematical sketch of a B6c-bridge inverter with battery connection on the left and motor connection on the right......................................................................38

Figure 2.17:

Schematic distribution of conduction losses between transistors and diodes, example for one phase. ..............39

Figure 2.18:

Plot and fit of the transistor resistance for silicon carbide metal-oxide-semiconductor field-effect transistors from Infineon and Semikron, for two different junction temperatures.......................................40

List of Figures

XIII

Figure 2.19:

Results for the parameter fitting of the linear switching loss model for silicon carbide metaloxide-semiconductor field-effect transistors in dependency of the modules maximum drain current, with transistor on-switching losses (top), transistor off-switching losses (mid) and diode reverse recovery losses (bottom). ...............................................43

Figure 2.20:

Switching losses for transistor (E_on, E_off) and diode (E_rr) for a 400 A example module over the current drain current of transistor ͳǡǡ”Ǥ and diode ͳǡǡ‹Ǥ. ...........................................................................44

Figure 2.21: 

Efficiency characteristic of the synthetic reference power electronics module fitted to the 85 kW permanent magnet synchronous reference motor. ..........45

Figure 2.22:

Scaling function for the characteristic efficiency map model of the power electronics module, referred to a 400 A reference module and fitted for a permanent magnet synchronous motor. ..........................45

Figure 2.23:

Open circuit voltage for anode materials (top) and cathode materials (bottom). ............................................47

Figure 2.24:

Relative capacity of a Li(Mn0.5Fe0.5)PO4-cathode over the battery cells C-rate............................................49

Figure 2.25:

Voltage and current characteristics of the medium power battery cell using a C-anode and a Li(Mn0.5Fe0.5)PO4-cathode. ............................................51

Figure 2.27:

Model structure of the fuel cell with sub-models. ..........53

Figure 2.28:

Polarization curve for a 2040 fuel cell design. ...............54

Figure 2.29:

Efficiency maps of compressor (top) and turbine (bottom) for two operating points of the 60 kW fuel cell stack. ........................................................................56

Figure 2.30:

Closed loop control of the fuel cell system. ...................59

Figure 2.31:

Overall fuel cell system efficiency referred to the lower heating value of hydrogen ....................................61

XIV

List of Figures

Figure 2.32: 

Single direction gear ratio dependent efficiency factor   for five different gearboxes, plotted over the single direction gear number  and the resulting quartic polynomial fit. .....................................63

Figure 2.33:

Fit of the synthetic gearbox efficiency Ʉ ǡ•›Ǥ (FIT) compared to real gear meshing efficiency Ʉ  (REAL) [51] [50] for five different automated gearboxes. .......................................................................65

Figure 2.34:

Temperature-dependent reference drag torque  ǡ†”ƒ‰ǡ”‡ˆǤ for automated gearboxes. ..............................66

Figure 2.35:

Temperature-dependent reference additional loss efficiency Ʉ ǡƒ††Ǥǡ”‡ˆǤ for automated gearboxes. ..............66

Figure 2.36:

Frequency and cumulative frequency of the deviation between the automated gearbox model and measurement data used for fitting of the model, with   = 40 °C (top) and   = 60 °C (bottom). .................67

Figure 2.37:

Gearbox efficiency for a synthetic automated gearbox at 200 Nm input torque and 2500 min-1 input speed for different gearbox temperatures over the gear number. .............................................................68

Figure 2.38:

Temperature-dependent reference drag torque  ǡ†”ƒ‰ǡ”‡ˆǤ for manual gearboxes. ...................................70

Figure 2.39:

Frequency and cumulative frequency of the deviation between the manual gearbox model and measurement data used for fitting of the model, with   = 30 °C (top) and   = 70 °C (bottom). .................71

Figure 2.40:

Gearbox efficiency for a synthetic manual gearbox at 200 Nm input torque and 2500 min-1 input speed for different gearbox temperatures. ................................72

Figure 2.41:

Temperature-dependent reference drag torque  ǡ†”ƒ‰ǡ”‡ˆǤ for bevel differential gearboxes. ...................74

Figure 2.42:

Temperature-dependent reference drag torque  ǡ†”ƒ‰ǡ”‡ˆǤ for spur differential gearboxes. .....................74

List of Figures

XV

Figure 2.43:

Gear ratio dependent synthetic gear meshing efficiency Ʉ ǡ•›Ǥ for spur differential gearboxes. ..........75

Figure 2.44:

Gearbox efficiency for bevel and spur differential gearboxes for different gear ratios ݅ at 400 Nm of gearbox input torque and 1000 min-1 of gearbox input speed. .....................................................................75

Figure 2.45:

Efficiency of the ECMC over the input torque. .............77

Figure 3.1: 

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the high efficiency gasoline engine. .............................................81

Figure 3.2: 

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the budget optimized gasoline engine. .............................................83

Figure 3.3: 

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the range extender gasoline engine. ...............................................84

Figure 3.4:

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the natural gas engine. ............................................................................85

Figure 3.5: 

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the diesel engine. ............................................................................87

Figure 3.6: 

Schematic cross section of the permanent magnet synchronous motor with reluctance rotor (left) and efficiency map of reference motor without inverter (right). .............................................................................90

Figure 3.7: 

Schematic cross section of the induction motor with squirrel cage rotor (left) and efficiency map of reference motor without inverter (right). ........................91

Figure 3.8:

Layout of the power electronics B6c-module (left) and efficiency map of the power electronics module for the permanent magnet synchronous reference motor (right). ..................................................................92

Figure 3.9: 

Voltage and Current characteristic of the high power battery cells in different States-of-Charge (SOC). .........94

XVI

List of Figures

Figure 3.10: 

Voltage and Current characteristic of the medium energy battery cells in different States-of-Charge (SOC)..............................................................................95

Figure 3.11: 

Voltage and Current characteristic of the high energy battery in different States-of-Charge (SOC). ......96

Figure 3.12: 

Fitted voltage characteristic of the fuel cell. ..................97

Figure 3.13: 

Air and fuel path (left) and overall system efficiency (right) of the fuel cell system. ........................................98

Figure 3.14: 

Electric charging system between grid and vehicle battery with energy accounting limit (1). .....................100

Figure 3.15: 

Charging efficiency (left) and charging time (right) over battery size for the high energy battery using fast charging. ................................................................102

Figure 3.16: 

Charging efficiency (left) and charging time (right) over battery size for the medium power battery using normal charging. .................................................102

Figure 4.1: 

State of charge over time comparison of dynamic programming (DP) and optimized Multi-Objective Equivalent Consumption Minimization Strategy (opt. MO-ECMS) algorithm on an example hybrid electric vehicle. .............................................................104

Figure 4.2: 

P0-hybrid electric powertrain architecture. ..................111

Figure 4.3: 

P2-hybrid electric powertrain architecture for a 48 V battery system, for front-wheel drive (top) and allwheel drive (bottom). ...................................................113

Figure 4.4: 

P2-hybrid electric powertrain architecture for a 400 V battery system, for front-wheel drive (top) and allwheel drive (bottom). ...................................................115

Figure 4.5: 

P2-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility, for front-wheel drive (top) and all-wheel drive (bottom). .......................................................................117

Figure 4.6: 

Serial-hybrid electric powertrain architecture for a 400 V battery system with external recharge

List of Figures

XVII possibility, for two-wheel drive (top) and all-wheel drive (bottom). ..............................................................119

Figure 4.7: 

Serial-parallel-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility, for front-wheel drive (top) and all-wheel drive (bottom). ..............................................122

Table 4.8: 

Simulation variations of the serial-parallel-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility for the light-duty-vehicle. ........................................................124

Figure 4.8: 

Fuel cell electric powertrain architecture for a 400 V battery system with external recharge possibility, for two-wheel drive (top) and all-wheel drive (bottom). ...125

Figure 4.9: 

Full electric powertrain architecture for an 800 V battery system, for front-wheel drive (top) and allwheel drive (bottom). ...................................................127

Figure 4.10:

Driving cycles and test conditions for sedan, sports utility vehicle (SUV) and light-duty vehicle (LDV). ...129

Figure 4.11: 

Speed and altitude of the Primary Driving Cycle with no speed limitation (top), 100 km / h speed limitation (mid) and 80 km / h speed limitation (bottom). .......................................................................131

Figure 4.12: 

Speed and altitude of the City Cycle. ...........................132

Figure 4.13: 

Speed and altitude of the Commuter Cycle. .................132

Figure 4.14: 

Speed and altitude of the Urban Delivery Cycle. .........133

Figure 4.15: 

Speed and altitude of the Maximum Range Cycle with no speed limitation (top), 100 km / h speed limitation (mid) and 80 km / h speed limitation (bottom). .......................................................................134

Figure 4.16:

Scheme of the Consumption Minimization Strategy for three control variables.............................................137

Figure 4.17: 

General approach for the equivalence factor in dependency of the state of charge. ...............................138

XVIII

List of Figures

Figure 4.18:

Equivalence factor functions for different conditions and different powertrains..............................................138

Figure 4.19: 

Scheme of Equivalent Consumption Minimization Strategy for serial/parallel hybrid powertrains. ............140

Figure 5.1: 

Energy consumption (top) and CO2-emissions (bottom) for the sedan driving the Primary Driving Cycle (open). ................................................................144

Figure 5.2: 

Energy consumption (top) and CO2-emissions (bottom) for the sedan driving the Maximum Range Cycle (open). ................................................................146

Figure 5.3: 

Energy consumption (top) and CO2-emissions (bottom) for the sedan driving the City Cycle. .............148

Figure 5.4: 

Energy consumption (top) and CO2-emissions (bottom) for the sedan driving the Commuter Cycle....149

Figure 5.5: 

Impact of different internal combustion engine powertrain architectures on the overall energy consumption of the sedan driving the Primary Driving Cycle (open) (top) and driving the City Cycle (bottom). .............................................................151

Figure 5.7: 

Impact of different measures and powertrain architectures on the electric energy consumption of the sedan driving the City Cycle. .................................153

Figure 5.8: 

Energy consumption (top) and CO2-emissions (bottom) for the SUV driving the Primary Driving Cycle (open). ................................................................155

Figure 5.9: 

Energy consumption (top) and CO2-emissions (bottom) for the SUV driving the Maximum Range Cycle (open). ................................................................157

Figure 5.10: 

Energy consumption (top) and CO2-emissions (bottom) for the SUV driving the City Cycle. ..............158

Figure 5.11: 

Energy consumption (top) and CO2-emissions (bottom) for the SUV driving the Commuter Cycle. ....160

List of Figures

XIX

Figure 5.12: 

Energy consumption (top) and CO2-emissions (bottom) for the SUV + trailer driving the Primary Driving Cycle (100 km / h). .........................................162

Figure 5.13: 

Energy consumption (top) and CO2-emissions (bottom) for the SUV + trailer driving the Max. Range Cycle (100 km / h).............................................163

Figure 5.14: 

Impact of different internal combustion engine powertrain architectures on the overall energy consumption of the SUV driving the Primary Driving Cycle (open) (top), the City Cycle (mid) and SUV + trailer driving the Primary Driving Cycle (100 km / h) (bottom). ..................................................165

Figure 5.15: 

Impact of different internal combustion engine powertrain architectures on the local CO2-emissions of the SUV driving the Primary Driving Cycle (open) (top), the City Cycle (mid) and SUV + trailer driving the Primary Driving Cycle (100 km / h) (bottom). .......................................................................166

Figure 5.16: 

Impact of different measures and powertrain architectures on the electric energy consumption of the SUV driving the City Cycle ...................................167

Figure 5.17: 

Energy consumption (top) and CO2-emissions (bottom) for the LDV driving the Primary Driving Cycle (80 km / h). .........................................................169

Figure 5.18: 

Energy consumption (top) and CO2-emissions (bottom) for the LDV driving the Maximum Range Cycle (80 km / h). .........................................................171

Figure 5.19: 

Energy consumption (top) and CO2-emissions (bottom) for the LDV driving the City Cycle. ..............172

Figure 5.20: 

Energy consumption (top) and CO2-emissions (bottom) for the LDV driving the Urban Delivery Cycle. ............................................................................174

Figure 5.21: 

Impact of different internal combustion engine powertrain architectures on the overall energy consumption of the LDV driving the Primary

XX

List of Figures Driving Cycle (80 km / h) (top) and the City Cycle (bottom). .......................................................................175

Figure 5.22: 

Impact of different internal combustion engine powertrain architectures on the local CO2-emissions of the LDV driving the Primary Driving Cycle (80 km / h) (top) and the City Cycle (bottom). ...................177

Figure 5.23: 

Impact of different measures and powertrain architectures on the electric energy consumption of the LDV driving the City Cycle. ..................................178

List of Tables Table 2.1: 

Estimated mass fraction and specific heat capacity of the materials used in the internal combustion engines. ...........................................................................15

Table 2.2: 

Values for the catalyst model in the internal combustion engines static characteristic model. ............16

Table 2.3:

Data sheet of the synthetic reference permanent magnet synchronous motor with 85 kW rated power. ....27

Table 2.4:

Data sheet of the synthetic reference induction motor with 150 kW rated power.....................................36

Table 2.6:

Fitting parameters for the different battery cell types................................................................................50

Table 2.7:

Stack configuration for an 80 kW and a 60 kW fuel cell stack. ........................................................................54

Table 2.8:

Estimated mass fraction and specific heat capacity of the materials used in the automated gearboxes, with and without torque converter (TQ). ........................69

Table 2.9:

Estimated mass fraction and specific heat capacity of the materials used in the manual and differential gearboxes ........................................................................73

Table 3.1:

Fuels used for the simulation of the internal combustion engines in GT-Power. .................................80

Table 3.2: 

Battery cell definitions and composition for the different powertrains with electric drive systems. ..........93

Table 3.3:

Gear ratios of the different transmission types. ..............99

Table 4.1: 

Model parameters for the different vehicles used in the simulation. ..............................................................106

Table 4.2: 

Simulation variations of the P0-hybrid electric powertrain architecture. ................................................112

Table 4.4: 

Simulation variations of the P2-hybrid electric powertrain architecture for a 400 V battery system. ....116

XXII

List of Tables

Table 4.5: 

Simulation variations of the P2-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility.................................118

Table 4.6: 

Simulation variations of the serial-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility.................................121

Table 4.7: 

Simulation variations of the serial-parallel-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility for sedan and SUV. ......................................................................123

Table 4.9: 

Simulation variations of the fuel cell electric powertrain architecture for a 400 V battery system with external recharge possibility.................................126

Table 4.10: 

Simulation variations of the full electric powertrain architecture for an 800 V battery system. .....................128

Table 5.1: 

CO2-emisson factors for the different used fuels. ........143

Table A.1:

Illustration of powertrain components in simulation data sheets.....................................................................196

List of Abbreviations AC B6c B7 BEV CA CNG CO2 DC DLR DOC DP E10 E-Axle ECMC EU FCEV FKFS GHG GT H2O HEV IGBT IM LDV LNT MEA MOSFET MTPA MTPV N NOx P0 P2

Alternating current Controlled six pulse bridge converter Diesel with up to 7 % fatty acid methyl esters Battery electric vehicle Crank angle Compressed natural gas Carbon Dioxide Direct current Deutsches Zentrum für Luft- und Raumfahrt Diesel oxidation catalyst Design Point Gasoline with 5 % to 10 % bio-ethanol Electric axle drive Electronically Controlled Multi-Plate Clutch European Union Fuel cell electric vehicle Forschungsinstitut für Kraftfahrwesen und Fahrzeugmotoren Stuttgart Greenhouse gas Gamma Technologies Water Hybrid electric vehicle Insulated-gate bipolar transistor Induction motor Light-Duty Vehicle Lean NOX Trap Membrane electrode assembly Metal-oxide-semiconductor field-effect transistor Maximum Torque per Ampere Maximum Torque per Volt Rated/Nominal Point Nitrogen Oxides Parallel-0 configuration Parallel-2 configuration

XXIV PHEV PMSM RDE RMS RMSE S SiC SOC SUV TWC UN WLTC

List of Abbreviations Plug-in hybrid electric vehicle Permanent magnet synchronous motor Real Driving Emissions Root-mean-squared Root-mean-square error Serial configuration Silicon carbide State of charge Sport Utility Vehicle Three-way catalyst United Nations Worldwide harmonized Light Duty Test Cycle

List of Symbols Formula signs

ƒ

Factor

-

„

Factor

-



Capacity

Wh



C-Rate

1/h



Energy

Wh

‡

Energy density

Wh / kg

ˆ

Function

various



Current

A

‹

Gear ratio

-



Moment of inertia

kg m²



Factor

-



Torque

Nm



Mass

kg

ˆ

Mass flow

kg / s



Rotational speed

min-1

’

Pressure

Pa



Power

W



Ohmic resistance

Ω

•

Slip

-



State of charge

-

List of Symbols

XXVI

–

Time

s



Temperature

°C



Voltage

V



Volume

cm³

š

Variable

various

›

Variable

various

œ

Number

-

Ʉ

Efficiency

-

Ԃ

Rotor displacement angle

rad

ɔ

Phase angle

rad

Ȳ

Magnetic flux linkage

Vs

ɘ

Angular speed

rad / s

Indices

ͳ

Stator-related size

ʹ

Rotor-related size

ƒ

Phase a of a 3-phase AC-motor

ƒ…–Ǥ

Activation overvoltage

ƒ††Ǥ

Additional losses

Ǥ

Anode

„

Phase b of a 3-phase AC-motor



Battery



Breakdown point (torque)

List of Symbols

XXVII

…

Phase c of a 3-phase AC-motor

ƒ–Ǥ

Cathode

…‡ŽŽ

Battery cell

…‘

Conduction (losses)

…‘”Ǥ

Corner of motor base speed range

…›ŽǤ

Cylinder

†

d-coordinate of the F2 coordinate system



Drain of power electronics module

‹Ǥ

Diode

‹‡•‡Ž

Diesel engine



Design Point

”ƒ‰

Drag (torque)

†›Ǥ

Dynamic value



Electronically controlled multi-plate clutch

‡††›

Eddy losses

‡ŽǤ

Electric

‡š…Ǥ

Excess losses

ʹ

Rotor-flux-fixed coordinate system



Fuel Cell System

‡

Iron(losses) of electric motors

ˆ‹–

Fitted values

ˆ—‡Ž

Fuel



Gearbox

List of Symbols

XXVIII



Gear meshing

‰”ƒ˜Ǥ

Gravimetric

Š

Displacement of one cylinder for combustion engines and main field for electric motors



Engine displacement

Š›•–Ǥ

Hysteresis losses



Internal combustion engine



Induction motor

‹

Input



Junction of power electronics module



Stator voltage oriented coordinate system

‹Ǥ

Kinetic overvoltage

Ž‘••

Summarised losses

ƒšǤ

Maximum

‡…ŠǤ

Mechanical

‹

Internal/Indicated

‹Ǥ

Minimum

”

Friction



Nominal Point



Open circuit voltage

‘ˆˆ

Off-state

‘ŠǤ

Ohmic losses

‘

On-state

List of Symbols

XXIX

––‘

Otto engine

‘—–

Output

’

Pole pairs



Power electronics

 

Planetary gear sets



Permanent magnet



Permanent Magnet synchronous motor



Pulse-width modulation

“

q-coordinate in the F2 coordinate system

”‡ˆǤ

Reference value



Root-mean-squared value

””

Reverse recovery (losses) of diode

•„

Stroke-to-bore ratio



Single direction

•–ƒ–Ǥ

Static (losses)

•™

Switching (losses)

•›Ǥ

Synthetic (losses)

–ƒ”‰Ǥ

Target value

”Ǥ

Transistor

–”ƒ•Ǥ

Transmission (losses)

Ɂ

Air gap (power)

ɐ

Stray (inductance)

Abstract The goal currently being implemented in the EU Parliament of only allowing climate-neutral vehicles for new registrations on the roads by 2035 is often associated with the conversion of motorized private transport to battery-electric vehicles. However, other vehicle concepts also offer great opportunities to reduce the manufacturers’ fleet consumption to zero, especially if e-fuels are used as an alternative to fossil fuels. To this end, various powertrain technologies are being investigated, comparing the purely battery-electric powertrain in three vehicle types with other possible powertrain architectures. The different powertrain architectures are then evaluated in the context of overall energy consumption and local CO2-emissions. The architectures studied, include hybrid powertrains, fuel cell powertrains and battery electric powertrains. The aim of this research study is to obtain reliable predictions of suitable powertrain architectures for the market in 2040. For this purpose, the simulation tools and the models generated with them are presented in a first chapter. The overall system simulation presented in a later chapter is carried out with the MATLAB software. The modelling of the individual components of the powertrain is carried out with various software tools. After modelling, the generated component models are converted into simplified, steady state models for the overall system simulation. First, the generation of the combustion engine models in GT-Power with the use of the FKFS UserCylinder is described. The modelling of various technologies that are considered ready for the market in 2040 is described. Since these technologies are not all fully integrated in the software, substitute models are implemented that represent the desired effects of the respective technology as accurately as possible. Examples include pre-chamber spark plugs, high-pressure injection and cylinder coatings. Technologies that can be implemented without substitute models include a variable compression ratio, water injection, a high-turbulence concept and exhaust gas recirculation systems. In addition, sub-models for exhaust gas aftertreatment, the thermal behaviour of combustion engines, the mass and rotating inertia of combustion engines are described. Subsequently, the modelling of the electric drive systems in MATLAB is discussed. The electric drive systems each consist of an electric motor and the power electronics. Two types of electrical motors are introduced, a permanent magnet synchronous motor and an induction motor. Both motors are modelled using

XXXII

Abstract

equivalent circuit models and then parameterised with an optimised control strategy. For the optimisation of the machine operation, the Maximum Torque Per Ampere (MTPA) strategy is used for the base speed range and additionally the Maximum Torque Per Volt (MTPV) strategy for the field weakening area. In the field weakening area, the motor is always operated at the maximum voltage limit in order to minimise the phase currents and thus the power losses. In addition, sub-models for the mass and rotating inertia of the electrical machines are presented. A controlled six-pulse bridge circuit (B6c) is used for the power electronics. The modelling of the power electronics is carried out with a power loss model, which takes into account the switching and conduction losses of the power semiconductors. A procedure is presented for parameterising the power loss model using manufacturer data in order to generate a generic model for the power electronics. The model generated in this way enables subsequent scaling of the power electronics on the basis of the maximum currents. The battery cells are modelled in Excel. For this purpose, the losses and occurring effects in the battery cell are subdivided into four different types and then approximated by mathematical functions. The nourished losses are then subtracted from the open-circuit voltage of the corresponding cell. The losses to be mentioned here are the activation overvoltage, the ohmic losses and the kinetic and mass transport losses. Furthermore, kinetic and mass transport effects are taken into account; these represent a limitation of the maximum usable capacity in the modelling and do not cause any losses. In addition, a model for the battery mass is provided. The fuel cell system is modelled in Simulink with subsequent measurement and export to a steady state MATLAB model. The fuel cell system uses an extrapolated cell model for the year 2040. The air supply is implemented with an electrically assisted turbocharger to achieve partial utilisation of the waste heat. The membrane is humidified via water injection. The anode circuit is designed as a closed circuit. The hydrogen is circulated by an ejector pump. To avoid increased nitrogen concentrations in the anode circuit, the anode circuit is additionally equipped with a purge valve. The voltage level is adjusted by a boost and buck converter. In addition, the model contains a sub-model for the mass of the fuel cell system. Finally, the different transmissions are modelled in Excel. For this purpose, the losses are divided into three loss types for automatic transmissions and two for the remaining transmissions. The loss types are gearing meshing losses, torque losses and additional losses. Generic gearboxes are generated from the available measurement data, which are scaled depending on the input torque, the number of gears, the gear ratios and the number of gear sets.

Abstract

XXXIII

In the following chapter, the expected technology development up to the year 2040 is forecast and implemented in the models already presented. For the combustion engines, five generic engine types are generated, which can be scaled via the mechanical output power. All engines feature direct injection. The natural gas engine is an exception. The first engine is a highly efficient gasoline engine with a comprehensive technology package, which is defined as a "high efficiency concept". This engine has a high specific output power as well as strong de-throttling in naturally aspirated operation and very short combustion durations. For de-throttling in naturally aspirated operation, a high-pressure exhaust gas recirculation system and a variable valve train with cylinder deactivation are provided. The technology package for implementing the short combustion times includes a pre-chamber spark plug, high-pressure injection, a high-turbulence concept and a tumble flap. To further increase efficiency, a variable compression ratio and a variable geometry turbocharger are used. The concept is rounded off with a Euro-7-compliant exhaust aftertreatment system with electrically heated three-way catalytic converter and particulate filter. The second engine concept is an engine with a reduced technology package to reduce engine costs. The engine shows medium specific power and is referred to as the "budget optimized concept". To reduce the combustion time, the same technology package is used as in the "high efficiency concept", but without high-pressure injection. As the engine is mainly used in hybrid operation, no measures for de-throttling in naturally aspirated operation are implemented. The engine has a variable geometry turbocharger. Since the engine has a fixed compression ratio, water injection is used to reduce the tendency to knock in full load operation. The exhaust gas aftertreatment is redundant to the "high efficiency concept". The third concept is a simple turbo engine for use in range extender operation. The engine has the lowest specific power of the gasoline engines presented. The engine uses a high-turbulence concept to reduce combustion time and a turbocharger without variable geometry. The fourth engine concept is a natural gas engine, which has almost the same technology package as the "high efficiency concept". Only direct injection has been replaced by intake manifold injection to achieve better mixture homogenisation. In addition, an electric turbocharger was added to achieve sufficient cylinder filling even at low engine speeds. This is due to the higher air requirement of natural gas. For exhaust gas aftertreatment, the engine uses only an electrically heated three-way catalytic converter, since almost no particulate emissions occur during the combustion of natural gas. The fifth engine is a diesel engine with a comprehensive technology package. The

XXXIV

Abstract

engine uses injection with increased injection pressure and combustion control. It also has high and low pressure exhaust gas recirculation. A coated piston is used to reduce wall heat loss. The engine uses a variable geometry turbocharger. To comply with the strict limits of Euro-7 legislation, the engine uses a comprehensive exhaust aftertreatment package. The exhaust aftertreatment consists of a combined DOC/LNT catalytic converter, a particulate filter, two SCR catalytic converters, one of which is electrically heated, and an ammonia slip catalytic converter. Two possible electric motors are used for the electric drives. A permanent magnet synchronous motor is used for low and medium power. This uses copper windings in the stator and permanent magnets in the rotor. The machine is designed as a reluctance motor. The assumption in the efficiency increase up to the year 2040 is made on the basis of current standards. The second electrical motor is an induction motor. The motor is used for medium and high loads. In order to reduce the costs and environmental impact of this type of motor, the copper windings in the stator and the rotor-cage are replaced by aluminium. The assumption regarding efficiency for the year 2040 is also carried out using common standards. For the power electronics, SiC-MOSFETs are used instead of IGBTs. This approach leads to a strong increase in efficiency for the power electronics modules used. Furthermore, three battery types are defined to meet the different requirements of the powertrain architectures. For all batteries, assumptions are made for the year 2040 from the current technical literature. The first battery type is a highpower battery for use in conventional hybrid electric vehicles without external charging possibility. To keep the installed battery capacity as low as possible, it has a high C-rate with low energy density. Lithium titanium oxide is used as the anode material, and a special configuration of lithium iron phosphate is used for the cathode. The second battery type is a compromise between a highperformance and a high-energy battery with medium C-rates. This battery type is used for hybrid electric vehicles with external charging possibility in order to achieve good efficiency despite limited battery capacity. Carbon in the form of nano graphite is used as the anode material. The cathode consists of lithium manganese iron phosphate. The third battery type is a high-energy battery, which is used exclusively in battery electric vehicles. The C-rate is correspondingly low, but is sufficient in this application due to the large battery size. Accordingly, however, this battery has the highest energy density of all the battery types presented. The anode consists of carbon and silicon to achieve higher energy densities. Lithium nickel manganese cobalt oxide with increased

Abstract

XXXV

nickel content is used as the cathode material. Various improvements are assumed for the fuel cell system compared to today's systems. The fuel cell system can be divided into four main components. The first main component is the stack. The improvements in the stack are mainly expected in the area of the polymer electrolyte membrane, where a greatly increased energy density, reduced platinum loading and increased durability are anticipated. The second main component is the air path. In the air path, an electrically assisted turbocharger is used, which has increased efficiencies for both compressor and turbine. In the third main component, the fuel path, no improvement is assumed compared to current systems. The fourth main component is the electrical side of the system and the control system. Here, improvements are assumed in the area of the boost and buck converter. This is due to the use of SiC-MOSFETs instead of IGBTs. No improvements are assumed for the transmissions compared to today's systems. In the area of charging infrastructure, an efficiency improvement is assumed through the use of SiC-MOSFETs instead of IGBTs in the charging columns. In addition, a strong increase in fast charging performance is assumed compared to today's systems, as corresponding systems are already in pilot production. In the following chapter, the simulation environment for calculating the energy consumption and CO2-emissions is presented. First of all, a brief presentation of the newly developed opt. MO-ECMS algorithm is given, which is used for optimising the energy management of the powertrains. The newly presented algorithm achieves comparable results to the dynamic programming algorithm, but has advantages in terms of computation time, stability and flexibility. Afterwards, the three vehicle types considered are presented: a sedan, an SUV and a converted 3.5 t light-duty vehicle with a total gross vehicle weight of 7.5 t (LDV). In addition, the SUV is simulated with and without a 2.0 t tarpaulin trailer. For the driving resistance parameters of the longitudinal dynamics simulation, assumptions are made for the year 2040. Seven different powertrain architectures are examined for the SUV and the LDV, and eight for the sedan, in order to cover as wide a spectrum of vehicle configurations as possible. Added to this are various transmission configurations, resulting in a total of 57 simulated vehicle variants. Each vehicle variant is to be considered for different driving situations. For this purpose, five driving cycles are used, some of which are further subdivided into different speed categories: none, 100 km/h and 80 km/h speed limit. The driving cycles are WLTP and RDE

XXXVI

Abstract

cycles. The driving cycles are then applied to the corresponding vehicle variants. Finally, the ECMS methods used are presented in this chapter. A distinction is made between purely electric drive, parallel hybrid, serial hybrid and serial/parallel hybrid. In the last chapter of the thesis, the generated energy consumption as well as local CO2-emissions of the powertrain architectures are evaluated for the different driving cycles. Looking at the combustion engines in isolation, the following conclusions emerge: For the overall energy consumption, the gasoline engines with E10 show the lowest energy consumption under almost all conditions; only for very high loads do the diesel engines become more efficient. This is noticeable, for example, with the combination SUV + trailer or with the light-duty vehicle. There are two reasons for the lower efficiency of diesel engines compared to gasoline engines: Firstly, gasoline engines are heavily de-throttled or hybridised, which reduces the efficiency advantage of diesel engines in naturally aspirated operation. In addition, the diesel engines have to provide large amounts of electrical energy to operate the electrically heated SCR catalytic converter due to the low exhaust gas temperatures. The natural gas engines are always slightly worse than the gasoline engines with E10 due to increased charge exchange losses caused by the higher air requirement of the fuel. In terms of local CO2-emissions, the natural gas engines always perform best, followed by the gasoline engines with E10 and the diesel engines. This is where the carbon to hydrogen ratio of the fuels comes into effect. Looking at the hybrid configurations, it is noticeable that the higher the installed electric drive power, the lower the energy demand of the corresponding powertrain. This is due to a more efficient use of the combustion engine in the hybrid system and a higher recuperation potential during braking. Due to the higher installed electric drive power, the parallel-PHEV configurations generally have lower energy consumption than the HEV architectures. In hybrid mode, the parallel PHEV architectures have the lowest energy consumption of all the combustion engine configurations studied, unless the engines are oversized. In addition, due to their external recharge capability, these architectures can drive purely electrically in urban driving, resulting in locally CO2-emission-free traffic. The serial-PHEVs and serial/parallel-PHEV architectures have very similar energy consumptions, although the serial/parallel-PHEV architectures always perform slightly better due to the additional operating mode for parallel operation. This is due to the better efficiency chain in parallel operation. Both architectures almost always have lower energy consumption than the parallel

Abstract

XXXVII

HEV architectures, only at very high loads the energy consumption worsens due to the smaller combustion engine compared to the parallel HEVs. In urban driving, both architectures can drive purely electric and are thus locally CO2emission-free. The serial/parallel PHEV architectures have the lowest electric energy consumption in urban traffic due to the high electric drive power and the low vehicle weight. The fuel cell PHEV architecture has the second lowest energy consumption in hybrid mode. This is due to the high efficiency of the energy converter (fuel cell system). The fuel cell vehicle can drive purely electric in city traffic, the energy consumption is similar to the serial-PHEV architecture. The powertrain with fuel cell system is locally CO2-emission-free under all conditions. The battery-electric powertrain architecture has the lowest energy consumption in normal operation (cf. hybrid operation) due to the high efficiency of the battery. In urban traffic, however, this powertrain architecture performs worse in terms of energy consumption compared to the PHEVs. This is due to the high additional weight caused by the large batteries. The battery electric vehicles are locally CO2-emission-free under all conditions. In conclusion, it remains to be said that in terms of energy consumption, all powertrain architectures have advantages and disadvantages. Which powertrain is the most efficient always depends strongly on the respective driving situation. However, there is a tendency for the following gradation from high to low energy consumption: HEV, PHEV, fuel cell PHEV and battery electric vehicle. Nevertheless, the recommendation is to approach the CO2-problem in motorised private transport in an open-minded way in order to keep energy consumption and CO2-emissions as low as possible. In terms of local CO2emissions, only the fuel cell PHEVs and the battery electric vehicle have zero local CO2-emissions under all driving situations. All PHEV vehicles with combustion engines can be operated in urban traffic purely electrically and thus with zero local CO2-emissions. If the local CO2-emissions are compensated for by the use of e-fuels with "direct air capture", the global CO2-footprint of vehicles with combustion engines can be greatly reduced and they become competitive again. However, this would require a corresponding legal basis for the introduction of e-fuels. Building on this work, a comparable analysis for heavy-duty commercial vehicles is considered useful. Under certain circumstances, further technologies can be used here that have not yet been considered in this study. Simulation data sheets are provided for each vehicle as downloadable content and described in Appendix. This should make it possible to retrieve the data

XXXVIII

Abstract

and use it further. Furthermore, data sheets with an RDE compliance check for each of the driving cycles are provided in the same form.

Kurzfassung Das derzeit im EU-Parlament umgesetzte Ziel, bis 2035 nur noch klimaneutrale Fahrzeuge auf den Straßen neu zuzulassen, wird häufig mit der Umstellung des motorisierten Individualverkehrs auf batterieelektrische Fahrzeuge assoziiert. Aber auch andere Fahrzeugkonzepte bieten große Chancen, den Flottenverbrauch der Hersteller auf null zu reduzieren, insbesondere, wenn EFuels als Alternative zu fossilen Kraftstoffen genutzt werden. Zu diesem Zweck werden verschiedene Antriebstechnologien untersucht, die den rein batterieelektrischen Antrieb in drei Fahrzeugtypen mit weiteren möglichen Antriebsstrangarchitekturen vergleichen. Die unterschiedlichen Antriebsstrangarchitekturen werden im Rahmen des Gesamtenergieverbrauchs und der lokalen CO2-Emissionen bewertet. Zu den untersuchten Architekturen gehören Hybrid-Antriebe, Brennstoffzellen-Antriebe und batterieelektrische Antriebe. Ziel dieser Forschungsstudie ist es, verlässliche Vorhersagen über geeignete Antriebsstrangarchitekturen für den Markt im Jahr 2040 zu erhalten. Hierzu werden zunächst in einem ersten Kapitel die Simulationswerkzeuge sowie die mit diesen erzeugten Modelle vorgestellt. Die in einem späteren Kapitel vorgestellte Gesamtsystemsimulation erfolgt mit der Software MATLAB. Die Modellierung der Einzelkomponenten des Antriebsstrangs erfolgt mit verschiedenen Softwaretools. Die erzeugten Komponentenmodelle werden nach der Modellierung in vereinfachte Modelle für stationäre Betriebspunkte überführt und somit in die Gesamtsystemsimulation implementiert. Als Erstes wird auf die Erzeugung der Verbrennungsmotorenmodelle in GTPower unter Zuhilfenahme des FKFS UserCylinder eingegangen. Es wird die Modellierung verschiedener Technologien beschrieben, die für 2040 als marktreif erachtet werden. Da diese Technologien nicht alle vollumfänglich in der Software integriert sind, werden Ersatzmodelle umgesetzt, welche die gewünschten Effekte der jeweiligen Technologie so genau wie möglich nachbilden. Als Beispiele sind hier Vorkammerzündkerzen, Hochdruckeinspritzung und Zylinder-beschichtungen zu nennen. Ohne Ersatzmodelle umsetzbare Technologien sind beispielsweise ein variables Verdichtungsverhältnis, eine Wassereinspritzung, ein Hochturbulenzkonzept und Abgasrückführsysteme. Zusätzlich werden Untermodelle für die Abgasnachbehandlung, das thermische Verhalten der Verbrennungsmotoren sowie die Masse und Massenträg-

XL

Kurzfassung

heit der Verbrennungsmotoren vorgestellt. Anschließend wird auf die Modellierung der elektrischen Antriebssysteme in MATLAB eingegangen. Die elektrischen Antriebssysteme bestehend jeweils aus einer elektrischen Maschine und einem Leistungselektronikmodul. Es werden zwei Typen elektrische Maschinen vorgestellt, eine permanent magnetisch erregte Synchronmaschine und eine Asynchronmaschine. Beide Maschinen werden anhand von Ersatzschaltbildmodellen modelliert und anschließend mit einer optimierten Regelstrategie parametriert. Für die Optimierung des Maschinenbetriebs werden die Maximum Torque Per Ampere (MTPA) Strategie für den Grunddrehzahlbereich und zusätzlich die Maximum Torque Per Volt (MTPV) Strategie für den Feldschwächebereich genutzt. Im Feldschwächebereich wird der Motor immer an der Maximalspannungsgrenze betrieben, um die Phasenströme und somit die Verlustleistungen zu minimieren. Zusätzlich werden Untermodelle für die Masse und die Massenträgheit der elektrischen Maschinen vorgestellt. Für die Leistungselektronik wird eine gesteuerte Sechspuls-Brückenschaltung (B6c) genutzt. Die Modellierung der Leistungselektronik erfolgt mit einem Verlustleistungsmodell, welches die Schalt- und Durchlassverluste der Leistungshalbleiter berücksichtigt. Es wird ein Vorgehen zur Parametrierung des Verlustleistungsmodells anhand von Herstellerdaten vorgestellt, um ein generisches Modell für die Leistungselektronik zu erzeugen. Das so erzeugte Modell ermöglicht eine anschließende Skalierung der Leistungselektronik anhand der Maximalströme. Die Modellierung der Batteriezellen erfolgt in Excel. Hierzu werden die Verluste und auftretenden Effekte in der Batteriezelle in vier verschiedene Typen untergliedert und anschließend durch mathematische Funktionen angenähert. Die genährten Verluste werden anschließend von der Leerlaufspannung der entsprechenden Zellkonfiguration subtrahiert. Als Verluste sind hierbei die Aktivierungsüberspannung, die ohmschen Verluste und die Kinetischen und Massentransportverluste zu nennen. Des Weiteren werden Kinetische und Massentransporteffekte berücksichtigt, diese stellen in der Modellierung eine Limitierung der maximal nutzbaren Kapazität dar und verursachen keine Verluste. Zusätzlich wird ein Modell für die Batteriemasse bereitgestellt. Die Modellierung des Brennstoffzellensystems erfolgt in Simulink mit anschließender Vermessung und Export in ein stationäres MATLAB-Modell. Das Brennstoffzellensystem nutzt ein extrapoliertes Zellmodell für das Jahr 2040. Die Luftzufuhr wird mit einem elektrisch unterstützten Trubolader ausgeführt, um eine teilweise Nutzung der Abwärme zu erzielen. Die Befeuchtung der Membran erfolgt über eine Wassereinspritzung. Der Anodenkreislauf ist als geschlossener Kreislauf ausgeführt. Der Wasserstoff

Kurzfassung

XLI

wird mit einer Ejektorpumpe zirkuliert. Um erhöhte Stickstoffkonzentrationen im Anodenkreis zu vermeiden, ist der Anodenkreis zusätzlich mit einem Spülventil ausgestattet. Die Anpassung des Spannungsniveaus erfolgt durch einen Hoch- und Tiefsetzsteller. Zusätzlich enthält das Modell ein Untermodell für die Masse des Brennstoffzellensystems. Abschließen erfolgt die Modellbildung der verschiedenen Getriebe in Excel. Hierzu werden die Verluste in drei für Automatikgetriebe bzw. zwei für die restlichen Getriebe Verlustarten unterteilt. Die Verlustarten sind Verzahnungsverluste, Verlustdrehmoment und zusätzlich Verluste. Durch vorliegende Messdaten werden generische Getriebe erzeugt, welche in Abhängigkeit des Eingangsdrehmoments, der Gangzahl, der Getriebeübersetzungen und der Anzahl der Zahnradsätze skaliert werden. Im folgenden Kapitel wird die erwartete Technologieentwicklung bis zum Jahr 2040 prognostiziert und in die bereits vorgestellten Modelle implementiert. Bei den Verbrennungsmotoren werden fünf generische Motortypen erzeugt, welche über die mechanische Ausgangsleistung skaliert werden können. Alle Motoren weißen eine Direkteinspritzung auf. Ausgenommen hiervon ist der Erdgasmotor. Beim ersten Motor handelt es sich um einen hocheffizienten Ottomotor mit einem umfangreichen Technologiepaket, welcher als „high efficiency concept“ definiert wird. Dieser Motor weist eine hohe spezifische Leistung sowie eine starke Entdrosselung im Saugbetrieb und sehr kurze Brenndauern auf. Für die Entdrosselung im Saugbetrieb ist eine hochdruck Abgasrückführung und ein variabler Ventiltrieb mit Zylinderabschaltung vogesehen. Das Technologiepaket zur Umsetzung der kurzen Brenndauern umfasst eine Vorkammzündkerze, eine Hochdruckeinspritzung, ein Hochtrubulenzkonzept, sowie eine Tumbleklappe. Zur weiteren Steigerung der Effizienz kommt ein variables Verdichtungsverhältnis sowie ein Trubolader mit variabler Geometrie zum einsatz. Abgerundet wird das Konzept mit einem Euro-7 konformen Abgasnachbehandlungssystem mit elektrisch beheiztem Drei-Wege-Katalysator und Partikelfilter. Beim zweite Motorkonzept handelt es sich um einen Motor mit reduziertem Technologiepaket, um die Motorkosten zu senken. Der Motor hat eine mittlere spezifische Leistung und wird im weiteren als „budget optimized concept“ bezeichnet. Zur Reduzierung der Brenndauer wird das gleiche Technologiepaket wie im „high efficiency concept“ eingesetzt, jedoch ohne Hochdruckeinspritzung. Da der Motor hauptsächlich im Hybridbetrieb eingesetzt wird, werden keine Maßnahmen zur Entdrosselung im Saugbetrieb umgesetzt. Der Motor verfügt über einen Turbolader mit variabler Geometrie.

XLII

Kurzfassung

Da der Motor ein festes Verdichtungsverhältnis aufweist, wird zur Reduzierung der Klopfneigung unter Volllast eine Wassereinspritzung eingesetzt. Die Abgasnachbehandlung ist redundant zum „high efficiency concept“. Das dritte Konzept ist ein einfacher Turbomotor für den Einsatz im Range-Extender-Betrieb. Der Motor weist die niedrigste spezifische Leistung der vorgestellten Ottomotoren auf. Der Motor nutzt ein Hochturbulenzkonzept zur Reduzierung der Brenndauer sowie einen Turbolader ohne variable Geometrie. Das vierte Motorkonzept ist ein Erdgasmotor, welcher das nahezu gleiche Technologiepaket wie das „high efficiency concept“ aufweist. Lediglich die Direkteinspritzung wurde durch eine Saugrohreinblasung ersetzt, um eine bessere Gemischhomogenisierung zu erreichen. Zusätzlich wurde ein elektrischer Turbolader hinzugefügt, um auch bei niedrigen Drehzahlen eine ausreichende Zylinderfüllung zu erreichen. Dies ist auf den höheren Luftbedarf von Erdgas zurückzuführen. Bei der Abgasnachbehandlung nutzt der Motor lediglich einen elektrisch beheizten Drei-Wege-Katalysator, da bei der Verbrennung von Erdgas so gut wie keine Partikelemissionen auftreten. Beim fünften Motor handelt es sich um einen Dieselmotor mit einem umfassenden Technologiepaket. Der Motor nutzt eine Einspritzung mit gesteigertem Einspritzdruck sowie eine Verbrennungsregelung. Zusätzlich verfügt er über eine Hoch- und Niederdruck-Abgasrückführung. Zur Reduzierung der Wandwärmeverluste kommt ein beschichteter Kolben zum Einsatz. Der Motor nutzt einen Trubolader mit variabler Geometrie. Um die strengen Grenzwerte der Euro-7 Gesetzgebung einzuhalten nutzt der Motor ein umfangreiches Abgasnachbehandlungspaket. Die Abgasnachbehandlung besteht aus einem kombinierten DOC/LNT-Katalysator, einem Partikelfilter, zwei SCR-Katalysatoren, einer davon elektrisch beheizt, sowie einem Ammoniakschlupf-Katalysator. Für die elektrischen Antriebe werden zwei mögliche elektrische Maschinen genutzt. Für kleine und mittlere Leistungen wird eine permanent magnetisch erregte Synchronmaschine eingesetzt. Diese nutz Kupferwicklungen im Stator und Permanentmagnete im Rotor. Die Maschine ist als Reluktanzmaschine ausgeführt. Die Annahme in der Effizienzsteigerung bis zum Jahr 2040 wird anhand aktueller Normen vorgenommen. Bei der zweiten elektrischen Maschine handelt es sich um eine Asynchronmaschine. Die Maschine wird für mittlere und hohe Lasten eingesetzt. Um die Kosten und die Umweltwirkung dieses Maschinentyps zu senken, werden die Kupferwicklungen in Stator sowie der Käfigläufer durch Aluminium ersetzt. Die Annahme bezüglich der Effizienz für das Jahr 2040 wird ebenfalls anhand gängiger Normen durchgeführt. Für die Leistungselektronik werden SiC-MOSFETs anstatt von IGBTs eingesetzt. Dieses Vorgehen

Kurzfassung

XLIII

führt zu einer starken Effizienzsteigerung für die verwendeten Leistungselektronikmodule. Des Weiteren werden drei Batterietypen definiert, um die unterschiedlichen Anforderungen der Antriebsstrangarchitekturen zu erfüllen. Für alle Batterien werden Annahmen für das Jahr 2040 aus der gängigen Fachliteratur getroffen. Beim ersten Batterietyp handelt es sich um eine Hochleistungsbatterie für den Einsatz in konventionellen Hybridfahrzeugen ohne externe Lademöglichkeiten. Um die installierte Batteriekapazität möglichst gering zu halten, weist diese eine hohe C-Rate bei niedriger Energiedichte auf. Als Anodenmaterial wird Lithiumtitanoxid genutzt, für die Kathode wird eine spezielle Konfiguration von Lithiumeisenphosphat eingesetzt. Beim zweiten Batterietyp handelt es sich um ein Kompromiss zwischen Hochleistungs- und Hochenergiebatterie mit mittleren C-Raten. Dieser Batterietyp wird für Hybridfahrzeuge mit externer Lademöglichkeit eingesetzt, um trotz begrenzter Batteriekapazität eine gute Effizienz zu erreichen. Als Anodenmaterial wird Kohlenstoff in Form von nano Grafit genutzt. Die Kathode besteht aus Lithium-Mangan-Eisenphosphat. Der dritte Batterietyp ist eine Hochenergiebatterie, welche ausschließlich in batterieelektrischen Fahrzeugen zum Einsatz kommt. Die C-Rate ist entsprechend gering, ist aber aufgrund der großen Batteriegröße in dieser Anwendung ausreichend. Entsprechend weist diese Batterie die höchste Energiedichte von allen vorgestellten Batterietypen auf. Die Anode besteht aus Kohlenstoff und Silizium, um höhere Energiedichten zu erzielen. Als Kathodenmaterial wird Lithium-Nickel-Mangan-Kobalt-Oxid mit erhöhtem Nickelanteil eingesetzt. Für das Brennstoffzellensystem werde verschiedene Verbesserungen im Vergleich zu heutigen Systemen angenommen. Das Brennstoffzellensystem lässt sich in vier Hauptkomponenten unterteilen. Die erste Hauptkomponente ist der Stack. Die Verbesserungen im Stack sind hauptsächlich im Bereich der Polymerelektrolytmembran zu erwarten, hier wird mit einer stark gesteigerten Energiedichte, einer reduzierten Platinbeladung sowie einer erhöhten Haltbarkeit gerechnet. Die zweite Hauptkomponente ist der Luftpfad. Im Luftpfad wird ein elektrisch unterstützter Turbolader eingesetzt, welcher erhöhte Wirkungsgrade sowohl für Kompressor als auch für die Turbine aufweist. In der dritten Hauptkomponente, dem Kraftstoffpfad, werden keine Verbesserung im Vergleich zu heutigen Systemen angenommen. Die vierte Hauptkomponente ist die elektrische Seite des Systems sowie deren Regelung. Hier werden Verbesserungen im Bereich des Hochund Tiefsetzstellers angenommen. Dies ist durch die Nutzung von SiC-MOSFETs anstatt von IGBTs zu begründen. Bei den Getrieben werden keine Verbesserungen im Vergleich zu heutigen Systemen angenommen. Im Bereich

XLIV

Kurzfassung

der Ladeinfrastruktur wird eine Effizienzverbesserung durch die Nutzung von SiC-MOSFETs anstatt von IGBTs in den Ladesäulen angenommen. Zusätzlich wird ein starker Anstieg der Schnellladeleistung im Vergleich zu heutigen Systemen angenommen, da sich entsprechende Systeme bereits in der Vorserie befinden. Im darauffolgenden Kapitel wird die Simulationsumgebung für die Berechnung des Energiebedarfs und der Emissionen vorgestellt. Hierzu erfolgt zunächst eine kurze Vorstellung des neu entwickelten opt. MO-ECMS-Algorithmus welcher für die Optimierung des Energiemanagements der Antriebsstränge zum Einsatz kommt. Der neu vorgestellte Algorithmus erzielt vergleiche Ergebnisse zum Dynamic Programming Algorithmus, weist jedoch Vorteile bei der Berechnungszeit, der Stabilität und der Flexibilität auf. Anschließend erfolgt die Vorstellung der drei betrachteten Fahrzeugtypen, hierbei handelt es sich um eine Limousine, einen SUV sowie ein konvertiertes 3,5 t Nutzfahrzeug mit einer zulässigen Gesamtmasse von 7,5 t (LDV). Zusätzlich wird der SUV mit und ohne 2,0 t Planenanhänger simuliert. Für die Fahrwiderstandsparameter der Längsdynamiksimulation werden Annahmen für das Jahr 2040 getroffen. Für den SUV und das LDV werden je sieben bzw. für die Limousine acht verschiedene Antriebsstrangarchitekturen untersucht, um ein möglichst großes Spektrum an Fahrzeugkonfigurationen zu erfassen. Hinzu kommen verschiedene Getriebekonfigurationen, was zu einer Gesamtanzahl von 57 simulierten Fahrzeugvarianten führt. Jede Fahrzeugvariante wird für verschieden Fahrsituationen betrachtet. Zu diesem Zweck werden fünf Fahrzyklen genutzt, welche teilweise nochmals in verschiedene Geschwindigkeitskategorien: keine, 100 km / h und 80 km / h Geschwindigkeitsbegrenzung unterteilt sind. Bei den Fahrzyklen handelt es sich und WLTP- und RDE-Zyklen. Die Fahrzyklen werden anschließend auf die entsprechenden Fahrzeugvarianten angewandt. Abschließend die genutzten ECMS-Verfahren vorgestellt. Hierbei wird zwischen rein elektrischen Antrieb, parallel Hybrid, seriellem Hybrid und seriell/parallel Hybrid unterschieden. Im letzten Kapitel der Arbeit werden die generierten Energieverbräuche sowie lokale CO2-Emissionen der Antriebsstrangarchitekturen für die verschiedenen Fahrzyklen ausgewertet. Betrachtet man die Verbrennungsmotoren isoliert, ergeben sich folgende Schlussfolgerungen: Im Bereich des Energieverbrauchs zeigen die Ottomotoren mit E10 unter fast allen Bedingungen den geringsten Energieverbrauch, lediglich für sehr hohe Lasten werden die Dieselmotoren effizienter. Dies fällt beispielsweise bei der Kombination SUV + Anhänger

Kurzfassung

XLV

bzw. beim leichten Nutzfahrzeug auf. Die schlechtere Effizienz der Dieselmotoren im Vergleich zu den Ottomotoren hat zwei Gründe: Zum Ersten sind die Ottomotoren stark entdrosselt oder entsprechend stark hybridisiert, was den Wirkungsgradvorteil der Dieselmotoren im Saugbetrieb schmälert. Zusätzlich müssen bei den Dieselmotoren aufgrund der niedrigen Abgastemperaturen große Mengen elektrischer Energie für den Betrieb des elektrisch beheizten SCR-Katalysators bereitgestellt werden. Die Erdgasmotoren sind aufgrund erhöhter Ladungswechselverluste bedingt durch den höheren Luftbedarf des Kraftstoffs, stets etwas schlechter als die Ottomotoren mit E10. Bei den lokalen CO2-Emissionen schneidet die Erdgasmotoren stets am besten ab, gefolgt von den Ottomotoren mit E10 und den Dieselmotoren. Hier kommt das Verhältnis von Kohlenstoff zu Wasserstoff, der Kraftstoffe zum Tragen. Betrachtet man die Hybrid-Konfigurationen, fällt auf, dass je höher die installierte elektrische Antriebsleistung ist, desto geringer wird der Energiebedarf des entsprechenden Antriebsstrangs. Dies ist durch eine effizientere Nutzung des Verbrennungsmotors im Hybridverbund und ein höheres Rekuperationspotenzial beim Bremsen zu begründen. Aufgrund der höhren installierten elektrischen Antriebsleistung weisen die Parallel-PHEV-Konfigurationen generell einen geringeren Energieverbrauch als die HEV-Architekturen auf. Im Hybridbetrieb haben die Parallel-PHEV-Architekturen den geringsten Energieverbrauch von allen untersuchten Konfigurationen mit Verbrennungsmotor, außer die Motoren sind zu groß dimensioniert. Zudem können aufgrund der externen Aufladbarkeit diese Architektur im Stadtbetrieb rein elektrisch fahren, was zu einem lokal CO2-emissionsfreien Verkehr führt. Die Seriellen-PHEVs und Seriell/Parallel-PHEV-Architekturen weisen sehr ähnliche Energieverbräuche auf, wobei die Seriell/Parallel-PHEV-Architekturen aufgrund des zusätzlichen Betriebsmodus für den parallelen Betrieb stets etwas besser abschneiden. Dies ist durch die bessere Wirkungsgradkette im Parallelbetrieb zu begründen. Beide Architekturen weisen fast immer einen geringeren Energieverbrauch als die Parallel-HEV-Architekturen auf, lediglich bei sehr hohen Lasten verschlechtert sich der Energieverbrauch aufgrund des kleineren Verbrennungsmotors im Vergleich zu den Parallel-HEVs. Im Stadtbetrieb können beide Architekturen rein elektrisch fahren und sind somit lokal CO2-emissionsfrei. Die Seriell/Parallel-PHEV-Architekturen weisen aufgrund der hohen elektrischen Antriebsleistung und des geringen Fahrzeuggewichts den niedrigsten elektrischen Energieverbrauch im Stadtverkehr auf. Die Brennstoffzellen-PHEV-Architektur weist den zweit niedrigsten Energieverbrauch im Hyb-

XLVI

Kurzfassung

ridbetrieb auf. Das ist durch die hohe Effizienz des Energiewandlers (Brennstoffzellensystem) zu begründen. Das Brennstoffzellenfahrzeug kann im Stadtverkehr rein elektrisch fahren, der Energieverbrauch ist ähnlich zur Seriellen-PHEV-Architektur. Der Antriebsstrang mit Brennstoffzellensystem ist unter allen Bedingungen lokal CO2-emissionsfrei. Die batterieelektrische Antriebsstrangarchitektur weist im Normalbetrieb (vgl. Hybridbetrieb) aufgrund der hohen Effizienz der Batterie den geringsten Energieverbrauch auf. Im Stadtverkehr schneidet diese Antriebsarchitektur bezüglich des Energieverbrauchs im Vergleich zu den PHEVs jedoch schlechter ab. Dies ist durch das hohe Zusatzgewicht aufgrund der großen Batterien zu begründen. Die batterieelektrischen Fahrzeuge sind unter allen Bedingungen lokal CO2-emissionsfrei. Abschließend bleibt zu sagen, dass bezüglich des Energieverbrauchs alle Antriebsstrangarchitekturen Vor- und Nachteile haben. Welcher Antrieb der effizienteste ist, ist immer stark von der jeweiligen Fahrsituation abhängig. Tendenziell stellt sich jedoch die folgende Abstufung von Hohem zu niedrigen Energieverbrauch dar: HEV, PHEV, Brennstoffzellen-PHEV und batterieelektrisches Fahrzeug. Dennoch ist die Empfehlung technologieoffen an die CO2-Problematik im motorisierten Individualverkehr heranzutreten, um den Energieverbrauch und die CO2-Emissionen so gering wie möglich zu halten. Bei den lokalen CO2-Emissionen weisen lediglich die BrennstoffzellenPHEVs und das batterieelektrische Fahrzeug unter allen Fahrsituationen null lokale CO2-Emissionen auf. Alle PHEV-Fahrzeuge mit Verbrennungsmotor können im Stadtverkehr rein elektrisch und somit lokal CO2-emissionsfrei betrieben werden. Erfolgt eine Kompensation der lokalen CO2-Emissionen durch die Nutzung von E-Fuels unter Einsatz von "direct air capture", kann der globale CO2-Fußabdruck der Fahrzeuge mit Verbrennungsmotor stark gesenkt werden und diese werden wieder wettbewerbsfähig. Hierfür wäre aber eine entsprechende gesetzliche Grundlage zur Einführung von E-Fuels notwendig. Aufbauend auf dieser Arbeit wird eine vergleichbare Analyse für schwere Nutzfahrzeuge als sinnvoll erachtet. Hier können unter Umständen weitere Technologien zum Einsatz kommen, die in dieser Arbeit noch nicht betrachtet wurden. Für jedes Fahrzeug wird ein präzises Simulationsdatenblatt als Download zur Verfügung gestellt. Eine Beschreibung hierzu befindet sich im Anhang. Hierüber soll die Möglichkeit gegeben werden, die Daten zu extrahieren und diese

Kurzfassung

XLVII

weiterzuverwenden. Des Weiteren werden Datenblätter mit einer RDE-Konformitätsprüfung für jeden der Fahrzyklen in der gleichen Form zur Verfügung gestellt.

1 Introduction One of the main problems of our time is the effect of global warming caused by the emission of greenhouse gases (GHG). The biggest share of the GHG emissions is CO2 which is formed and emitted by each combustion process of carbon fuels. Our energy supply is mostly based on the combustion of fossil fuels in technical combustion processes for generating grid power, industrial processes or for means of transportation. To reduce the overall emission of GHG, in particular CO2, reduction measures for the concerned sectors need to be applied. To find adequate measures, a UN Climate Change Conference is held every year to determine reduction goals for the future in order to limit the temperature rise caused by GHG emissions to a certain level. A result of these conferences are protocols in which signing countries obligate themselves to certain measures for reducing GHG emissions. The first protocol was the Kyoto Protocol, which was in action from 1997 until 2015, followed in 2015 by the Paris Agreement, which is currently in force. The Paris Agreement obligates the signed countries to CO2-reduction measures for limiting the global temperature rise to 2°C compared to pre-industrial levels [1]. To reach those goals the EU puts laws into place that limit the CO2-emissions of the different sectors. One sector is road traffic, which in 2017 was responsible for 21 % of the overall CO2-emissions in the EU. This sector is further divided into heavyduty vehicles with an overall share of 6 % and light-duty vehicles with an overall share of 15 % in total CO2-emissions. 12 % of those 15 % are caused by passenger cars [2]. For heavy-duty vehicles, reduction values are being discussed but are not applied yet. The formulated reduction goals are 15 % reduction for 2025 and 30 % for 2030 referred to the CO2-emission levels between July 2019 and June 2020. Due to the fact that the investigated heavy-duty vehicle, with 7.5 t maximum gross weight, is a converted van, it is assumed that the emission targets for light-duty vehicles apply. For the light-duty vehicles, the new CO2-emission standards for cars and vans apply since 14 July 2022. In Fig. 1, the different targets for car CO2-emissions can be seen. The reduction values for passenger cars are 15 % for 2025, 55 % for 2030 and 100 % for 2035, starting from a value of 95 g CO2 / km. For vans, the formulated reduction goals are 15 % for 2025, 50 % for 2030 and 100 % for 2035, starting from a value of 147 g CO2 / km. These reduction values apply to the fleet-consumption of Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/978-3-658-42168-7_1.

© The Author(s), under exclusive license to Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2023 T. Stoll, A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040, Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart, https://doi.org/10.1007/978-3-658-42168-7_1

1 Introduction

2

each car manufacturer and if not reached, can lead to delicate penalty payments for the manufacturers [2] [3] [4]. 180 160 140 g CO2 / km

120 100 80 60 40 20 0 2000

Figure 1.1:

Historical trend line EU-27 reduction target values EU-27 reduction target line

2010

2020 year

2030

2040

CO2-emissions of newly registered passenger cars: Historical trend line [5], EU-27 reduction targets [2] [3] [4].

It can be seen that the legislation will lead to zero CO2-emissions for fleet consumption by 2040. This can only be achieved through two measures. The first measure is the decarbonisation of the vehicle fleet by converting vehicle powertrains to battery electric and fuel cell, which produces zero CO2 emissions locally. The second measure is the usage of carbon neutral fuel generated with electricity and further referred as e-fuel. The e-fuels generated with electricity from regenerative sources and CO2 form the air (direct air capture) are almost carbon neutral. Nevertheless, it has to be considered that the CO2-emissions of the e-fuels are strongly dependent on the CO2-emissions of the used electricity. It has to be said, that this measure needs to be forced by infrastructure to only deliver these e-fuels to the supply chain instead of fossil fuels. Due to current legislative proposals in the EU, introducing e-fuels for marine and aviation applications, a tendency to transform the infrastructure to e-fuels seems plausible [6] [7]. In order to obtain utilizable data for estimations of the CO2-emissions, this study investigates the most promising powertrain architectures in terms of overall energy consumption and local CO2-emissions. With the generated data

1 Introduction

3

it will be possible to calculate fleet consumptions by using the different investigated powertrains with a custom composition of the different powertrain technologies. Chapter 2 provides a general overview of the modelling approach for the simulations used in this study. The different models are introduced in detail. Furthermore, the scaling of the models for different variants and sizes as well as the export into steady state characteristic models for the simulation in MATLAB is described. Chapter 3 deals with the expected development of powertrain technologies until 2040. For this, the different components used in the powertrain architectures are simulated with estimated developments until 2040. First, five different combustion engines, including three gasoline engine concepts, a diesel engine concept and a natural gas engine concept are introduced. For each concept, precise simulative estimations are made to predict their future fuel consumption and to achieve Euro-7 emission goals. Second, two different electric drive systems are presented, one with a permanent magnet synchronous motor and one with an induction motor to match the different demands of the powertrains. Three different types of battery cells are also introduced to match the different powertrain technologies, as well as a fuel cell system for the fuel cell electric vehicles. In addition, nine different types of gearboxes are estimated and simulated. In the last section, the tank systems and charging columns for the battery electric vehicles are introduced. In chapter 4, the simulations used in this study are explained. Three vehicle types are considered, a sedan car, a sport utility vehicle (SUV) with and without a 2.0 t tarpaulin trailer, as well as a converted 3.5 t light-duty vehicle with a maximum gross weight of 7.5 t. Due to its field of application, the converted light-duty vehicle is henceforth referred to as light-duty vehicle (LDV). Eight different powertrain architectures are introduced, resulting in the simulation of 57 different vehicles. Each vehicle is evaluated for different driving cycles. This study executes nine different drive cycles for the evaluation of the different powertrain configurations and vehicle types. Each vehicle is simulated with an optimized operating strategy to minimize its energy consumption. Chapter 5, provides an evaluation of the different powertrain architectures for the investigated vehicles. Each vehicle is evaluated in terms of its energy consumption and local CO2-emissions. The evaluation is conducted for all relevant driving cycles of each vehicle type. Finally, a comparison of the different

4

1 Introduction

powertrain architectures is made to elaborate the advantages and disadvantages of each variation. In chapter 6, a summary is given, complemented by an outlook for future research and further investigations. Appendix A1 and A2 and the downloadable content contain data sheets for each vehicle and each driving cycle to give the possibility of retrieving the simulation data.

2 Modelling Methods To perform the simulations effectively, an appropriate model approach for each component of the powertrain and for the overall powertrain is defined. For the internal combustion engines, a static engine simulation using GTPower from Gamma Technologies and the FKFS UserCylinder is created. The other components are modelled in MATLAB, MATLAB Simulink and Excel. All component models are then converted into steady state characteristic map models for the full longitudinal dynamics simulation of the vehicles and the powertrains. Next, a fitting control strategy for each powertrain is developed and implemented into the vehicle simulation. While developing the control strategy, the packaging of each vehicle and powertrain combination is optimized for minimum energy consumption and minimum powertrain costs. Finally, the vehicle is simulated on a virtual test bench to generate the results. The model development and simulation process is illustrated in Figure 2.1. This chapter deals with the development of component models, while the development of the vehicle model and the control strategy is explained in chapter 4. GT-POWER

MATLAB

ICE Development

Optimal Control Strategy Development

FC Development

& Optimized Packaging O ng

•Fuel Consumption (el./chem.) •CO2-emissions

adjusted for good drivability MATLAB/SIMULINK/EXCEL

MATLAB

MATLAB

Component Development

Vehicle Model

Virtual Test Bench

Figure 2.1:

Model development and simulation process.

© The Author(s), under exclusive license to Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2023 T. Stoll, A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040, Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart, https://doi.org/10.1007/978-3-658-42168-7_2

2 Modelling Methods

6

2.1

Internal Combustion Engines

For each internal combustion engine concept a full GT-Power model is created and parametrized. To model the effects that take place in the internal combustion engine a wide range of sub models in the FKFS UserCylinder is used. If these effects are not modelled in the FKFS UserCylinder, estimations retrieved from published research and measurement data are included. The most important models, effects and parameter estimations are: Knock: Knock describes a self-ignition of the fuel in the cylinder during the compression phase of the cylinder or during combustion without an external ignition source. Due to the rapid combustion during knock, this effect potentially damages the combustion engine and has to be avoided. Knock events limit the efficiency of spark ignited combustion engines, as they limit temperature and pressure in the combustion chamber. High pressures and temperatures lead to higher efficiency, but also to higher knock probability. To model the knock event properly, two implemented knock models in the FKFS UserCylinder are used. These models are based on measurement data and provide high accuracy in calculating knock events. The phenomenological knock model for the Otto engines using liquid fuel was created by Alexander Fandakov [8]. In contrast to other knock models, the knock model created by Fandakov provides high accuracy over changing conditions in the combustion chamber and over the use of different engine technologies, like variable compressions ratio, exhaust gas recirculation and water injection, which are used in this study. For Otto engines using gaseous hydrocarbon fuels like methane, a phenomenological knock model created by Lukas Urban [9] is used. Water injection: Water injection is a measure to reduce knock due to the evaporation enthalpy of the water and the higher charge mass in the cylinder. The first effect lowers the in-cylinder temperature, the second effect improves the overall heat capacity of the cylinders charge. Water injection also shows a positive effect on the exhaust gas temperature which protects the turbocharger against overheating [10]. The water is injected during the compression stroke of the cylinder at the intake manifold of the cylinders. The injected water mass is limited to 40 % of the injected fuel mass to prevent condensation of the injected water in the compressor of the turbocharger. This may happen due to the blow-by mass flow during combustion, which is fed back to the air box at

2.1 Internal Combustion Engines

7

the engines intake. Water condensation in the compressor of the super- or turbocharger would lead to damage on the compressor blades. Water injection is modelled in the FKFS UserCylinder. Pre-chamber spark plug: For this study, an active pre-chamber spark plug for the Otto engines is estimated. In general, a pre-chamber plug replaces the standard spark plug. The pre-chamber spark plug contains a fuel injector and a spark plug, as well as multiple holes to the combustion chamber for delivering ignition jets. With this configuration, a small amount of fuel is injected into the pre-chamber, which is then ignited by the spark plug. Through the holes in the pre-chamber spark plug, already burning ignition jets enter the combustion chamber. This leads to fast burn rates because the charge in the cylinder is ignited in multiple positions and at high turbulence levels generated by the burning jets. Due to the significantly improved burn rates, the knock probability is reduced because the fuel burns faster than the pre-reactions for knock can take place. The pre-chamber spark plug is not modelled in the FKFS UserCylinder. An approach to estimate its effect needs to be found. The most relevant effect is the decrease in burn duration. This effect is modelled by increasing the charge motion parameter in the FKFS UserCylinder. To give an estimation of how much the turbulence needs to be increased, the published measurement data from [11] is analysed. The 10 % -90 % burn duration for a λ = 1 concept is reduced by around 50 % by using a pre-chamber spark plug, compared to a standard spark plug. This 50 % reduction is now applied to the internal combustion engine model without pre-chamber spark plug by improving the charge turbulence level until the 10 % - 90 % burn duration is reduced by 50 % at an engine speed of 1500 min-1 and indicated mean effective pressure of 8.0 bar. High-pressure injection: High-pressure injection uses a 1000 bar injector for direct injection. First evidence in [12] shows that knock can be decreased by injecting the fuel close to top dead centre of the compression stroke. The start of injection for this measurement data [12] and for the estimations in this study is at 30 ° before top dead centre. Two effects that lead to lower knock are identified in [12]. The first effect is the increase in charge motion due to the high-pressure injection. This increase is already modelled with the pre-chamber spark plug and the high turbulence concept. Furthermore, this effect is not taken into account for increasing knock resistance because the minimum burn duration is already modelled up to a maximum pressure rise of 8 bar / °CA. If this effect should be modelled later on, the charge motion has to be reduced

8

2 Modelling Methods

by reducing the turbulence generation by the high turbulence concept. The second effect that leads to higher knock resistance is that the pre-reactions in the air-fuel mixture are delayed due to the late start of injection. To describe this effect the Livengood-Wu integral is used, which calculates how far the chemical pre-reactions that lead to knock are proceeded in the air-fuel mixture. It calculates the formation of critical chemical reactants in relation to the critical reactant concentration for auto ignition, resulting in a maximum value for auto-ignition of 1.0. For the high-pressure injection, with an injection start at 30° before top dead centre, the value for the Livengood-Wu integral is around 20 % lower than for the same combustion without high-pressure injection. With that knowledge, the high-pressure injection is applied to the model by delaying knock until the Livengood-Wu integral reaches a value of 1.20. Combustion Control: The diesel engine has between one and two pilot injections and one rate shaped main injection. The pilot injections are used for nitrogen oxide emission reduction and for limiting the maximum pressure rise during combustion to 8 bar / °CA. The first flank of the main injection is rate shaped and controlled for controlling the pressure rise during combustion to 8 bar / °CA. The combustion control for the diesel engine is modelled with the FKFS UserCylinder and the GT-Power environment. High turbulence: The high turbulence concept targets high tumble levels for the intake charge of the engine. A high turbulence level can be achieved by special intake manifold design as well as with a tumble flap. Also, a design with two stages using a tumble flap to switch between those stages is possible. A higher turbulence level leads to faster burn rates of the air-fuel mixture in the combustion chamber [13]. For the concepts presented here, the maximum turbulence level from [13] is set as a baseline. The turbulence is then reduced until an in-cylinder pressure rise of a maximum of 8 bar / °CA is achieved. In addition, concepts with a tumble flap are intended to switch between two different turbulence levels. The turbulence is modelled in the FKFS UserCylinder by tuning the parameter for the charge motion. Exhaust gas recirculation: The exhaust gas recirculation is applied to Otto engines as well as to diesel engines. For the Otto engines, the exhaust gas recirculation system is used for dethrotteling. The exhaust gas recirculation system adds a certain amount of inert gas to the cylinder charge. Hence, the amount of air-fuel mixture is reduced and at the same operation point, the pressure at the intake of the cylinder can be increased. For the Otto engines the

2.1 Internal Combustion Engines

9

maximum exhaust gas recirculation rate is 30 % of the intake charge to always provide a flammable mixture in the combustion chamber. For the diesel engines the exhaust gas recirculation system is used to lower the nitrogen oxide emissions. This happens by mixing the intake air with a certain amount of inert exhaust gas. To burn the same amount of fuel in the same way the intake airflow has to be kept constant in first approximation. This leads to an increase of the in-cylinder charge. The higher in-cylinder charge leads to a higher heat capacity and consequently to lower temperatures in the combustion chamber. This slows the reaction of nitro oxide formation after the Zel'dovich mechanism and results in lower nitrogen oxide emissions of the engine. The exhaust gas recirculation systems are modelled in the standard GT-Power environment. For the calculation of the combustion processes the FKFS UserCylinder is used. Engine friction: For the engine friction, either base values from a FKFS internal source or from [13] are used and then adapted to a friction model with basic data retrieved from [14]. First, the base value for the engine friction is adapted to 2040 levels. Therefore, the engine friction for the Otto engine with liquid fuels is set to the friction values of [13]. The engine friction of the Otto engine with gaseous fuels, also with the baseline from [13], is raised by 10 % due to high cylinder pressure; causing a more robust engine construction. The friction for the diesel engine is equal to the baseline from a FKFS internal source and is kept constant to current levels, estimating a more robust engine construction for 2040 due to higher in cylinder pressures. After adapting the base values, the friction model is introduced. The friction model is scalable over the displacement, the stroke-to-bore ratio and the cylinder number of the engine. The default values for the displacement of one cylinder are 375 cm³ for the Otto engine and 500 cm³ for the diesel engine. The default value for the stroke-to-bore ratio is 1.15 for the Otto engine and 1.13 for the diesel engine. Both engine types use four cylinders as reference. The functions used to scale the friction mean effective pressure of the engines in the GT-Power models are shown in Figure 2.2.

2 Modelling Methods

10 1.075

Kpmr,Vh,Diesel Poly. (Kpmr,Vh,Diesel)

1.050 Kpmr,Vh [-]

1.025

Kpmr,Vh,Otto Poly. (Kpmr,Vh,Otto)

y = 0.6415x2 - 0.8087x + 1.244

1.000 0.975 y = 0.6222x2 - 0.7844x + 1.2067 0.950 0.30 0.35 0.40 0.45 0.50 0.55 0.60 cylinder displacement Vh [cm³]

0.65

2.500 Kpmr,sb [-]

2.000

Kpmr,sb,Otto

Kpmr,sb,Diesel

Poly. (Kpmr,sb,Otto)

Poly. (Kpmr,sb,Diesel)

1.500

y = 2.0036x2 - 5.5004x + 4.6526

1.000 0.500 0.70

y = 1.9854x2 - 5.4886x + 4.6863

0.80

0.90 1.00 1.10 stroke-to-bore ratio s/b [-]

1.20

1.30

1.100 Kpmr,zcyl.,Otto Poly. (Kpmr,zcyl.,Otto)

1.080

Kpmr,zcyl.,Diesel Poly. (Kpmr,zcyl.,Diesel)

Kpmr,zcyl. [-]

1.060 y = 0.012x2 - 0.101x + 1.2115

1.040 1.020 1.000

y = 0.0073x2 - 0.0652x + 1.1445

0.980 1

Figure 2.2:

2

3 4 number of cylinders zcyl. [-]

5

Factors for scaling of friction mean effective pressure over the cylinder displacement (top), stroke-to-bore ratio (mid) and the number of cylinders (bottom) in the GT-Power models.

2.1 Internal Combustion Engines

11

Variable valve train: The variable valve train is used for the Otto engine concepts to dethrottle the engine. For this, two functions of the variable valve train are implemented. The first function is an intake valve shift to enable the Miller cycle. In the Miller cycle the intake valve closes before bottom dead centre of the intake stroke. This leads to a lower in-cylinder charge at the same intake pressure. The second function is the possibility to switch to a zero cam for certain cylinders, for the intake as well as the exhaust valves. The cylinders that are switched to the zero cam are no longer fired. With this function certain cylinders are deactivated. The number of active cylinders is thereby reduced and their load point is shifted in direction to higher loads. For this, a higher amount of air is needed and the intake pressure can be increased in contrast to the possibility without deactivated cylinders. These two functions are modelled in the standard GT-Power environment. Variable compression ratio: The variable compression ratio is used for the Otto engines to reach their full efficiency potential at low and medium loads. For high loads, that are already close to knock events the compression ratio is held low. For low and medium loads the compression ratio is increased as much as occurring knock events allow. The variable compression ratio is modelled in the standard GT-Power environment. Heat insulated piston: The diesel engine uses a heat insulated piston proposed in [15], to minimize wall-heat losses during combustion. The low heat capacity and conductivity proposed in this publication leads to a large decrease in heat transfer through the piston. The temperature of the piston surface strongly follows the current temperature of the cylinder charge, leading to high temperatures during combustion and low temperatures during charge exchange. During combustion, the heat loss is minimized, resulting in higher combustion efficiencies. The values for heat capacity and heat conductivity of the piston are adapted to the values introduced in the publication [15] and are modelled in the FKFS UserCylinder. Super and Turbochargers: Three different kind of chargers are used for the engine simulation in GT-Power, respectively an electrical powered supercharger, a variable-geometry turbocharger and a standard turbocharger. For the super- and turbocharger efficiency of the compressor an improvement of 11 % is assumed until 2040, referring to current levels. For the turbines of the turbochargers, this improvement is assumed by 14 %. The assumptions are retrieved from [13] and implemented in the standard GT-Power environment.

12

2 Modelling Methods

The combustion engine models in GT-Power are exported to steady state characteristic models in MATLAB. This approach is required to keep simulation times for the virtual test bench at acceptable levels and still provide data with the necessary accuracy. Therefore, the assumption is made that the influence of dynamic load shifts can be neglected onto the characteristic maps of the engines. The main objective for dynamic simulation is the simulation of exhaust gas raw emissions caused by operation with richer mixture during acceleration. For the assumed Euro-7 emission standards, the Otto engines are always operated with stoichiometric air-to-fuel ratio to reduce carbon monoxides and hydrocarbon emissions. The stoichiometric operation of the Otto engine in combination with an electrically heated catalytic converter provides low emission levels that match the Euro-7 emission standards. In conclusion, for the Otto engines the emissions are not modelled. The Diesel engines are always operated with lean mixture which also leads to emissions similar to the static simulations. For the diesel engines, the nitro oxide emissions are estimated in the simulation, while neglecting dynamic effects. For the full longitudinal dynamic simulation of the vehicle and powertrain later on, the combustion engines have to be scaled to different engine powers resulting in different sizes of the engines. For that purpose, the maximum and minimum size of the cylinders is defined. For the Otto engines, the cylinder swept volume is between 250 cm³ and 500 cm³, for the diesel engines, this value ranges between 450 cm³ and 650 cm³. To scale the engines correctly three scaling functions are introduced. The first scaling function scales the specific injected fuel mass over the engines displacement. Therefore, the engines, modelled in GT-Power, are simulated and optimized with different cylinder swept volumes. During the simulations, the friction mean effective pressure is kept constant for all configurations, while the size of the super- and turbochargers as well as the stroke-to-bore ratio are optimized for maximum efficiency. This proceeding is chosen because the scaling of the friction mean effective pressure is realized in another function. After that, the specific injected fuel mass is measured at the point of minimum brake specific fuel consumption to create the scaling function for the influence of different cylinder swept volumes. The first scaling function for an example Otto engine is shown in Figure 2.3 (top). The second scaling function scales the friction mean effective pressure over the engines displacement, using the correlations for the engine friction presented earlier in this section [14]. The second scaling function for an example Otto engine is shown in Figure 2.3 (mid). The third scaling

2.1 Internal Combustion Engines

13

function is the influence of the engine temperature on the friction mean effective pressure. For this purpose, the engine is simulated with different temperatures and the values for the friction mean effective pressure are measured at the point of minimum brake specific fuel consumption. The corresponding third scaling function for an example Otto engine is presented in Figure 2.3 (bottom). The rotating inertia of the combustion engines reduced to the crankshaft is retrieved from [16] and calculates as follows:

‫ܬ‬ூ஼ா ൌ ͺǤ͸ͷͻͺͳͲ ‫ିͲͳ כ‬ଵ଴ ‫ܸ כ‬ு ଶ ൅ ͲǤͲ͵Ͳ͹ ‫ିͲͳ כ‬ଷ ‫ܸ כ‬ு ൅ ͲǤͲͺ͵ͳ

‡“ǤʹǤͳ

‫ܬ‬ூ஼ா : Combustion engine rotating inertia reduced to crankshaft [kg m²] ܸு :

Engine displacement [cm³]

The engine warm up and cooling-off is modelled with the basic equations of heat transfer. The cooling-off against the ambient is parametrized with measurement data retrieved from [17]. The engine is assumed with a high degree of engine insulation to delay its cooling-off. For the warm up of the engine, the losses due to wall heat transfer and engine friction are taken into account, modelled with characteristic maps exported from GT-Power. Assuming a rapid warm-up of the engine, only the small cooling circuit of the internal combustion engine is used until the engine has reached its optimum operating temperature of 95 °C. Furthermore, it is assumed that no heat is dissipated via the engine radiator during the warm-up phase of the combustion engine. The engine mass is calculated with an equation retrieved from [18]:

݉ூ஼ா ൌ ͲǤʹͻͻͺ

݇݃ ‫ܯ  כ‬௠௔௫ ൅ ͷͻǤʹͶ݇݃ ܰ݉

݉ூ஼ா : Engine weight [kg] ‫ܯ‬௠௔௫ : Maximum engine torque [Nm]

‡“ǤʹǤʹ

2 Modelling Methods

14

Kmffuel,VH [-]

1.060 1.040 1.020 1.000

Kpmr,VH [-]

0.980 250

500

750 1000 1250 1500 engine displacement VH [cm³]

1750

2000

250

500

750 1000 1250 1500 engine displacement VH [cm³]

1750

2000

-40

-20

0 20 40 60 engine temperature TICE[°C]

80

100

1.250 1.200 1.150 1.100 1.050 1.000

Kpmr,TICE[-]

6.000 4.000 2.000 0.000

Figure 2.3: Scaling functions of the characteristic map model for an example Otto engine for the influence of change in engine displacement on the specific injected fuel mass (top), for the change in engine displacement on the friction mean effective pressure (mid) and for the change in engine temperature on the friction mean effective pressure (bottom).

2.1 Internal Combustion Engines

15

The composition of the engine is reduced to four basic materials, steel, aluminium, polymer and oil. The estimated specific heat capacity as well as the mass composition for the internal combustion engines are summarized in Table 2.1. Table 2.1:

Estimated mass fraction and specific heat capacity of the materials used in the internal combustion engines.

Material

Mass fraction [%]

Specific heat capacity [J / kg / K]

Steel

52.2

490

Aluminium

33.2

896

Polymer

11.4

1700

Oil

3.2

2090

The catalytic converters use a thermal model to predict the catalyst temperature during simulation with the static characteristic model of the internal combustion engines. The model only consists of a simple thermal model without reaction enthalpy of the emissions in the exhaust gas. In addition, for the diesel engines the conversion (SCR) and storage (LNT) of nitrogen oxides is modelled with standard temperature dependent catalytic conversion curves. For modelling the catalytic converters, their dimensions are standardized. The results of the standardization are listed in Table 2.2. The thermal model for the catalytic converters uses the basic equations of heat transfer. For this purpose, the heat transfer coefficient for the heat transfer between the catalytic converter and the environment (outside) and between exhaust gas and catalytic converter (inside) is defined. The heat transfer coefficient for the outside of the catalytic converters is retrieved from [19]. The model for the outer heat transfer coefficient contains values in dependency of the vehicle speed and the position of the catalytic converter in the underbody of the vehicle. The values for the heat transfer coefficient on the outside of the catalytic converters are shown in Figure 2.4.

2 Modelling Methods

16 Table 2.2:

Values for the catalyst model in the internal combustion engines static characteristic model.

Description

Value

Unit

Material carrier

Steel

-

Carrier cell density

400

cpsi

Specific substrate weight per l catalyst volume [20]

0.40

kg / l

Specific wash coat weight per l catalyst volume [21]

0.15

kg / l

Specific inner surface per l catalyst volume

2.45

m² / l

Specific volume per kW engine power

0.012 – 0.024

l / kW

Length

80 - 160

mm

Coated monolith heat capacity [21]

435

J / kg / K

Specific electric heating power per kW engine power for electrical heated catalysts

33

W / kW

For the heat transfer coefficient on the inside of the catalytic converter, two GT-Power simulations are evaluated. The results for the heat transfer coefficients plotted over the engine speed are shown in Figure 2.5. In the first GTPower simulation (V1) the heat transfer coefficient of the three-way catalyst installed in an Otto engine is evaluated. In the second (V5), the heat-transfer coefficient of the diesel oxidation catalyst (DOC) and the selective catalytic reduction catalyst (SCR1) in a Diesel engine are examined. Based on the good matching measurements, an overall potency fit for the heat transfer coefficient on the inside of the catalytic converters is made.

2.1 Internal Combustion Engines

heat transfer coefficient αout [W/m²/K]

100

Engine Manifold Underbody Linear (Engine) Linear (Manifold) Linear (Underbody)

80 60 40

17

y = 0.4212x + 6.7135 y = 0.2889x + 5.803 y = 0.2132x + 6.0426

20 0 0

heat transfer coefficient αin [W/m²/K]

Figure 2.4:

50 45 40 35 30 25 20 15 10 5 0

50

100 150 vehicle speed [km / h]

200

Heat transfer coefficient for the outer surface of the catalytic converters against the environment, dependent on vehicle speed and installation position of the catalytic converts, implemented for the static characteristic model of the internal combustion engines.

TWC - V1 (GT) DOC - V5 (GT) SCR1 - V5 (GT) y = 0.3367x

0 Figure 2.5:

2000

4000 engine speed [min-1]

6000

0.5512

8000

Heat transfer coefficient for the inner surface of the catalytic converters against the exhaust gas, dependent on the rotational speed of the engine, implemented for the static characteristic model of the internal combustion engines.

2 Modelling Methods

18

2.2

Electric Drive Systems

For the electric drive systems three models are created. For the two electric motor types equivalent circuit models based on differential equations are used. Both motor models are based on the induction law:

ܷ ൌ ܴ‫ ܫ‬൅ 

݀ߖ  ݀‫ݐ‬

‡“ǤʹǤ͵

This equation is applied on each phase of the 3-phase AC currents of the motors stator and for the induction motor also on the rotor. The equations are afterwards converted into equivalent circuit models by coordinate transformation as shown in [22]. For each motor type a different coordinate system is used and will be introduced in the following section. The power electronics model uses a simple model approach, calculating switching and conducting losses of each transistor.

2.2.1

Permanent Magnet Synchronous Motor

The permanent magnet synchronous motor (PMSM) model is based on the equivalent circuit model in the rotor flux fixed F2 or dq coordinate system, as it can be retrieved from [22]. First, the model transfers the 3-phase alternating current into a 2-phase stator fixed alternating current (S coordinate system), using the so-called Clarke transformation. In a second step, the 2-phase stator fixed equations are transferred into a 2-phase rotor flux fixed coordinate system (F2), which rotates with the angular frequency ߱ிଶ of the rotor flux field. This transformation results into direct current values for the equations. This coordinate transformation is called Park transformation. In the following section all values for current I, voltage U and flux linkage Ψ are in space-vector notation. The resulting set of equations for the permanent magnet synchronous motor can be written as:

ܷௗ ൌ ሺܴଵ ൅ ܴி௘ ሻ‫ܫ‬ௗ ൅

݀ߖௗ െ ߱ிଶ ߖ௤  ݀‫ݐ‬

‡“ǤʹǤͶ

2.2 Electric Drive Systems

19

ܷ௤ ൌ ሺܴଵ ൅ ܴி௘ ሻ‫ܫ‬௤ ൅

݀ߖ௤ ൅ ߱ிଶ ߖௗ  ݀‫ݐ‬

‡“ǤʹǤͷ

ߖௗ ൌ ߖ௉ெ ൅ ‫ܮ‬ௗ ‫ܫ‬ௗ 

‡“ǤʹǤ͸

ߖ௤ ൌ ‫ܮ‬௤ ‫ܫ‬௤ 

‡“ǤʹǤ͹

These equations describe the relation between voltage ܷ, current ‫ܫ‬, flux linkage ߖ and ohmic resistances ܴ as specified in eq. 2.3. The “d” and “q” indices describe the direction of the related values, “1” describes the value for one electric stator phase and “Fe” the overall value for the iron losses. For the angular velocity of the coordinate system the following relations apply:

߱ிଶ ൌ ‫ݖ‬௣ ߱௠௘௖௛Ǥ ؆  ߱ଵ

‡“ǤʹǤͺ

Where ߱ଵ is the angular velocity of the 3-phase alternating current of the stator, ‫ݖ‬௣ the number of pole pairs of the motor and ߱௠௘௖௛Ǥ the mechanical angular velocity of the rotor. For steady state operating points ߱ிଶ equals ߱ଵ .For instationary operating points ߱ଵ may differ from the value of ߱ிଶ due to the change of the electric rotor displacement angle ϑ. In Figure 2.6 the equivalent circuit model for both axes of the F2 coordinate system can be seen. For the further modeling process the equivalent circuit model is not needed. Only the formula relationships (eq. 2.4 – eq. 2.7) are used. In Figure 2.7 the most important motor parameters during motor operation of the permanent magnet synchronous motor are shown. The electromagnetic air gap torque for 3-phase synchronous motors is defined in [22] and can be written as:

2 Modelling Methods

20

͵ ‫ܯ‬௠௜ ൌ ‫ݖ‬௣ ሺߖௗ ‫ܫ‬௤ ൅ ߖ௤ ‫ܫ‬ௗ ሻ ʹ

‡“ǤʹǤͻ

When the equations for the flux linkage are inserted into eq. 3.9, the results for the permanent magnet synchronous motor can be written as: ͵ ‫ܯ‬௠௜ ൌ ‫ݖ‬௣ ൫ߖ௉ெ ‫ܫ‬௤ ൅ ሺ‫ܮ‬ௗ െ ‫ܮ‬௤ ሻ‫ܫ‬ௗ ‫ܫ‬௤ ൯ ʹ

‫ܫ‬ௗ

ܴଵ

‡“ǤʹǤͳͲ

ܴி௘ ݀ߖௗ ݀‫ݐ‬

൅

െ

൅ െ

ܷௗ ߱ிଶ ߖ௤

‫ܫ‬௤

ܴଵ

ܴி௘ ݀ߖ௤ ݀‫ݐ‬

൅

െ

൅ െ

ܷ௤ ߱ிଶ ߖௗ

Figure 2.6:

Equivalent circuit model of the permanent magnet synchronous motor in the F2 coordinate system, with d-axis (top) and q-axis (bottom).

2.2 Electric Drive Systems

21

ሬሬԦௗ ߖ

ሬሬԦଵ ߖ d

߱ிଶ

ߴ

ሬԦ௣ ܷ

q ߴ

‫ܫ‬Ԧଵ

ϭϮ͕Ϯ

߮ଵ ሬԦଵ ܷ

Figure 2.7:

Schematic sketch of a permanent magnet synchronous motor with one pole pair (ɘ ʹൌɘ‡…ŠǤ), showing important motor parameters in the F2 rotor fixed coordinate system in steady state motor operation (ɘ ʹൌɘͳሻǤ

The angular velocity of the F2 coordinate system can be calculated by integrating the angular acceleration of the rotor.

߱ிଶ ൌ ‫ݖ‬௣ න

ͳ ሺ‫ ܯ‬െ ‫ܯ‬௠௥ ሻ ݀‫ݐ‬ ‫ܬ‬௅ ௠௜

‡“ǤʹǤͳͳ

2 Modelling Methods

22

‫ܯ‬௠௥ is the motors friction torque and ‫ܬ‬௅ the rotating inertia of the connected load. The value for the iron loss resistance is not constant, due to the used iron loss model. The value for the iron loss resistance in the current load point can be calculated as follows:

ܴி௘ ൌ

ʹ ܲி௘  ͵ ൫‫ܫ‬ௗଶ ൅ ‫ܫ‬௤ଶ ൯

‡“ǤʹǤͳʹ

The model for the power loss due to iron losses is based on the model developed by Bertotti [23] and can be written as:

ܲி௘ ൌ ͲǤͶ ‫ܲ כ‬ி௘ǡ௥௘௙Ǥ ቆ

݊௠௘௖௛Ǥ

ଵǤହ

݊௠௘௖௛Ǥ ൅ͲǤʹ ‫ܲ כ‬ி௘ǡ௥௘௙Ǥ ቆ ቇ ݊௠௘௖௛Ǥǡ௥௘௙Ǥ ൅ͲǤͶ ‫ܲ כ‬ி௘ǡ௥௘௙Ǥ ቆ

ଶ

ଶ

ߖ௛ ቇ ቆ ቇ ݊௠௘௖௛Ǥǡ௥௘௙Ǥ ߖ௛ǡ௥௘௙Ǥ

ଵǤହ

ߖ௛ ቆ ቇ ߖ௛ǡ௥௘௙Ǥ

‡“ǤʹǤͳ͵

݊௠௘௖௛Ǥ

ߖ௛ ቇቆ ቇ ݊௠௘௖௛Ǥǡ௥௘௙Ǥ ߖ௛ǡ௥௘௙Ǥ

The distribution of the different losses was achieved by adjusting the reference motor to a measured characteristic map of the motor published in [24]. The development and fitting of the reference motor for the permanent magnet synchronous motor will be explained later in this subsection. The iron losses can be divided into three types: x

x

Eddy losses: After fitting the losses to the reference motor, the eddy losses have a total share of 40 % on the overall iron losses in the reference operating point ൫͵ିଵ ݊௠௘௖௛Ǥǡ௠௔௫Ǥ ȁߖ௛ǡ௠௔௫Ǥ ൯ of the motor. The eddy losses are quadratic depended on the rotational speed of the motor ݊௠௘௖௛Ǥ , or more precisely on the angular speed of the stators electric field ߱ଵ , and also quadratic dependent on the main flux linkage of the motor ߖ௛ . Excess losses: The excess losses have a total share of 20 % on the overall iron losses in the reference operating point. The power of the excess losses

2.2 Electric Drive Systems

x

23

depends on the rotational speed of the motor ݊௠௘௖௛Ǥ and its main flux linkage ߖ௛ǡே by a factor of 1.5. Hysteresis losses: The hysteresis losses have a total share of 40 % on the iron losses in the nominal operating point. The power of the hysteresis losses depends linear on the rotational speed of the motor ݊௠௘௖௛Ǥ and its main flux linkage ߖ௛ǡே .

For the definition of the iron losses only one reference point ܲி௘ǡ௥௘௙Ǥ is needed. Using this point and the reference points for rotational speed and the flux linkage of the motor, the iron loss model automatically scales the iron losses to the corresponding operating point. The mechanical rotational speed of the permanent magnet synchronous motor is calculated as:

݊௠௘௖௛Ǥ ൌ

͸Ͳ ߱  ʹߨ ௠௘௖௛Ǥ

‡“ǤʹǤͳͶ

The main flux linkage is described as: ଶ

ߖ௛ ൌ ቤටሺߖ௉ெ ൅ ‫ܮ‬ௗ ‫ܫ‬ௗ ሻଶ ൅ ൫‫ܮ‬௤ ‫ܫ‬௤ ൯ ቤ

‡“ǤʹǤͳͷ

The flux linkage for the reference operating point, respectively the maximum flux linkage ߖ௛ǡ௠௔௫ can be written as: ଶ

ଶ ߖ௛ǡ௥௘௙Ǥ ൌ ቤටߖ௉ெ ൅ ൫‫ܮ‬௤ ‫ܫ‬ଵǡ௠௔௫Ǥ ൯ ቤ

‡“ǤʹǤͳ͸

The permanent magnet synchronous motor uses a 3-phase power electronics module with impressed voltage to modulate amplitude and frequency of the machine voltage ܷ௔ , ܷ௕ and ܷ௖ . The signal flow chart of the complete torque controlled electric drive system for the permanent magnet synchronous motor can be seen in Figure 2.8. It consists of a motor control, the coordinate transformation as well as the power electronics module and the permanent magnet synchronous motor.

2 Modelling Methods

24

DŽƚŽƌŽŶƚƌŽů ܷௗǡ௧௔௥௚Ǥ ܷ௤ǡ௧௔௥௚Ǥ

‫ܫ‬ௗǡ௧௔௥௚Ǥ ‫ܫ‬௤ǡ௧௔௥௚Ǥ

ƵƌƌĞŶƚ ŽŶƚƌŽů

‫ܫ‬ௗ ‫ܫ‬௤

‫ܯ‬௧௔௥௚Ǥ

dŽƌƋƵĞ ŽŶƚƌŽů

WĂƌŬdƌĂŶƐĨŽƌͲ ŵĂƚŝŽŶ

ůĂƌŬĞdƌĂŶƐĨŽƌͲ ŵĂƚŝŽŶ /ŶǀĞƌƐĞWĂƌŬ dƌĂŶƐĨŽƌŵĂƚŝŽŶ

߱௠௘௖௛Ǥ ‫ܫ‬௔ ǡ ‫ܫ‬௕ ǡ ‫ܫ‬௖

/ŶǀĞƌƐĞůĂƌŬĞ dƌĂŶƐĨŽƌŵĂƚŝŽŶ

Figure 2.8:

ܷ௔ǡ௧௔௥௚Ǥ ܷ௕ǡ௧௔௥௚Ǥ ܷ௖ǡ௧௔௥௚Ǥ

WŽǁĞƌůĞĐƚƌŽͲ ŶŝĐƐ

WD^D ϯΕ ܷ௔ ܷ௕ ܷ௖

Signal flow chart of torque controlled permanent magnet synchronous motor (PMSM) electric drive system with field-oriented control.

The motor control has two separate controllers: one torque controller that calculates the required motor current ‫ܫ‬ௗǡ௧௔௥௚Ǥ ǡ ‫ܫ‬௤ǡ௧௔௥௚Ǥ depending on the torque requirement ‫ܯ‬௧௔௥௚Ǥ and a current controller that compensates the spread between target and real current. In addition, the current controller calculates the voltage values ܷௗǡ௧௔௥௚Ǥ and ܷ௤ǡ௧௔௥௚Ǥ for the control of the power electronics module. The current controller works with eq. 2.4 – Eq.2.7, without the time-variable proportion of the flux linkage. The resulting equations describe the motor in steady state operation. The dependent part of the voltage on the rotational

2.2 Electric Drive Systems

25

motor torque M [Nm]

speed is calculated by a so-called decoupling network [22]. The target values for required motor currents ‫ܫ‬ௗǡ௧௔௥௚Ǥ ǡ ‫ܫ‬௤ǡ௧௔௥௚Ǥ need to be calculated optimally. For this purpose, different possibilities are known. The most efficient ones are the MTPA (Maximum Torque per Ampere), and for the field weakening area, a MTPA that slowly merges into MTPV (Maximum Torque per Volt). A schematic display of the different areas on the motor map can be seen in Figure 2.9.

base speed range MTPA

motor rotational speed n [min-1] Figure 2.9:

Schematic display of the different control areas of the PMSM motor map (1. quadrant).

The base speed range describes the area where the motor still has voltage reserves and the current can be controlled optimally. In the field weakening area the voltage limit for optimal current is reached. The permanent magnet field of the motor (flux linkage ߖ௉ெ ) needs to be weakened by raising the motor current ‫ܫ‬ௗ . This is continued until the torque cannot be realized anymore without extending the voltage limit of the motor (MTPV). Only one quadrant is shown in the figure, but the procedure is the same for the other three quadrants. The MTPA can be calculated with the torque and power equations of the motor by Lagrange multipliers. The merged area and the MTPV is calculated with a brute-force optimization. Nevertheless, the MTPA does not account for the iron losses and will deviate slightly from the optimal solution. Therefore, the

2 Modelling Methods

26

brute-force optimization is also used for the MTPA-area of the motor map. The phase current ‫ܫ‬ଵ over the distribution of motor currents ‫ܫ‬ௗ and ‫ܫ‬௤ shows a concave behavior. This results in only one minima of this function for a certain torque requirement at a certain rotational speed of the motor. With this knowledge, the brute-force optimization can calculate the optimal currents in each operating point of the motor map. On this basis, a synthetic motor is created by fitting the parameters of the differential equations eq. 2.4 – Eq.2.16 to the motor characteristic in [24] and adapting them to IE5-efficiency level from [25]. The synthetic motor has a rated output power of 85 kW (N) and is adapted to the rotational speed of an Otto engine. The design point (DP) for the efficiency is at 90 % of the maximum corner speed and 45 % of the maximum torque. The results for the efficiency map can be seen in Figure 2.10 and the motor parameters of the fitting in Table 2.3.

N

DP

Figure 2.10:

Efficiency characteristic of the synthetic reference permanent magnet synchronous motor with 85 kW rated power.

2.2 Electric Drive Systems Table 2.3:

27

Data sheet of the synthetic reference permanent magnet synchronous motor with 85 kW rated power.

General Data Rated power

ܲே

85

kW

Pole pair number

‫ݖ‬௣

4

-

Corner speed

݊௖௢௥Ǥ

2500

min-1

Maximum speed

݊௠௔௫Ǥ

6000

min-1

Maximum torque

‫ܯ‬௠௔௫Ǥ

330

Nm

Efficiency in DP

ߟ஽௉

96.56

%

Electric Data Maximum current (RMS)

‫ܫ‬ଵǡ௠௔௫Ǥ

400

A

Maximum voltage (RMS)

ܷଵǡ௠௔௫Ǥ

135

V

Stator resistance

ܴଵ

9

mΩ

Inductivity

‫ܮ‬ௗ

125

μH

Inductivity

‫ܮ‬௤

175

μH

Rotor flux linkage

ߖ௉ெ

95

mWb

Eddy losses in DP

ܲ௘ௗௗ௬ǡ஽௉

112

W

Hysteresis losses in DP

ܲ௛௬௦௧ǡ஽௉

136

W

Excess losses in DP

ܲ௘௫௖ǡ஽௉

62

W

Friction torque

‫ܯ‬௠௥

0.09

Nm

2 Modelling Methods

28

The results of the brute-force optimization of the synthetic motor are used for the simulation of the permanent magnet synchronous motor. These results are then exported into a MATLAB steady state model. This approach is required to keep simulation times for the virtual test bench at acceptable levels and still provide data with the necessary accuracy. Therefore, the assumption is made that the influence of dynamic load shifts can be neglected onto the characteristic maps of the permanent magnet synchronous motor. For the full longitudinal dynamic simulation of the vehicle and powertrain later on, the permanent magnet synchronous motor needs to be scaled to different powers and different maximum rotational speeds (e.g. Diesel engine). The scaling for the rotational speed is realized by keeping the rated power constant and decreasing the maximum rotational speed, resulting in an increase in maximum torque. This will require another set of motor parameters, but the results for the scaled efficiency map stay similar. The scaling of the efficiency as a function of the nominal power is done with the IE-5 function from [25]. The scaling factors referenced to the synthetic reference motor can be seen in Figure 2.11. 1.01 KPN,PMSM [-]

1 0.99 0.98 0.97 0.96 0.95 0

Figure 2.11:

25

50 75 100 125 Maximum output power PN [kW]

150

Scaling function for the characteristic efficiency map model as a function of the nominal power of the permanent magnet synchronous motor (IE5-efficiency class), referred to the 85 kW reference motor.

2.2 Electric Drive Systems

29

The rotating inertia of the permanent magnet synchronous motor is retrieved from [16] and calculated as follows:

‫ܬ‬௉ெௌெ ൌ ͷ ‫ ଻ିͲͳ כ‬

‰; ‫ܲ כ‬ே ൅ ͲǤͲͳ‰; ܹ

‡“ǤʹǤͳ͹

‫ܬ‬௉ெௌெ : Rotating inertia of the PMSM [kg m²] ܲே :

Rated power of the PMSM [W]

The mass of the permanent magnet synchronous motor is calculated with an equation retrieved from [18]. The resulting function is afterwards fitted to real machine data retrieved from [26], for permanent magnet synchronous motor drive systems. The resulting mass for the drive system calculates as follows:

݉௉ெௌெ ൌ ͲǤ͵͹͵ ‫ିͲͳ כ‬ଷ

݇݃ ‫ܲ  כ‬ே ൅ ͺǤͻ݇݃ ܹ

‡“ǤʹǤͳͺ

݉௉ெௌெ : Weight of the PMSM drive system [kg] ܲே : 2.2.2

Rated power of the PMSM [W]

Induction Motor

The induction motor (IM) model is based on the equivalent circuit transformer model in the stator field fixed K or AB coordinate system, as it can be retrieved from [22]. First, the model transfers the 3-phase alternating current into a 2phase stator fixed alternating current (S coordinate system), using the so-called Clarke transformation. In a second step, the 2-phase stator fixed equations are transferred into a 2-phase stator coordinate system (K), which rotates with the angular frequency of the 3-phase alternating current ߱ଵ . This transformation results into direct current values for the equations. In the following section all values for current I, voltage U and flux linkage Ψ are in space-vector notation. This coordinate transformation is referred to as Park transformation. The resulting set of equations for the induction motor can be written as:

2 Modelling Methods

30

ሬሬሬሬറ ܷଵ ൌ ܴଵ ሬሬറ ‫ܫ‬ଵ ൅ ݆߱ଵ ‫ܮ‬ఙଵ ሬሬറ ‫ܫ‬ଵ ൅

݆߱ଵ ‫ܮ‬௛ ܴி௘ ቀ‫ܫ‬ሬሬറ ൅ ‫ܫ‬ሬሬሬറଶᇱ ቁ ݆߱ଵ ‫ܮ‬௛ ൅ ܴி௘ ଵ

ሬሬሬሬറ ܷଶ ܴଶᇱ ሬሬሬሬറ ݆߱ଵ ‫ܮ‬௛ ܴி௘ ൌ ‫ ܫ‬Ԣ ൅ ݆߱ଵ ‫ܮ‬ᇱఙଶ ሬሬሬറ ‫ܫ‬ଶᇱ ൅ ቀ‫ܫ‬ሬሬറ ൅ ‫ܫ‬ሬሬሬറଶᇱ ቁ ൌ Ͳ ‫ݏ‬ ‫ ݏ‬ଶ ݆߱ଵ ‫ܮ‬௛ ൅ ܴி௘ ଵ

‡“ǤʹǤͳͻ

‡“ǤʹǤʹͲ

where ‫ ܮ‬is the inductivity, R the ohmic resistance, I the current, U the voltage and ‫ ݏ‬the induction motors current slip. All “ ’ ” values are transformed to the stator side. The index “h” labels the main field inductivity and “σ” the stray field inductivity of the stator (1) and the rotor (2). The rotor is a squirrel-cage rotor which results into a rotor voltage ሬሬሬሬറ ܷଶ of 0. For the angular velocity of the coordinate system the following relations apply:

߱௄ ൌ  ߱ଵ ൌ

߱ଶ ߱௠௘௖௛Ǥ ൌ ‫ ݖ‬ ሺͳ െ ‫ݏ‬ሻ ሺͳ െ ‫ݏ‬ሻ ௣

‡“ǤʹǤʹͳ

where ߱ଵ is the angular velocity of the 3-phase alternating current of the stator, ߱ଶ the angular velocity of the rotor field, ‫ݖ‬௣ the number of pole pairs of the motor and ߱௠௘௖௛Ǥ the mechanical angular velocity of the rotor. In Figure 2.12 the equivalent circuit model of the induction motor in the K coordinate system can be seen. For further investigations the equivalent circuit model is used. First, the torque characteristics of the induction motor must be considered in more detail. The characteristic torque over the slip can be seen in Figure 2.13. For positive slip values the induction motor works in motor operation, for negative slip values in generator operation. The operating range of the motor is between the breakdown torques (BP). The electromagnetic air gap torque is defined by the load on the motor and results in a slip value s. The electromagnetic air gap torque for 3-phase induction motors can be derived from the power transferred through the air gap and the mechanical output power [27] [28]. The power transferred through the air gap can be written as:

2.2 Electric Drive Systems

31

ܲఋ ൌ

ܴଵ

‫ܫ‬ଵ

͵ ܴଶᇱ ᇱ ‫; ܫ‬ ʹ ‫ ݏ‬ଶ

‡“ǤʹǤʹʹ

݆ɘଵ ‫ܮ‬ᇱ஢ଶ

݆ɘଵ ‫ܮ‬஢ଵ

ܴி௘

ܷଵ

Figure 2.12:

ܴଶᇱ ‫ݏ‬

‫ܫ‬ଶᇱ

௎మ ௦

݆ɘଵ ‫ܮ‬௛

сϬ

Equivalent circuit model of the induction motor in the K coordinate system. Motor

Generator

Mmi [Nm]

MBP

operating range 1

0.5

0

-0.5

- MBP slip s [-] Figure 2.13:

Schematic torque/slip characteristic of an induction motor.

-1

2 Modelling Methods

32

The mechanical output power now can be written from the electric side (eq. 2.23) of the rotor and from the mechanical side (eq. 2.24) of the machines air gap torque:

ܲ௠௜ ൌ ܲఋ െ ܲଶǡோమ ൌ

͵ ܴଶᇱ ᇱ ଶ ͵ ᇱ ᇱ ଶ ͵ ܴଶᇱ ᇱ ଶ ‫ܫ‬ଶ െ  ܴଶ ‫ܫ‬ଶ ൌ ሺͳ െ ‫ݏ‬ሻ ‫ ܫ‬ ʹ ‫ݏ‬ ʹ ʹ ‫ ݏ‬ଶ

ܲ௠௜ ൌ ‫ܯ‬௠௜ ߱௠௘௖௛Ǥ ൌ ‫ܯ‬௠௜ ߱ଵ

ሺͳ െ ‫ݏ‬ሻ  ‫ݖ‬௣

‡“ǤʹǤʹ͵

‡“ǤʹǤʹͶ

Form eq. 2.23 and 2.24 the air gap torque of the motor can be derived as:

‫ܯ‬௠௜ ൌ

ͳ ܴଶᇱ ᇱ ଶ ͵ ‫ݖ‬௣ ‫ ܫ‬ ʹ ߱ଵ ‫ ݏ‬ଶ

‡“ǤʹǤʹͷ

The angular velocity of K coordinate system can be calculated by integrating the angular acceleration of the rotor.

߱௄ ൌ ‫ݖ‬௣ න

ͳ ሺ‫ ܯ‬െ ‫ܯ‬௠௥ ሻ ݀‫ݐ‬ ‫ܬ‬௅ ௠௜

“ǤʹǤʹ͸

where ‫ܯ‬௠௥ is the motors friction torque and ‫ܬ‬௅ the rotating inertia of the connected load. The value for the iron loss resistance is not constant, due to the used iron loss model. For the induction motor, the same iron loss model as for the permanent magnet synchronous motor is used (eq. 2.12 and eq. 2.13). The distribution of the different iron losses shares was kept constant to the permanent magnet synchronous motor. The fitting of the iron losses to the reference motor will be done later in this subsection. The mechanical rotational speed of the induction motor is calculated as:

2.2 Electric Drive Systems

݊௠௘௖௛Ǥ ൌ

33

͸Ͳ ߱  ʹߨ ௠௘௖௛Ǥ

‡“ǤʹǤʹ͹

The main flux linkage is described as:

ߖ௛ ൌ ȁ‫ܮ‬௛ ሺ‫ܫ‬ଵ ൅ ‫ܫ‬ଶᇱ ሻȁ

‡“ǤʹǤʹͺ

The flux linkage for the reference operating point, respectively the maximum flux linkage ߖ௛ǡ௠௔௫ is written as:

ߖ௛ǡ௥௘௙Ǥ ൌ ห‫ܮ‬௛ ‫ܫ‬ଵǡ௠௔௫Ǥ ห

‡“ǤʹǤʹͻ

The induction motor uses a 3-phase power electronics module with impressed voltage to modulate amplitude and frequency of the machine voltage ܷ௔ , ܷ௕ and ܷ௖ . The signal flow chart of the complete torque controlled electric drive system is the same as for the permanent magnet synchronous motor which can be seen in Figure 2.8. Only the dq-coordinates are replaced by the AB-coordinates of the K coordinate system. The whole drive system for the induction motor consists of a motor control, the coordinate transformation as well as the power electronics module and the induction motor itself. The motor control has two separate controllers: one torque controller that calculates the required motor current ‫ܫ‬Ԧଵ depending on the torque requirement ‫ܯ‬௧௔௥௚Ǥ and a current controller that compensates the spread between target and real current. The required motor current for using a squirrel cage rotor is calculated as follows: ܷ௛ ܷ௛ ܷ௛ ሬሬሬԦᇱ ሬሬԦ ‫ܫ‬ଵ ൌ ሬሬሬሬሬԦ ‫ܫ‬ி௘ ൅ ‫ܫ‬ሬሬሬԦ ൅ ൅  ௛ ൅ ‫ܫ‬ଶ ൌ ܴி௘ ݆߱ଵ ‫ܮ‬௛ ܴଶᇱ ൅ ݆߱ଵ ‫ܮ‬ᇱఙଶ ‫ݏ‬ with the main field voltage:

‡“ǤʹǤ͵Ͳ

2 Modelling Methods

34

ܷ௛ ൌ อȁ‫ܫ‬ଶᇱ ȁ ቆ

ܴଶᇱ ൅ ݆߱ଵ ‫ܮ‬ᇱఙଶ ቇอ ‫ݏ‬

‡“ǤʹǤ͵ͳ

and the translated rotor current from the air gap torque (eq. 2.25):

ȁ‫ܫ‬ଶᇱ ȁ ൌ ඨ‫ܯ‬௧௔௥௚Ǥ

ʹ ߱ଵ ‫ݏ‬ ʹ ߱௠௘௖௛Ǥ ‫ݏ‬  ൌ ඨ‫ܯ‬௧௔௥௚Ǥ   ͵ ‫ݖ‬௣ ܴଶᇱ ͵ ሺͳ െ ‫ݏ‬ሻ ܴଶᇱ

‡“ǤʹǤ͵ʹ

In addition, the current controller calculates the voltage value ሬሬሬሬԦ ܷଵ for the control of the power electronics module:

ሬሬሬሬԦ ܷଵ ൌ ܷ௛ ൅ ሬሬԦ ‫ܫ‬ଵ ሺܴଵ ൅ ݆߱ଵ ‫ܮ‬ఙଵ ሻ

‡“ǤʹǤ͵͵

The target values for the required motor current ሬሬԦ ‫ܫ‬ଵ need to be calculated optimally. The same relations as for the permanent magnet synchronous motor appear. The MMPA in the bases speed range slowly devolves into the MMPV in the field weakening area, as already shown in Figure 2.9. Again, the optimization of ‫ܫ‬ሬሬԦଵ , is done by brute-force optimization. During the optimization the slip s is optimized until the minimum power consumption for a given torque is reached. On this basis, a synthetic motor is created by fitting the parameters of the preliminary equations in this subsection to the motor characteristic in [29]. Afterwards, the motor efficiency is adapted to IE3-efficiency level from [25]. The copper resistance of the stator is replaced by aluminum, which increases the stator resistance ܴଵ by the approximate factor of 1.567. For the rotor the aluminum windings do not have an impact on the resistance, due to possible compensation by design. The aluminum windings are chosen to achieve a cost and environmental impact optimal design of the induction motor. The synthetic motor has a rated output power of 150 kW (N) and a maximum rotational speed of 12000 min-1. The design point (DP) for the efficiency is at 90 % of the maximum corner speed and 45 % of the maximum torque. The results for

2.2 Electric Drive Systems

35

the efficiency map can be seen in Figure 2.14, and the motor parameters of the fitting in Table 2.4.

N

DP

Figure 2.14:

Efficiency characteristic of the synthetic reference induction motor with 150 kW rated power.

The results of the brute-force optimization of the synthetic motor are used for the simulation of the induction motor. These results are then exported into a MATLAB steady state model. This approach is required to keep simulation times for the virtual test bench at acceptable levels and still provide data with the necessary accuracy. For the full longitudinal dynamic simulation of the vehicle and powertrain later on, the induction motor needs to be scaled to different powers and different maximum rotational speeds. The scaling for the rotational speed is realized by keeping the rated power constant and decreasing the maximum rotational speed, resulting into an increase in maximum torque. This will require another set of motor parameters, but the results for the scaled efficiency map stay similar. The scaling of the power is done with the IE-3 function from [25]. The scaling factors referring to the synthetic reference motor can be seen in Figure 2.15.

2 Modelling Methods

36 Table 2.4:

Data sheet of the synthetic reference induction motor with 150 kW rated power.

General Data Rated power

ܲே

150

Pole pair number

‫ݖ‬௣

2

Corner speed

݊௖௢௥Ǥ

5300

min-1

Maximum speed

݊௠௔௫Ǥ

12000

min-1

Maximum torque

‫ܯ‬௠௔௫Ǥ

270

Efficiency in DP

ߟ஽௉

kW -

Nm

94.81

%

Electric Data Maximum current (RMS)

‫ܫ‬ଵǡ௠௔௫Ǥ

700

A

Maximum voltage (RMS)

ܷଵǡ௠௔௫Ǥ

135

V

Stator resistance

ܴଵ

7.8

mΩ

Rotor resistance

ܴଶᇱ

3.6

mΩ

Inductivity

‫ܮ‬௛

746

μH

Inductivity

‫ܮ‬ఙଵ

17

μH

Inductivity

‫ܮ‬ᇱఙଶ

32

μH

Eddy losses in DP

ܲ௘ௗௗ௬ǡ஽௉

64

W

Hysteresis losses in DP

ܲ௛௬௦௧ǡ஽௉

155

W

Excess losses in DP

ܲ௘௫௖ǡ஽௉

50

W

Friction torque

‫ܯ‬௠௥

0.65

Nm

2.2 Electric Drive Systems

37

1.01 1 KPN,IM [-]

0.99 0.98 0.97 0.96 0.95 0

Figure 2.15:

25

50 75 100 125 Maximum power output PN [kW]

150

Scaling function for the characteristic efficiency map model as a function of the nominal power of the induction motor (IE3efficiency class), referred to the 150 kW reference motor.

The rotating inertia of induction motor is retrieved from [16] and calculated as follows:

‫ܬ‬ூெ ൌ ͷ ‫ ଻ିͲͳ כ‬

݇݃݉; ‫ܲ כ‬ே ൅ ͲǤͲͳ݇݃݉; ܹ

‡“ǤʹǤ͵Ͷ

‫ܬ‬ூெ : Rotating inertia of the induction motor [kg m²] ܲே : Rated power of the induction motor [W] The mass of the induction motor is calculated with an equation retrieved from [18]. The resulting function is afterwards fitted to real machine data retrieved from [30] [31] [32], for induction motor drive systems. The resulting mass for the drive system calculates as follows:

݉ூெ ൌ ͲǤ͵Ͷ ‫ିͲͳ כ‬ଷ

݇݃ ‫ܲ  כ‬ே ൅ ʹͳǤͺ݇݃ ܹ

݉ூெ :

Weight of the induction motor drive system [kg]

ܲே :

Rated power of the induction motor [W]

‡“ǤʹǤ͵ͷ

2 Modelling Methods

38 2.2.3

Power Electronics

Figure 2.16:

DŽƚŽƌĐŽŶŶĞĐƚŝŽŶ

ĂƚƚĞƌLJĐŽŶŶĞĐƚŝŽŶ

For the power electronics module of the electric drive systems, a B6c-bridge inverter is used. It consists of six complementary switching transistors with counter parallel diode and one intermediate circuit capacitor for smoothing the input voltage. The inverter is controlled with pulse-width modulation using a switching frequency of 10 kHz. The pole pair number ‫ݖ‬௣ of the electric motors is adapted to this frequency to deliver an acceptable sinusoidal voltage signal to the motor at maximum motor speed. For this model silicon carbide metaloxide-semiconductor field-effect transistors with a maximum drain voltage of 1200 V are used. In Figure 2.16 a schematical sketch of the power electronics module (B6c) without control electronics can be seen.

Schematical sketch of a B6c-bridge inverter with battery connection on the left and motor connection on the right.

In [33] the power loss of the B6c-bridge inverter is defined as follows:

ܲ௉ாǡ௟௢௦௦ ൌ ͸൫ܲ௉ாǡ௖௢ǡ்௥Ǥǡ௢௡ ൅ ܲ௉ாǡ௖௢ǡ஽௜Ǥǡ௢௡ ൅ ܲ௉ாǡ௦௪ǡ்௥Ǥ ൅ ܲ௉ாǡ௦௪ǡ஽௜Ǥ ൯

ܲ௉ாǡ௖௢ǡ்௥Ǥǡ௢௡ :

Conduction losses of transistor [W]

ܲ௉ாǡ௖௢ǡ஽௜Ǥǡ௢௡ :

Conduction losses of diode [W]

ܲ௉ாǡ௦௪ǡ்௥Ǥ :

Switching losses of transistor [W]

ܲ௉ாǡ௦௪ǡ஽௜Ǥ :

Switching losses of diode [W]

‡“ǤʹǤ͵͸

2.2 Electric Drive Systems

39

For the loss model, ten datasheets of silicon carbide metal-oxide-semiconductor field-effect transistors from the manufactures Infineon and Semikron are evaluated. Due to eq. 2.36 two loss mechanisms are considered, the conduction losses and the switching losses. The reverse power losses as well as the power consumption for the power electronics control are neglected. The conduction losses appear when a current runs through a transistor or diode. The conduction losses are ohmic losses. In Figure 2.17 a schematic distribution between transistor and diode for carrying the current can be seen. If voltage and current have the same indicator (++ / --) the current flows through the transistors. If voltage and current have the opposite indicator (+- / -+) the diodes carry the current.

U [V], I [A]

transistors are current carrying diodes are current carrying

0

time [ms] Figure 2.17:

Schematic distribution of conduction losses between transistors and diodes, example for one phase.

The loss model for the conducting losses of the transistors and the diodes are fitted by plotting the conduction losses over the maximum root mean squared (RMS) drain current ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ for two temperatures, 25 °C and 150 °C. For all operating points the junction temperature of the transistors and diodes is set to ܶ௉ாǡ௃ ൌ ͳʹͷι. The conduction loss power for one transistor calculates as follows:

2 Modelling Methods

40

ଶ

‡“ǤʹǤ͵͹

ܲ௉ாǡ௖௢ǡ்௥Ǥǡ௢௡ ൌ  ൫‫ܫ‬ଵǡோெௌ …‘• ߮൯ ܴ௉ாǡ௖௢ǡ்௥Ǥǡ௢௡ 

‫ܫ‬ଵǡோெௌ ǣ    RMS current in current operating point [A] ߮:

Electrical phase shift between voltage and current [°]

ܴ௉ாǡ௖௢ǡ்௥Ǥǡ௢௡ : Resistance of one transistor [Ω] For the resistance of one transistor ܴ௉ாǡ்௥Ǥǡ௢௡ , the resistance values for different sizes of the transistors from the datasheets are plotted over their maximum drain current ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ . Subsequently the data points are fitted to a function for each of the two temperatures as can be seen in Figure 2.18. 90

25 °C 100 °C

y = 1697x-0.952

80 RPE,c,Tr.,on [mΩ]

70 60 50 40 30 20

y = 1155x-0.964

10 0 0

Figure 2.18:

100

200

300 IPE,D,max. [A]

400

500

600

Plot and fit of the transistor resistance for silicon carbide metal-oxide-semiconductor field-effect transistors from Infineon and Semikron, for two different junction temperatures.

With the knowledge of the fit functions for the transistor resistance, depending on temperature and size, the following function can be derived:

2.2 Electric Drive Systems

ି଴Ǥଽ଺ସ ܴ௉ாǡ௖௢ǡ்௥Ǥǡ௢௡ ൌ ͳͳͷͷ ቆ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ ൅

41

ି଴Ǥଽହଶ ି଴Ǥଽ଺ସ ͳǤͶ͸ͺ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ െ  ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ ܶ௉ாǡ௃ ቇ ͳʹͷι‫ܥ‬

‡“ǤʹǤ͵ͺ

‫ିͲͳ כ‬ଷ ߗ

The maximum drain current ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ is calculated with the maximum RMS current of the electric motor ‫ܫ‬ଵǡோெௌǡ௠௔௫Ǥ as well as the current electric phase shift of that operating point …‘• ߮൫‫ܫ‬ଵǡோெௌǡ௠௔௫Ǥ ൯. The maximum drain current is assumed 25 % higher than needed, to have a certain safety factor, resulting in the following equation:

‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ ൌ ͳǤʹͷ‫ܫ‬ଵǡோெௌǡ௠௔௫Ǥ …‘• ߮൫‫ܫ‬ଵǡோெௌǡ௠௔௫Ǥ ൯

‡“ǤʹǤ͵ͻ

For the diodes only the data from Infineon is evaluated, due to the more optimized conduction losses compared to the diode design from Semikron. The conduction losses for the diodes are calculated as follows: ଶ

ܲ௉ாǡ௖௢ǡ஽௜Ǥǡ௢௡ ൌ  ൣ‫ܫ‬ଵǡோெௌ ሺͳ െ …‘• ߮ሻ൧ ൬ͶǤ͸ െ

ͶǤ͵ െ ͶǤ͸ ܶ ൰ ‫ିͲͳ כ‬ଷ ߗ ͳʹͷι‫ ܥ‬௉ாǡ௃

‡“ǤʹǤͶͲ

The switching losses occur each time a transistor is switched on or off. In addition the reverse recovery losses of the diodes are taken into account. The reverse recovery losses occur when the diodes are switched on. The switching losses for the off-switching of the diodes are neglected. It is assumed that in each switching cycle with the frequency ݂௉ாǡ௉ௐெ ൌ ͳͲœ all three components of the switching losses occurring depending on the current RMS current ‫ܫ‬ଵǡோெௌ and the current electric phase shift ߮. The assumptions are similar to the ones for the conduction losses, as already illustrated in Figure 2.17. For the modelling of the switching losses, a linear fit of the switching losses dependent on the current RMS current and the maximum RMS drain current ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ is assumed. The power loss for the switching of the transistors can be written as:

2 Modelling Methods

42

ܲ௉ாǡ௦௪ǡ்௥Ǥ ൌ  ݂௉ாǡ௉ௐெ ൣܽ௉ாǡ்௥Ǥǡ௢௡ ൫‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ ൯‫ܫ‬ଵǡோெௌ …‘• ߮ ൅ ܾ௉ாǡ்௥Ǥǡ௢௡ ൅  ܽ௉ாǡ்௥Ǥǡ௢௙௙ ൫‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ ൯‫ܫ‬ଵǡோெௌ …‘• ߮ ൅ ܾ௉ாǡ்௥Ǥǡ௢௙௙ ൧

ܽ௉ாǡ்௥Ǥǡ௢௡ , ܽ௉ாǡ்௥Ǥǡ௢௙௙ :

‡“ǤʹǤͶͳ

Line gradient for linear transistor switching loss ௃

model (on/off) ቂ ቃ ஺

ܾ௉ாǡ்௥Ǥǡ௢௡ , ܾ௉ாǡ்௥Ǥǡ௢௙௙ :

Zero line offset for linear transistor switching loss model (on/off) [J]

The switching losses for the diodes are written as:

ܲ௉ாǡ௦௪ǡ஽௜Ǥ ൌ  ݂௉ாǡ௉ௐெ ൣܽ௉ாǡ஽௜Ǥǡ௥௥ ൫‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ ൯‫ܫ‬ଵǡோெௌ ሺͳ െ …‘• ߮ሻ ൅ ܾ௉ாǡ஽௜Ǥǡ௥௥ ൧

ܽ௉ாǡ஽௜Ǥǡ௥௥ :

‡“ǤʹǤͶʹ

Line gradient for linear diode switching loss model ௃

(reverse recovery) ቂ ቃ ஺

ܾ௉ாǡ஽௜Ǥǡ௥௥ :

Zero line offset for linear diode switching loss model (reverse recovery) [J]

The switching losses are modelled without a temperature dependence, due to the lack of temperature dependent data in the data sheets from Infineon and Semikron. The switching losses are modelled for a junction temperature of ܶ௉ாǡ௃ ൌ ͳʹͷι‫ܥ‬. Some data is only available for ܶ௉ாǡ௃ ൌ ͳͷͲι‫ܥ‬, but due to the available data from Infineon, no change in the dissipated energy is assumed between these two temperatures. For the modelling of the switching losses, ten different transistors and five different diodes are evaluated for fitting the line gradients and the zero line offset in dependence of the maximum drain current ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ . The results can be seen in Figure 2.19, with transistor on-switching losses (top), transistor off-switching losses (mid) and diode reverse recovery losses (bottom). The zero line offset of the off-switching losses is 0 and therefore not shown in the figure. With the results of the parameter fitting, the values for linear switching loss model can be calculated by using the maximum drain current of the power electronics module ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ . In Figure 2.20 the different switching losses over the current drain current of the transistor

2.2 Electric Drive Systems

43

0.05 0.04 0.03 0.02 0.01 0

1.5

ܽܲ‫ܧ‬,ܶ‫ݎ‬.,‫݋‬ff [mJ / A]

y = 0.0018x + 0.0373

1 0.5 0

0

200 400 ‫ܧܲܫ‬,‫ܦ‬,݉ܽ‫ݔ‬. [A]

600

0

200 400 ‫ܧܲܫ‬,‫ܦ‬,݉ܽ‫ݔ‬. [A]

600

0.06 y = 8E-05x + 0.0027

0.04 0.02 0 0

ܽܲ‫ܧ‬,Di.,rr [mJ / A]

bܲ‫ܧ‬,ܶ‫ݎ‬.,‫[ ݊݋‬mJ]

y = 6E-05x + 0.0065

200 400 ‫ܧܲܫ‬,‫ܦ‬,݉ܽ‫ݔ‬. [A]

600

0.02

0.3

0.015

bܲ‫ܧ‬,Di.,rr [mJ]

ܽܲ‫ܧ‬,ܶ‫ݎ‬.,‫[ ݊݋‬mJ / A]

‫ܫ‬ଵǡோெௌǡ்௥Ǥ and the diode ‫ܫ‬ଵǡோெௌǡ஽௜Ǥ for an exemplary module with 500 A of maximum drain current ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ can be seen.

y = 6E-06x + 0.0066

0.01 0.005

0.2 0.1 y = 0.0003x + 0.0762

0

0 0

Figure 2.19:

200 400 ‫ܧܲܫ‬,‫ܦ‬,݉ܽ‫ݔ‬. [A]

600

0

200 400 ‫ܧܲܫ‬,‫ܦ‬,݉ܽ‫ݔ‬. [A]

600

Results for the parameter fitting of the linear switching loss model for silicon carbide metal-oxide-semiconductor field-effect transistors in dependency of the modules maximum drain current, with transistor on-switching losses (top), transistor off-switching losses (mid) and diode reverse recovery losses (bottom).

2 Modelling Methods

44

25 E_on [mJ] E_off [mJ]

20 power loss [mJ]

E_rr [mJ] 15 10 5 0 0

Figure 2.20:

100

200 300 I1,RMS,Tr. [A]; I1,RMS,Di. [A]

400

500

Switching losses for transistor (E_on, E_off) and diode (E_rr) for a 400 A example module over the current drain current of transistor ͳǡǡ”Ǥ and diode ͳǡǡ‹Ǥ.

Based on the introduced power loss equations and the motor data generated in section 2.2.1 and section 2.2.2 a synthetic power electronics loss model for each reference motor is created. The efficiency map for fitting the loss model to the 85 kW permanent magnet synchronous reference motor can be seen in Figure 2.21. The results of the power electronics efficiency mapping are exported into a MATLAB steady state model. This approach is required to keep simulation times for the virtual test bench at acceptable levels and still provide data with the necessary accuracy. For the full longitudinal dynamic simulation of the vehicle and powertrain later on, the power electronics efficiency needs to be scaled to different motor characteristics. This is done by scaling to different maximum drain currents, depending on the maximum voltage level of the battery, the maximum rated power of the related motor as well as the motors current characteristic. For that purpose, different permanent magnet synchronous motors are created resulting into maximum drain currents ‫ܫ‬௉ாǡ஽ǡ௠௔௫Ǥ between 20 A and 1000 A. Afterwards

2.2 Electric Drive Systems

45

a mean efficiency over the whole characteristic map is generated and referenced to a 400 A module. The resulting scaling function for the efficiency of the power electronics module in dependency of the maximum drain current can be seen in Figure 2.22. The model for the power electronics module does not contain additional models. The weight for the module is already included in the weight for the motors from section 2.2.1 and section 2.2.2.

KIPE,D,max.,PE [-]

Figure 2.21:

Efficiency characteristic of the synthetic reference power electronics module fitted to the 85 kW permanent magnet synchronous reference motor.

1.02 1 0.98 0.96 0

Figure 2.22:

200 400 600 800 Maximum drain current IPE,D,max. [A]

1000

Scaling function for the characteristic efficiency map model of the power electronics module, referred to a 400 A reference module and fitted for a permanent magnet synchronous motor.

2 Modelling Methods

46

2.3

Battery Cells

The battery cells are modelled by fitting literature data [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] to a newly developed excel model. The model only contains a loss model for the battery cells depending on voltage and current (C-rate). For the battery cells three different cell types are modelled, one high gravimetric energy density battery cell with low C-rates, a medium power battery with medium gravimetric energy density and C-rates, and a high power battery with low gravimetric energy density but high C-rates. In a first step, the material composition and general key data of the cells are defined by choosing suitable cell designs from [44]. The results of that definition can be seen in Table 2.5. Table 2.5: General assumptions for the battery cell models. Type

High Energy

Anode

Si0.2C0.8

C

Li4Ti5O12

NMC9.5.5

Li(Mn0.5Fe0.5)PO4

LiFePO4

Max. C-rates

±4 C

±15 C

±60 C

Lifetime [cycles]

500

1000

> 8000

Grav. energy density [Wh / kg]

300

220

140

Cathode

Medium Power

High Power

In a second step, the open circuit voltage of the battery half-cell materials against lithium, for the anode materials ܷை஼௏ǡ஺௡Ǥ ሺܱܵ‫ܥ‬ሻ as well as for the cathode materials ܷை஼௏ǡ஼௔௧Ǥ ሺܱܵ‫ܥ‬ሻ, are retrieved from the literature data. The results for the open circuit voltage of the different materials can be seen in Figure 2.23, with the anode materials (top) and the cathode materials (bottom). The open circuit voltage of the battery full cell is then calculated as follows:

2.3 Battery Cells

47

ܷை஼௏ ሺܱܵ‫ܥ‬ሻ ൌ  ܷை஼௏ǡ஼௔௧Ǥ ሺܱܵ‫ܥ‬ሻ െ ܷை஼௏ǡ஺௡Ǥ ሺܱܵ‫ܥ‬ሻ

‡“ǤʹǤͶ͵

ܷܱ‫ܸܥ‬,An. [V]

2.0 1.5 1.0

C Si(0.2)C(0.8) Li(4)Ti(5)O(12)

0.5 0.0 0

0.2

0.4

SOC [-]

0.6

0.8

1

ܷܱ‫ܸܥ‬,Cat. [V]

4.5 4.0 3.5 3.0

Li[Mn(0.5)Fe(0.5)]PO(4) NMC9.5.5 LiFePo(4)

2.5 2.0 0

Figure 2.23:

0.2

0.4

SOC [-]

0.6

0.8

1

Open circuit voltage for anode materials (top) and cathode materials (bottom).

For modelling the battery characteristics, four types of effects caused by the charge and discharge currents are defined: x x x x

Activation overvoltage Ohmic losses Kinetic and mass transport losses Kinetic and mass transport effects

First, these four effects are mathematically modelled with suitable equations to fit the battery characteristic for the different battery types. A fit for the voltage characteristic is defined, to take into account the influence of these effects

2 Modelling Methods

48

on the battery voltage. The influence of the activation overvoltage, the ohmic losses and the kinetic and mass transport losses on the battery voltage result into irreversible power losses that are dissipated into thermal energy in the battery cells. The influence of the other kinetic and mass transport effects are assumed to be fully reversible and only lead to a reduction in the usable battery capacity. The preliminary fit of the voltage characteristic of the battery due to the three loss types calculates as:

ܷ஻஺்ǡ௟௢௦௦ ൌ ܷை஼௏ െ ൬

ܷ஻஺்ǡ௔௖௧Ǥ ൅  ܷ஻஺்ǡ௢௛௠Ǥ ൅ ܷ஻஺்ǡ௞௜௡Ǥ ൰ ʹ

‡“ǤʹǤͶͶ

The activation overvoltage only consist of ܷ஻஺்ǡ௔௖௧Ǥ , describing the hysteresis between the open circuit voltage for charge and discharge. It results in a voltage offset of

௱௎ಳಲ೅ǡೌ೎೟Ǥ ଶ

, referred to the open circuit voltage of the battery cell.

For the ohmic losses the following equation is used:

ܷ஻஺்ǡை௛௠Ǥ ൌ  ‫ܫ‬஻஺் ൣܴ஻஺்ǡ௢௛௠Ǥǡ௦௧௔௧Ǥ ൅ ሺͳ െ ܱܵ‫ܥ‬ሻܴ஻஺்ǡ௢௛௠Ǥǡௗ௬௡Ǥ ൧

‫ܫ‬஻஺் :

‡“ǤʹǤͶͷ

Current battery cell current [A]

It consists, of a static share with the application parameter ܴ஻஺்ǡை௛௠Ǥǡ௦௧௔௧Ǥ and a SOC-dependent dynamic share with the application parameter ܴ஻஺்ǡை௛௠Ǥǡௗ௬௡Ǥ . For the kinetic and mass transport losses the following voltage equation is used:

ܷ஻஺்ǡ௞௜௡Ǥ ൌ  οܷ஻஺்ǡ௞௜௡Ǥǡ௦௧௔௧Ǥ ܱܵ‫ ܥ‬௔ ‫ܴܥ‬

ܽ:

Adaptive polynomial coefficient for application [-]

CR:

Current C-rate of the battery cell [1 / h]

‡“ǤʹǤͶ͸

With the knowledge of ܷ஻஺்ǡ௟௢௦௦Ǥ , the influence of the kinetic and mass transport effects on the batteries capacity is now taken into account. These are

2.3 Battery Cells

49

no losses but limitations in the usable capacity of the battery cells, due to reaction and mass transport kinetics. To calculate the reduction in capacity the following one-dimensional approach is used: ‫ܧ‬஻஺் ሺܱܵ‫ܥ‬ሻ ൌ  ‫ܧ‬஻஺் ሺܱܵ‫ ܥ‬േ ͲǤͲʹሻ േ  ܷ஻஺்ǡ௟௢௦௦ ሺܱܵ‫ܥ‬ሻ‫ܫ‬஻஺் ߂ܱܵ‫ܥ‬ሺͳ െ ݂ሺ‫ܴܥ‬ሻܾሻ

‡“ǤʹǤͶ͹

݂ሺ‫ܴܥ‬ሻ: Adaptive C-rate dependent function for application ܾ:

Adaptive coefficient for application [-]

For the equation a resolution of ߂ܱܵ‫ ܥ‬ൌ ʹΨ is used, starting from 100 % SOC to 0 % SOC for discharge (+) and from 0 % SOC to 100 % SOC for charge (-). The starting values for the capacity function ‫ܧ‬஻஺் ሺܱܵ‫ܥ‬ሻ are zero for discharge and the nominal capacity of the battery cell for charging. The adaptive C-rate function ݂ሺ‫ܴܥ‬ሻ is retrieved from the literature cell data mentioned in the beginning of this section. An example for fitting the adaptive CRate function for the Li(Mn0.5Fe0.5)PO4-cathode can be seen in Figure 2.24.

relative capacity [-]

1 0.8 y = -0.083ln(x) + 0.7781

0.6 0.4 0.2

LMFP@Graphite 0 0

Figure 2.24:

5

10 C-rate [1 / h]

15

20

Relative capacity of a Li(Mn0.5Fe0.5)PO4-cathode over the battery cells C-rate.

The relative capacity is defined as the ratio of usable capacity to nominal capacity. Next, all application parameters are manually fitted to reach the required cell characteristics [39] [41] [42] [43] for the three different cell types. In addition, the maximum and minimum cell voltage is defined to avoid deep

2 Modelling Methods

50

discharge and overvoltage of the battery cells. The resulting parameters for the model fit can be seen in Table 2.6. The result for the fit of overall cell voltage of the medium power battery cell is shown in Figure 2.25. Table 2.6:

Fitting parameters for the different battery cell types.

Type

High Energy

Medium Power

0.05

0.06

0.06

ܴ஻஺்ǡ௢௛௠Ǥǡ௦௧௔௧Ǥ [Ω]

7.2 10-4

2.5 10-4

3.43 10-5

ܴ஻஺்ǡ௢௛௠Ǥǡௗ௬௡Ǥ [Ω]

0

4.2 10-4

6.30 10-5

οܷ஻஺்ǡ௞௜௡Ǥǡ௦௧௔௧Ǥ [V]

0

0.05

0.10

ܽ [-]

0

0.01

0.085

0

-0.083 ln(CR) + 0.7781

(0.0008 CR³ 0.0889 CR² + 3.4323 CR + 1.4891) 10-2

1

8

8

ܷ஻஺்ǡ௠௔௫Ǥ [V]

4.15

4.35

2.30

ܷ஻஺்ǡ௠௜௡Ǥ [V]

2.70

2.70

1.00

ܷ஻஺்ǡ௔௖௧Ǥ [V]

݂ሺ‫ܴܥ‬ሻ [-]

ܾ [-]

High Power

For the full longitudinal dynamic simulation of the vehicle and powertrain later on, the battery cells use the characteristic maps as seen in Figure 2.25. The cell size can be easily defined by using the full battery capacity and the needed battery voltage to match the intermediate circuit voltage. The battery model does not have a scaling function. The weight of the cells can be calculated with the estimated gravimetric cell density in Table 2.5. For the whole battery system the estimated weight calculates as follows:

2.4 Fuel Cell System

51

݉஻஺் ൌ ͳǤʹʹ‫ܥ‬஻஺் 

ͳ  ݁௚௥௔௩Ǥ

‡“ǤʹǤͶͺ

‫ܥ‬஻஺் :

Nominal capacity of the battery system [Wh]

݁௚௥௔௩Ǥ :

Gravimetric energy density of the used battery cell type [Wh / kg]

Due to the lack of available data, the extra weight for all full battery systems is assumed by the factor of 1.22 referred to the cell weight.

cell voltage [V]

4.2 3.7 3.2

-15 C -1 C 5C

2.7 0 Figure 2.25:

2.4

0.2

0.4 SOC [-] 0.6

-10 C OCV 10 C

0.8

-5 C 1C 15 C

1

Voltage and current characteristics of the medium power battery cell using a C-anode and a Li(Mn0.5Fe0.5)PO4-cathode.

Fuel Cell System

For the fuel cell system a MATLAB/Simulink model is used that was developed by Entenmann and Reichenbacher [45]. The design and application of the fuel cell system is done in MATLAB/Simulink and GT-Power. Embedded into the fuel cell system model is a zero-dimensional cell model which was developed by Kamgang Nzengang [46]. The models are afterwards adapted to fit the fuel cell estimation for the year 2040 [47] [48]. In Figure 2.26 the general configuration of the fuel cell system with related components can be seen.

Silencer

Ambient

Dilution line

Turbine

PMSM

Water tank

Gas dryer

Compressor

Air filter

DC/DCconverter

Fuel Cell Stack

Water injection

Demoisturizer

Ejector

Low-pressure control valve

Diluter

Purge valve

Pressure reducer

Hydrogen tank

52 2 Modelling Methods

Figure 2.26: Overview over the fuel cell system configuration with the related components used [45].

2.4 Fuel Cell System

53

The setup can be split up in five functional lines that will be introduced in the following sub-sections. The five functional lines are the fuel cell stack, the air supply, the fuel supply, the electrical path and the control of the fuel cell stack. In the last sub-section the implementation into the full longitudinal dynamic simulation of the vehicle and powertrain, as well as the additional model parameters are described.

2.4.1

Fuel Cell Stack

The fuel cell stack contains the fuel cells as well as the bipolar plates and the cooling of the stack. The setup of the cell model can be seen in Figure 2.27. The fuel cell model is a zero-dimensional semi-physical model that calculates the cell voltage and the mass transport for the anode, cathode and the membrane electrode assembly (MEA). The polarization curve of the fuel cell model is composed of four well known loss mechanism: x x x x

Ohmic conduction losses Activation overvoltage Concentration overvoltage Hydrogen permeation

These loss mechanisms reduce the open circuit voltage given by the reversible voltage of the electrochemical reaction that results into 1.23 V, with the assumption that the water in the reaction is formed in liquid phase.

Figure 2.27:

Model structure of the fuel cell with sub-models.

2 Modelling Methods

54

The results for the fit of the polarization curve at 80 °C stack temperature and 1.5 bar cathode/anode pressure, supplied with air and direct hydrogen can be seen in Figure 2.28, where i is the current density of the cell per cm².

cell voltage [V]

1.0 0.8 0.6 0.4 0.2 0.0 0

Figure 2.28:

500

1000 1500 i [mA / cm²]

2000

2500

Polarization curve for a 2040 fuel cell design.

The fuel cell stack uses bipolar plates with multiple serpentines for distribution of air and hydrogen to the MEA. The bipolar plates also contain the cooling lines for temperature control of the fuel cell stack. The configuration of the fuel cell stack for 80 kW and 60 kW of maximum electrical output power can be seen in Table 2.7. Table 2.7:

Stack configuration for an 80 kW and a 60 kW fuel cell stack.

Description

Unit

Value

Value

80 kW

60 kW

Cell area

mm²

400

400

Cell number (serial)

-

180

220

Maximum voltage of the Stack

V

180

220

Operating temperature

°C

80

80

Number of gas channels in bipolar plates

-

5

5

Number of serpentines in bipolar plates

-

7

7

2.4 Fuel Cell System 2.4.2

55

Air Supply

The air supply has the task to deliver an adequate amount of air to the fuel cell stack. The stoichiometry ratio of the intake should have at least a value of ߣ ൌ 1.1 to achieve sufficient air supply for the electrochemical reaction. In some operating points this value has to be higher to prevent condensation in the exhaust air at the turbine of the turbocharger or even flooding of the cathode. In the following, the functions of the components in the air supply are described in more detail. Air Filter: The air filter has to be designed in a way that the separation rate is much higher than for a comparable filter of a combustion engine. This is justified by the fact that the small channels in the bipolar plates can easily clog and by that throttle the efficiency and the maximum power of the fuel cell system. In addition, big particles could damage the compressor blades of the turbocharger. The air filter is designed to minimize pressure losses during filtering [45]. Electrically assisted Turbocharger (Compressor/PMSM/Turbine): The electrically assisted turbocharger is chosen to reach maximum efficiency for the air supply of the fuel cell system. Nevertheless, the turbocharger needs to be properly designed according to different particularities for the usage in the fuel cell system: x

x

x

Due to lower exhaust air temperatures compared to a combustion engine, the usable enthalpy for the turbine gets lower, leading to the need of an electrical motor assistance for the compressor. The operation pressure of the fuel cell is strongly influenced by the operating points of compressor and turbine and thus by the proper design of the turbocharger. The stoichiometry ratio of the intake air is set to λ = 1.1 to minimize the compressor work, or to higher values, if condensation in the turbine occurs in certain operating points.

A first optimization for maximum system efficiency leads to a compressor diameter of 45 mm and a turbine wheel diameter of 27.5 mm [45]. Nevertheless, the efficiency of the turbine could be more optimal, but due to the marginal influence on the overall system efficiency a further optimization is not performed. The maximum rotational speed of the turbocharger is around 140,000 min-1. The turbocharger characteristic for the 60 kW fuel cell stack for two

2 Modelling Methods

56

different electrical output powers, 15 kW and 60 kW, can be seen in Figure 2.29, with compressor (top) and turbine (bottom). ηc [%] 0

2.0

60 kW

10

Pressure ratio [-]

1.8

20

15 kW

1.6

30 40

1.4

50 60

1.2

70 80

1.0 0

0.01

0.02

0.03

0.04

Reduced speed [rpm / K0.5]

Corrected mass flow rate [kg / s]

7500

ηT [%] 0 5

60 kW

15 kW

10 15 20

5000

25 30 2500

35

0

40 45 50

1.0

1.1

1.2

1.3

1.4

1.5

Pressure ratio [-]

Figure 2.29:

Efficiency maps of compressor (top) and turbine (bottom) for two operating points of the 60 kW fuel cell stack.

The permanent magnet synchronous motor (PMSM) at the shaft of the turbocharger delivers a maximum torque of 0.35 Nm and has a rated power of 5.5 kW.

2.4 Fuel Cell System

57

Water injection: For a precise and easy control of the humidity of the intake air a water injector is used. The water injector requires high pressure to reach a fine spray for better vaporisation of the water in the intake air. For that, the injector operates in short on/off cycles to always keep a high injection pressure and to evenly distribute the water to the intake air. The water injector is supplied by the water tank and a pump that delivers the required pressure. The water injection is controlled by the intake air humidity control [45]. Water tank: The water tank represents the central part of the water management of the fuel cell system. The water level of the water tank is controlled by the gas dryer that only produces as much water as necessary to keep the water tank on a certain level. In addition, water is delivered by the dehumidification of the fuel supply and taken out of the circuit by the water injector for intake air humidification Gas dryer: The gas dryer delivers the required amount of water to the water tank. For that, an air-to-air heat exchanger is used that partially dehumidifies the air exhausted by the turbocharger. The amount of air dehumidified is controlled by a throttle valve. Dilution line: In the dilution line the purged hydrogen from the fuel supply is fed to the exhaust air. The molar concentration of hydrogen is reduced below 4 % to avoid ignition of the purged hydrogen. Silencer: The silencer is designed for the absorption of sound energy in the medium and high frequency range. It dampens the noise caused by the turbochargers turbine [45].

2.4.3

Fuel Supply

The fuel supply delivers the fuel in the form of hydrogen to the anode. The fuel path is a closed loop except for purging processes. The composition of the gas in the anode side consists only of hydrogen, water and diffused nitrogen. Due to the concentration gradient between cathode and anode, nitrogen is slowly diffusing from the cathode to the anode side of the MEA. The nitrogen in the anode circuit needs to be purged regularly to avoid a drop in fuel cell performance. In the following the components of the fuel path are described in more detail.

58

2 Modelling Methods

Hydrogen Tank: The pressure of the hydrogen tank for light-duty vehicle application is set to 700 bar, to minimize the needed space for the tank in the vehicle. At 700 bar pressure and standard conditions, 25.23 l tank volume per kg hydrogen is needed [45]. Pressure Reducer: The pressure reducer reduces the pressure from the 700 bar tank system to 5 bar. The maximum pressure in the anode circuit is 3 bar, so that there is a sufficient pressure gradient between pressure reducer and anode circuit [45]. Low-pressure control valve: The low-pressure control valve precisely controls the anode pressure to always match the mean cathode pressure. This is done to minimize gas diffusion and mechanical stress in the MEA. The input pressure of the pressure control valve is 5 bar and the output pressure adapts to the requested anode pressure up to 3 bar. The mass flow is controlled by adapting the throttle diameter of the low-pressure control valve. The low-pressure control valve is controlled by the anode pressure control [45]. Ejector: The ejector or jet pump recirculates the hydrogen in the anode circuit and delivers the needed mass flow to compensate the consumed fuel in the anode circuit. The ejector is optimised in GT-Power to the conditions for use in the fuel cell system [45]. Demoisturizer: The passive water demoisturizer is installed to separate the water in the anode circuit, which is transported to the anode side by diffusion. The separated water is directed to the water tank, where it is used for humidification of the intake air. The water demoisturizer has a volume of 4.0 l [45]. Purge valve: The purge valve is used to purge nitrogen out of the anode circuit, when a molar nitrogen concentration of 3 % is reached. The purge process lasts until the molar nitrogen concentration is lowered to 1 %. During purge events, the pressure in the anode circuit is kept constant by the low-pressure control valve. The purged anode gas is then directed to a passive diluter. The purge valve is controlled by the anode purge control [45]. Diluter: A passive diluter conducts the purged hydrogen from the anode circuit into the dilution line. The diluter uses a throttle point to continually deliver the purged anode gas into the exhaust air. The diluter is designed to keep the molar hydrogen concentration in the exhaust air below 4 %, which corresponds to the lower flammability limit of hydrogen.

2.4 Fuel Cell System 2.4.4

59

Electrical Functionalities and Control

The fuel cell systems stack current is controlled by a DC/DC-converter that transforms the electric energy to a requested voltage level for the intermediate circuit of the powertrain. The DC/DC-converter is a buck-boost converter that uses silicon carbide metal-oxide-semiconductor field-effect transistors to control the output voltage. Due to more stable voltage conditions, the electrically assisted turbocharger collects its current out of the intermediate circuit of the vehicles powertrain [45]. The fuel cell system has a separate system controller and can be seen as a closed system. External inputs are for example the on/off/standby switch, the power request and the requested voltage level for the electric power output. These parameters are provided by the energy management system of the vehicle or by the powertrain controller. A schematic display of the control instances for the closed loop control of the fuel cell system can be seen in Figure 2.30.

Figure 2.30:

Closed loop control of the fuel cell system.

The power request controller passes on the feed forward values to the four main controllers of the fuel cell system. The first controller is the cathode pressure control. It controls the electrically assisted turbocharger and manages the airflow to provide the needed air for the cathode. Additionally, it provides safety functions for the turbocharger to avoid compressor stall. For the control

2 Modelling Methods

60

a PI-controller with feed forward is used. The second controller is the intake air humidity control. It controls the water injection, in a manner that the intake air has optimal humidity for humidification of the membrane but avoids condensation in the turbochargers turbine. For the intake air humidity control a PI-controller with feed forward is used as well. The third controller is the anode pressure control. It controls the low-pressure control valve in a manner that the anode pressure always follows the mean cathode pressure. The anode pressure control system works faster than the cathode pressure control which always assures a pressure equilibrium between anode and cathode. The anode pressure control, too, uses a PI-controller with feed forward. The fourth controller is the anode purge controller. It uses a two-point control that opens at 3 % molar fraction of nitrogen in the anode circuit and closes at 1 % molar fraction of nitrogen in the anode circuit.

2.4.5

Implementation in Simulation

For the full longitudinal dynamic simulation of the vehicle and powertrain later on, the fuel cell system model is reduced to characteristic curves for the overall system efficiency. For the resulting overall system efficiency the fuel cell system with controller is put on a virtual test bench. The fuel cell system works at 20 °C ambient temperature, 1.0 bar ambient pressure and a mean stack temperature of 80 °C. Each operating point is then measured for 1400 s to also include dynamic effects like the purging events. The efficiency is then calculated by setting the lower heating value of the consumed hydrogen in relation to the electric power output of the system. The electric energy consumption of the electrically assisted turbocharger is also included in the fuel cell system efficiency. The overall system efficiency for the 60 kW and 80 kW fuel cell system can be seen in Figure 2.31. The weight of the fuel cell system without tank is retrieved from [47] and results in 1.1 kg / kW. The hydrogen tank data retrieved from [49] result in a value of 0.74 kg / l. The fuel cell system does not have a thermal model included.

2.5 Transmission

61

70% system efficiency

60% 50% 40% 30% 20% 80 kW 60 kW

10% 0% 0

Figure 2.31:

2.5

20 40 60 electrical output power [kW]

80

Overall fuel cell system efficiency referred to the lower heating value of hydrogen

Transmission

For each gearbox type a generic model is created by fitting FKFS internal and literature gearbox data to a newly developed excel optimization model. Different models are introduced for the automated gearboxes, manual gearboxes and differential gearboxes as well as for the electronically controlled multiplate clutch. Each gearbox model contains a model for the gear-meshing losses, the friction torque and a temperature model.

2.5.1

Automated gearboxes

The efficiency model for the synthetic automated gearboxes is created by fitting measurement and literature data of six different gearboxes to the excel optimization model. The results of this fitting are presented in the following section. First, the general efficiency equation for the automated gearbox model can be written as:

2 Modelling Methods

62

ߟீ஻ ൌ

‫ீܯ‬஻ǡ௜௡ െ ‫ீܯ‬஻ǡ஽௥௔௚ ߟீ஻ǡ௦௬௡Ǥ ߟீ஻ǡ௔ௗௗǤ  ‫ீܯ‬஻ǡ௜௡

ߟீ஻ :

Current efficiency of the gearbox [-]

‫ீܯ‬஻ǡ௜௡ :

Input torque of the gearbox [Nm]

‡“ǤʹǤͶͻ

‫ீܯ‬஻ǡ஽௥௔௚ : Current drag torque of the gearbox [Nm] ߟீ஻ǡ௦௬௡Ǥ :

Synthetic gear meshing losses [-]

ߟீ஻ǡ௔ௗௗǤ :

Additional loss efficiency [-]

Due to good availability of data for the automated gearboxes, a generic model for the gear meshing efficiency ߟீ஻ǡ௦௬௡Ǥ is developed. For this, the gear meshing efficiencies ߟீெ from [50] [51] for five different gearboxes are retrieved. First, it is assumed, that for a gear ratio of 1.00 the gear meshing efficiency is 100 %. This point is defined as reference point 0. A gearbox does not necessarily have a reference point 0. This point only exists if a direct drive through of the gearbox is available (݅ ൌ ͳǤͲͲሻ. Second, the single direction gear number ܰௌ஽ is defined. The single direction gear number only has a gear number ܰௌ஽ ൌ Ͳ if a direct drive-through exists (݅ ൌ ͳǤͲͲ). If there is no direct drivethrough, the reference point ܰௌ஽ ൌ Ͳ is only theoretically between the two gear ratios closest to ݅ ൌ ͳǤͲͲ. The reference point ܰௌ஽ ൌ Ͳ is used for counting the single direction gear number ܰௌ஽ . Starting from the reference point ܰௌ஽ ൌ Ͳ, the single direction gear number is counted upward in both directions for ݅ ൏ ͳǤͲͲ (starting from the highest gear ratio) and ݅ ൐ ͳǤͲͲ (starting from the lowest gear ratio), resulting in only positive numbers for the single direction gear number ܰௌ஽ . Third, the single direction gear ratio is defined as follows:

݅ௌ஽ ൌ ൝

݅ௌ஽ ൌ ݅݅ ൒ ͳ ͳ  ݅ௌ஽ ൌ ݅ ൏ ͳ ݅

݅ௌ஽ :

Single direction gear ratio [-]

݅:

Gear ratio [-]

‡“ǤʹǤͷͲ

2.5 Transmission

63

With knowledge of the single direction gear ratio ݅ௌ஽ , a normalized efficiency factor is defined:

‫ீܭ‬஻ ൌ ሺͳ െ ߟீெ ሻ ‫݅ כ ͲͲͳ כ‬ௌ஽ 

‡“ǤʹǤͷͳ

‫ீܭ‬஻ :

Single direction gear ratio dependent efficiency factor [-]

ߟீெ :

Gear meshing efficiency [-]

The normalization is carried out to reach ‫ீܭ‬஻ values around 1. In Figure 3.16, ‫ீܭ‬஻ for each gear of the five examined gearboxes is plotted over the single direction gear number ܰௌ஽ , starting from 1, excluding the direct drive-through 0. Then a general fit for ‫ீܭ‬஻ over ܰௌ஽ is created, using a polynomial fit. For the polynomial fit, a quartic curve fit for the ‫ீܭ‬஻ values shows the best results. 2.50

KGB [-]

2.00 1.50 1.00 0.50 y = 0.0238x4 - 0.3265x3 + 1.4561x2 - 2.3207x + 1.9462 0.00 1

2

3

4

5

6

NSD [-] Figure 2.32:

Single direction gear ratio dependent efficiency factor   for five different gearboxes, plotted over the single direction gear number  and the resulting quartic polynomial fit.

2 Modelling Methods

64

After creating the polynomial fit ‫ீܭ‬஻ǡ௙௜௧ , eq. 2.50 and eq. 2.51 are inverted and extended with the influence of the number of planetary gear sets ‫ݖ‬௉ீௌ to calculate the synthetic gearbox efficiency:

ߟீ஻ǡ௦௬௡Ǥ

‫ீܭ‬஻ǡ௙௜௧ ሺௌ஽ ሻ ‫ ͳۓ‬െ ݅ ൐ ͳǤͲͲ ݅ ‫כ‬ ͳͲͲ ‫Ͳ כ‬Ǥʹͷ ‫ݖ כ‬௉ீௌ  ۖ ൌ ͳ ݅ ൌ ͳǤͲͲ ‫۔‬ ‫ீܭ‬஻ǡ௙௜௧ ሺௌ஽ ሻ݅ ۖ ͳെ ݅ ൏ ͳǤͲͲ ͳͲͲ ‫Ͳ כ‬Ǥʹͷ ‫ݖ  כ‬௉ீௌ ‫ە‬

‫ீܭ‬஻ :

Single direction gear ratio dependent efficiency factor [-]

ߟீெ :

Gear meshing efficiency [-]

‫ݖ‬௉ீௌ :

Number of planetary gear sets in the gearbox [-]

‡“ǤʹǤͷʹ

In Figure 2.33, the deviation between the gear meshing efficiency ߟீெ (REAL) [50] [51] and the synthetic gearbox efficiency ߟீ஻ǡ௦௬௡Ǥ (FIT) can be seen. The equation for ߟீ஻ǡ௦௬௡Ǥ is fitted in advance, due to the fact that the fitting of the gear meshing efficiency is based on different data than the fitting of gearbox drag torque and additional loss efficiency. The efficiency parameters for the automated gearbox model, ‫ீܯ‬஻ǡ஽௥௔௚ and ߟீ஻ǡ௔ௗௗǤ , are fitted simultaneously. These parameters are fitted to the measurement data of an 8-speed gearbox, with 700 Nm of maximum input torque, retrieved from a FKFS internal source. The gearbox drag torque model uses reference values for the maximum input torque as well as for the current input speed of the gearbox. The fit for the gearbox drag torque leads to the following equation:

‫ீܯ‬஻ǡ஽௥௔௚ ൌ ‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ ሺܶீ஻ ሻ

଴Ǥ଻ଶସଶ ‫ீܯ‬஻ǡ௠௔௫ ݊ீ஻ǡ௜௡Ǥ ቀ ݅ ି଴Ǥ଺଴ସ଻  ቁ ିଵ ͹ͲͲܰ݉ ͳͲͲͲ݉݅݊

‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ : Reference drag torque of the gearbox [Nm] ܶீ஻ :

Mean gearbox temperature [°C]

‫ீܯ‬஻ǡ௠௔௫ :

Maximum input torque of the gearbox [Nm]

݊ீ஻ǡ௜௡Ǥ :

Rotational speed of the gearbox input [min-1]

݅:

Gear ratio of current gear [-]

‡“ǤʹǤͷ͵

2.5 Transmission

Gearbox 1

1.00 ηGM [-]; ηGB,syn. [-]

1.00 ηGM [-]; ηGB,syn. [-]

65

0.98 0.96 REAL

0.94

FIT

0.98 0.96

1 2 3 4 5 6 7 8 9 gear number [-]

1 2 3 4 5 6 7 8 9 gear number [-]

ZF 9 HP ηGM [-]; ηGB,syn. [-]

ηGM [-]; ηGB,syn. [-]

0.96 REAL FIT

0.98 0.96 REAL FIT

0.94 0.92

0.92

1 2 3 4 5 6 7 8 9 gear number [-]

1 2 3 4 5 6 7 8 9 gear number [-]

ZF 6 HP

1.00 ηGM [-]; ηGB,syn. [-]

GM 9T50 E

1.00

0.98

0.94

REAL FIT

0.94 0.92

0.92

1.00

Benz 9G-Tronic

0.98 0.96 REAL FIT

0.94 0.92 1

Figure 2.33:

2 3 4 5 gear number [-]

6

Fit of the synthetic gearbox efficiency Ʉ ǡ•›Ǥ (FIT) compared to real gear meshing efficiency Ʉ  (REAL) [51] [50] for five different automated gearboxes.

2 Modelling Methods

66

MGB,Drag,ref. [Nm]

The temperature-dependent reference drag torque ‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ is assumed to be linearly dependent [52] from the temperature reaching its minimum value at 60 °C, as it can be seen in Figure 2.34. 8.0 7.0 6.0 5.0 4.0 3.0 -40 -30 -20 -10 0

Figure 2.34:

10 20 30 40 50 60 70 80 90 100 TGB [°C]

Temperature-dependent reference drag torque  ǡ†”ƒ‰ǡ”‡ˆǤ for automated gearboxes.

The equation for the fit of the additional loss efficiency uses the same approach as for the gearbox drag torque and can be written as:

ߟீ஻ǡ௔ௗௗǤ ൌߟீ஻ǡ௔ௗௗǤǡ௥௘௙Ǥ ሺܶீ஻ ሻቀ

௡ಸಳǡ೔೙Ǥ ଵ଴଴଴௠௜௡షభ

ቁ

଴Ǥ଴ଵସହ

݅ ି଴Ǥ଴଴ଷହ 

‡“ǤʹǤͷͶ

ߟீ஻ǡ௔ௗௗǤǡ௥௘௙Ǥ : Reference additional loss efficiency of the gearbox [-]

ηGB,add,ref. [-]

For the temperature-dependent reference additional loss efficiency ߟீ஻ǡ௔ௗௗǤǡ௥௘௙Ǥ the same assumptions as for the temperature-dependent reference drag torque ‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ are made. The results are shown in Figure 2.35. 0.95 0.94 0.93 0.92 0.91 0.90 -40 -30 -20 -10 0 Figure 2.35:

10 20 30 40 50 60 70 80 90 100 TGB [°C]

Temperature-dependent reference additional loss efficiency Ʉ ǡƒ††Ǥǡ”‡ˆǤ for automated gearboxes.

2.5 Transmission

67

100% 80% 60% 40% 20% 0%

Cumulative frequency

70 60 50 40 30 20 10 0 -30 -20 -10 -5 -4 -3 -2 -1 0 1 2 3 4 5 10 20 30

Frequency [-]

The fit of the gearbox drag torque ‫ீܯ‬஻ǡ஽௥௔௚ and the additional loss efficiency ߟீ஻ǡ௔ௗௗǤ are carried out for two temperatures, 40 °C and 60 °C, with 240 measuring points each. To validate the quality of the fit, two histogram plots for the deviation between model and measurement points are drawn up in Figure 2.36. One for ܶீ஻ = 40 °C (top) and one for ܶீ஻ = 60 °C (bottom).

100% 80% 60% 40% 20% 0%

Cumulative frequency

70 60 50 40 30 20 10 0 -30 -20 -10 -5 -4 -3 -2 -1 0 1 2 3 4 5 10 20 30

Frequency [-]

Deviation [%]

Deviation [%]

Figure 2.36:

Frequency and cumulative frequency of the deviation between the automated gearbox model and measurement data used for fitting of the model, with   = 40 °C (top) and   = 60 °C (bottom).

It can be seen that 80 % of the fitted points have a deviation of less than ±4 % compared to the measurement data. The created generic automated gearbox model is now used to create synthetic automated gearboxes for the simulation. The model is already usable for steady state modelling and is integrated into

2 Modelling Methods

68

the full longitudinal simulation of the vehicle. The scaling is realized with the maximum gearbox torque ‫ீܯ‬஻ǡ௠௔௫ at gearbox input. An example for the overall gearbox efficiency for different temperatures can be seen in Figure 2.37. The operation point for the efficiency mapping is at 200 Nm of input torque and 2500 min-1 input speed. The used gear ratio will be introduced in section 3.5 and can be checked in Table 3.3 (9-speed). 0.92 0.90

ηGB [-]

0.88 0.86 0.84 -40 °C 0 °C 40 °C

0.82

-20 °C 20 °C 60 °C - 100 °C

0.80 1

Figure 2.37:

2

3

4 5 6 gear number [-]

7

8

9

Gearbox efficiency for a synthetic automated gearbox at 200 Nm input torque and 2500 min-1 input speed for different gearbox temperatures over the gear number.

The rotating inertia of the automated gearboxes is retrieved from FKFS internal data and is assumed with 0.025 kg m² for positive torques and 0.020 kg m² for negative torques. The rotating inertia refers to the gearbox input. The mass for the 9-speed automated gearbox is fitted after data from [53], resulting in the following equation:

2.5 Transmission

݉ீ஻ ൌ ͲǤͲͷͷͶ

69

݇݃ ‫ܤܩܯ  כ‬ǡ݉ܽ‫ ݔ‬൅ ͸ͳǤͶͲ݇݃ ܰ݉

‡“ǤʹǤͷͷ

The mass for the 6-speed automated gearbox is fitted after data from [54], resulting in the following equation:

݉ீ஻ ൌ ͲǤͲͷͳͺ

݇݃ ‫ܤܩܯ  כ‬ǡ݉ܽ‫ ݔ‬൅ ͷͻǤͶͺ݇݃ ܰ݉

‡“ǤʹǤͷ͸

The torque converter is estimated with 8.0 kg weight, already included in the gearbox weight. If no torque converter is used, the 8.0 kg need to be subtracted from the gearbox mass. The automated gearbox model uses a thermal mass model to calculate the heating and cooling of the gearbox. The composition of the gearbox materials and heat capacities can be seen in Table 2.8. The cooling of the gearbox uses the heat transfer coefficient (underbody) from Figure 2.4. Table 2.8:

Estimated mass fraction and specific heat capacity of the materials used in the automated gearboxes, with and without torque converter (TQ).

Material

Mass fraction [%] + TQ

Mass fraction [%]

Specific heat capacity [J / kg / K]

Steel

66

72

490

Aluminium

28

26

896

Oil

6

2

2090

2.5.2

Manual gearboxes

The efficiency model for the synthetic manual gearboxes is created by fitting measurement data of a gearbox to the excel optimization model. Due to missing publicly available data the gear meshing losses for the manual transmission

2 Modelling Methods

70

are not calculated separately. The synthetic gear meshing losses ߟீ஻ǡ௦௬௡Ǥ are calculated the same way as the additional loss efficiency for the automated gearbox, resulting in the following equation for the gearbox efficiency:

ߟீ஻ ൌ

‫ீܯ‬஻ǡ௜௡ െ ‫ீܯ‬஻ǡ஽௥௔௚ ߟீ஻ǡ௦௬௡Ǥ  ‫ீܯ‬஻ǡ௜௡

‡“ǤʹǤͷ͹

The efficiency parameters for the manual gearbox model, ‫ீܯ‬஻ǡ஽௥௔௚ and ߟீ஻ǡ௦௬௡Ǥ are fitted simultaneously to measurement data of a 7-speed gearbox, with 500 Nm of maximum torque input, retrieved from a FKFS internal source. The fitting of the gearbox drag torque results into the following equation: ‫ீܯ‬஻ǡ஽௥௔௚ ൌ ‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ ሺܶீ஻ ሻ

଴Ǥଶ଼ଽ଻ ‫ீܯ‬஻ǡ௠௔௫ ݊ீ஻ǡ௜௡Ǥ ቀ ݅ ି଴Ǥହହଽ଼  ቁ ିଵ ͷͲͲܰ݉ ͳͲͲͲ݉݅݊

‡“ǤʹǤͷͺ

MGB,Drag,ref. [Nm]

The temperature-dependent reference drag torque ‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ is assumed to be linear dependent [52] from the temperature reaching its minimum value at 60 °C, as it can be seen in Figure 2.38. 5.0 4.0 3.0 2.0 -40 -30 -20 -10 0

Figure 2.38:

10 20 30 40 50 60 70 80 90 100 TGB [°C]

Temperature-dependent reference drag torque  ǡ†”ƒ‰ǡ”‡ˆǤ for manual gearboxes.

The fitted equation for the additional loss efficiency can be written as:

2.5 Transmission

71

ߟீ஻ǡ௦௬௡Ǥ ൌߟீ஻ǡ௦௬௡Ǥǡ௥௘௙Ǥ ቀ

௡ಸಳǡ೔೙Ǥ

ଵ଴଴଴௠௜௡షభ

ቁ

଴Ǥ଴଴଼଴

݅ ି଴Ǥ଴଴଼ଶ 

‡“ǤʹǤͷͻ

100%

60

80% 60%

40

40%

20

20%

0

0%

Frequency [-]

-10

-3

-1 -0.4 -0.2 0 0.2 0.4 Deviation [%]

1

3

10

80

100%

60

80% 60%

40

40%

20

20%

0

0% -10 -3

Figure 2.39:

-1 -0.4 -0.2 0 0.2 0.4 Deviation [%]

1

3

Cumulative frequency

80

Cumulative frequency

Frequency [-]

The results for the reference temperature-dependent synthetic gear meshing efficiency does not show a temperature dependency and is fitted to ߟீ஻ǡ௦௬௡Ǥǡ௥௘௙Ǥ ൌ ͲǤͻ͸Ͷ. The fit of the gearbox drag torque ‫ீܯ‬஻ǡ஽௥௔௚ and the additional loss efficiency ߟீ஻ǡ௦௬௡Ǥ was carried out for two temperatures, 30 °C and 70 °C, with 210 measuring points each. To validate the quality of the fit, two histogram plots for the deviation between model and measurement points are created, which can be seen in Figure 2.39: one for ܶீ஻ = 30 °C (top) and one for ܶீ஻ = 70 °C (bottom).

10

Frequency and cumulative frequency of the deviation between the manual gearbox model and measurement data used for fitting of the model, with   = 30 °C (top) and   = 70 °C (bottom).

2 Modelling Methods

72

It shows that 80 % of the fitted points have a deviation of less than ±1 % compared to the measurement data. It can also be seen that the fit shows an offset. A compensation of that offset would lead to a higher root-mean-square error (RMSE) of the model fit and is therefore not carried out. The scaling of the deviation is not linear for better legibility. The created generic manual gearbox model is now used to create synthetic manual gearboxes for the simulation. The model is already usable for steady state modelling and is integrated into the full longitudinal simulation of the vehicle. The scaling is done by the maximum gearbox torque ‫ீܯ‬஻ǡ௠௔௫ at gearbox input. An example for the overall gearbox efficiency for different temperatures can be seen in Figure 2.40. The operation point for the efficiency mapping is at 200 Nm of input torque and 2500 min-1 input speed. 0.98

ηGB [-]

0.96

0.94

0.92

-40 °C

-20 °C

0 °C

20 °C

40 °C

60 °C - 100 °C

0.90 1 Figure 2.40:

2

gear number [-]

3

4

Gearbox efficiency for a synthetic manual gearbox at 200 Nm input torque and 2500 min-1 input speed for different gearbox temperatures.

The used gear ratio will be introduced in section 3.5 and can be checked in Table 3.3 (4-speed). The rotating inertia the of manual gearbox is retrieved from FKFS internal data and is assumed with 0.025 kg m² for positive torques

2.5 Transmission

73

and 0.020 kg m² for negative torques. The rotating inertia refers to the gearbox input. The mass for the different gearboxes is assumed with 20 kg for a 1speed, 25 kg for a 2-speed and 35 kg for a 4-speed gearbox. The manual gearbox model uses a thermal mass model to calculate the heating and cooling of the gearbox. The composition of the gearbox materials and heat capacities can be seen in Table 2.9. The cooling of the gearbox uses the heat transfer coefficient (underbody) from Figure 2.4. Estimated mass fraction and specific heat capacity of the materials used in the manual and differential gearboxes

Table 2.9:

2.5.3

Material

Mass fraction [%]

Specific heat capacity [J / kg / K]

Steel

66

490

Aluminium

26

896

Oil

2

2090

Differential gearboxes

For the differential gearboxes the same approach as for the automated and the manual gearboxes is used. Again, data retrieved from [55] [56] is adapted to the excel optimization model. The equation for calculating the efficiency of the differential gearboxes calculates as in eq. 2.57 for the manual gearboxes. First, a closer look to the bevel differential gearbox is taken. The drag torque of the bevel gearbox is assumed to be constant over the gear ratio ݅ and maximum input torque ‫ீܯ‬஻ǡ௠௔௫ and calculates as follows: ଴Ǥଷ଺ଵଶ ݊ீ஻ǡ௜௡Ǥ ‫ீܯ‬஻ǡ஽௥௔௚ ൌ ‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ ሺܶீ஻ ሻቀ  ቁ ͳͲͲͲ݉݅݊ିଵ

‡“ǤʹǤ͸Ͳ

2 Modelling Methods

74

MGB,Drag,ref. [Nm]

The temperature-dependent reference drag torque ‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ is modelled precisely, due to good availability of measurement data from [55]. The reference drag torque of the bevel gear differential gearbox reaches its minimum value at a temperature of 80 °C, as can be seen in Figure 2.41. 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 -40 -20 Figure 2.41:

0

20

40

60 80 100 120 140 160 180 200 TGB [°C]

Temperature-dependent reference drag torque  ǡ†”ƒ‰ǡ”‡ˆǤ for bevel differential gearboxes.

MGB,Drag,ref. [Nm]

The synthetic gear meshing efficiency is estimated with ߟீ஻ǡ௦௬௡Ǥ = 0.98. The fit of the gearbox drag torque ‫ீܯ‬஻ǡ஽௥௔௚ is carried out for three temperatures, 40 °C, 60 °C and 80 °C, with 5 measuring points each. The maximum deviation between fit and measurement data is ±10 %. Next, the spur differential gearbox is modelled. The spur differential gearbox uses two sets of planetary gears for transmission of the torque. The drag torque ‫ீܯ‬஻ǡ஽௥௔௚ is calculated in the same way as for the bevel differential gearbox (eq. 2.60). The reference drag torque ‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ is reduced according to measurement data from [56] and can be seen in Figure 2.42. 3.0 2.0 1.0 0.0 -40 -20

Figure 2.42:

0

20

40

60 80 100 120 140 160 180 200 TGB [°C]

Temperature-dependent reference drag torque  ǡ†”ƒ‰ǡ”‡ˆǤ for spur differential gearboxes.

2.5 Transmission

75

For the synthetic gear meshing efficiency the gear meshing loss model of the automated gearbox is used (‡“ǤʹǤͷʹሻ. The synthetic gear meshing efficiency over the gear ratio can be seen in Figure 2.43. The model for synthetic gear meshing efficiency has no temperature dependency.

ηGB,syn. [-]

1 0.98 0.96 0.94 1

2

3

4

5

6

7

8

9

10

i [-] Figure 2.43:

Gear ratio dependent synthetic gear meshing efficiency Ʉ ǡ•›Ǥ for spur differential gearboxes.

For the spur differential gearboxes, no fit was carried out, due to the lack of measurement data. A gearbox drag torque ‫ீܯ‬஻ǡ஽௥௔௚ǡ௥௘௙Ǥ reduction based on a relative measurement of drag reduction from bevel to spur differential gearboxes [56] as well as an adoption of the synthetic gear meshing efficiency ߟீ஻ǡ௦௬௡Ǥ is carried out. An example for the overall gearbox efficiencies for different temperatures and gearbox configurations is shown in Figure 2.44. The operation point for the efficiency mapping is at 400 Nm of input torque and 1000 min-1 input speed.

ηGB [-]

0.99 0.98 0.97 bevel (i = 3)

spur (i = 3)

spur (i = 5)

0.96 -40

-20

0

20

40

60

80

100

T [°C] Figure 2.44:

Gearbox efficiency for bevel and spur differential gearboxes for different gear ratios ݅ at 400 Nm of gearbox input torque and 1000 min-1 of gearbox input speed.

2 Modelling Methods

76

The rotating inertia of the differential gearboxes is neglected. The mass for the different differential gearboxes is assumed with 15 kg for the bevel gear and 10 kg the spur gear differential gearbox. The differential gearbox model uses a thermal mass model to calculate the heating and cooling of the gearbox. The composition of the gearbox materials and heat capacities can be seen in Table 2.9. The cooling of the gearbox uses the heat transfer coefficient (underbody) from Figure 2.4.

2.5.4

Electronically Controlled Multi-Plate Clutch

For the Electronically Controlled Multi-Plate Clutch (ECMC), again, the same approach as for the other gearboxes is used, resulting in the efficiency equation:

ߟா஼ெ஼ ൌ

‫ܯ‬ா஼ெ஼ǡ௜௡ െ ‫ܯ‬ா஼ெ஼ǡ஽௥௔௚ ߟா஼ெ஼ǡ௧௥௔௡௦Ǥ  ‫ܯ‬ா஼ெ஼ǡ௜௡

ߟா஼ெ஼ :

Efficiency of the ECMC [-]

‫ܯ‬ா஼ெ஼ǡ௜௡ :

Input torque of the ECMC [Nm]

‡“ǤʹǤ͸ͳ

‫ܯ‬ா஼ெ஼ǡ஽௥௔௚ : Drag torque of the ECMC [Nm] ߟா஼ெ஼ǡ௧௥௔௡௦Ǥ : Transmission efficiency of the ECMC [-] The ECMC drag torque is calculated with the same equation as for the differential gearbox (eq. 2.60). The reference ECMC drag torque is estimated from [57], with ‫ܯ‬ா஼ெ஼ǡ஽௥௔௚ǡ௥௘௙Ǥ = 0.35 Nm. The transmission efficiency of the ECMC is calculated with the slip between input and output of the ECMC. The slip is needed to transmit the power between in- and output. To reach optimal efficiency of the ECMC, the minimum design slip from [58] was retrieved, resulting in a value of 1.7 %. This slip value results in a value for the transmission efficiency of ߟா஼ெ஼ǡ௧௥௔௡௦Ǥ = 0.983. The efficiency of the ECMC over the input torque can be seen in Figure 2.45. The weight of the ECMC is estimated with 15 kg. The model of the ECMC does not have a model for rotating inertia or temperature.

2.5 Transmission

1 ηECMC [-]

0.98 0.96 0.94 0.92 0.9 0

Figure 2.45:

100 200 300 400 500 600 700 800 900 1000 MECMC,in [Nm] Efficiency of the ECMC over the input torque.

77

3 Development in Powertrain Technology The aim of this chapter is to analyse the development in powertrain technology until 2040. For this purpose, different powertrain components are defined. The major components for the simulation of the different powertrain configurations in this project are: x x x x x

Internal Combustion Engines Electric Drive Systems Battery Systems Fuel Cell Systems Transmissions

For each of these powertrain components a precise assumption is stated. In transmission no major developments are expected until 2040, but different types of gearboxes are introduced. After considering the most important powertrain components an overview of the charging systems, the charging columns as well as the tank systems is given.

3.1

Internal Combustion Engines

For the gasoline engines in this section a four-cylinder base engine from [13] is used as a baseline for the four Otto engine concepts. Three of these engines are gasoline engines with different technology packages to match the demands of the different powertrain configurations by keeping the costs for the engine concepts on acceptable levels. The fourth engine is a natural gas engine with an extensive technology package comparable to the gasoline engine with the most extensive technology package henceforth referred to as high efficiency concept. The diesel engine uses a state-of-technology 2020 four-cylinder engine as baseline. All engines are simulated and optimized in GT-Power. The fuel parameters used for the simulation are shown in Table 3.1. In this section a closer look on the internal combustion engine concepts is provided.

© The Author(s), under exclusive license to Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2023 T. Stoll, A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040, Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart, https://doi.org/10.1007/978-3-658-42168-7_3

3 Development in Powertrain Technology

80 Table 3.1:

Fuels used for the simulation of the internal combustion engines in GT-Power.

Engine type

Fuel

Lower heating value

Air-fuel ratio (stoic)

Density (1 bar / 20 °C)

Gasoline

E10

41.73 MJ / kg

14.50:1

744.6 kg / m³

Natural gas

Methane

50.00 MJ / kg

17.12:1

0.668 kg / m³

Diesel

B7

43.00 MJ / kg

14.33:1

826.8 kg / m³

3.1.1

Gasoline Engine (High Efficiency Concept)

The base engine for the High Efficiency Concept is a 2.0 l engine with 4 cylinders and 188 kW of maximum brake system power. The engine has a specific power of 94 kW / l, direct injection, a variable compression ratio between 8:1 and 22:1 and a stroke-to-bore ratio of 1.30. The maximum indicated efficiency of the engine is 46.5 %. The maximum steady state exhaust temperature is 965 °C. The maximum in-cylinder pressure during combustion is 135 bar and the maximum pressure rise during combustion is limited to 8 bar / °CA. To achieve maximum conversion rates of the three-way catalyst, the concept always works with air-to-fuel ratio of 1.0. The air and exhaust path as well as the engine map for the brake specific fuel consumption are shown in Figure 3.1. The engine uses a turbocharger with a variable nozzle turbine and an intercooled high pressure exhaust gas recirculation. For the exhaust gas aftertreatment an electrical heated three-way catalyst and a gasoline particle filter are intended. Furthermore, the engine concept includes an extensive technology package to achieve low levels of fuel consumption over wide parts of the operating range. The impact of these technologies is further divided into three measures.

3.1 Internal Combustion Engines

81

400 350 240 220

torque [Nm]

300 250 200 230 150

210

200

100 230

50 300 0 1000

Figure 3.1:

240 2000

400 3000 4000 speed [min-1]

250 600 5000

6000

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the high efficiency gasoline engine.

The first measure is the minimization of charge exchange losses. This effect can be achieved by reducing the throttling of the intake air in naturally aspirated operating points. For this purpose, a variable valvetrain with cylinder deactivation and an intercooled high pressure exhaust gas recirculation are intended. Using the variable valvetrain a Miller cycle is implemented into the engine operating strategy. In areas where the engine works naturally aspirated, the intake valves close early, thus reducing the amount of charge in the cylinder. In addition, two of the four cylinders can be deactivated by switching to a “zero” cam, for the intake as well as for the exhaust valves of these cylinders. Furthermore, a certain amount of exhaust gas can be recirculated. The recirculated exhaust gas is not an active part of the combustion but raises the amount of charge in the cylinder. The second measure to obtain a low fuel consumption is the reduction of knock probability while keeping a high combustion efficiency. That results in high pressures and high temperatures during combustion by keeping the optimal ignition timing. There are three possibilities to reduce knock. The first possibility is to shorten the burn duration. The mixture burns faster than the prereactions for the self-ignition can take place. The second possibility is to use

82

3 Development in Powertrain Technology

evaporation enthalpy or inert gas, to reduce the temperature in the combustion chamber. The third possibility is to delay the critical chemical reactions that cause knock. The engine uses a high turbulence concept with tumble flap, a pre-chamber spark plug and a high-pressure injection to keep the burn duration on low levels over the whole operating range. With the high-pressure injection, the fuel is injected close before the top dead centre of the compression stroke. The critical chemical reactions that lead to knock are delayed because the prereactions cannot take place in the preceding compression stroke. The third measure includes mechanical improvements of the engine. A variable compression ratio is used, which allows maximum efficiency over the whole operating range while providing optimum conditions for the other two measures. In addition, the improvement in engine friction and turbocharger efficiency from [13] is used for the baseline of the engine. The High Efficiency Concept is applied to the P0- and P2-hybrid electric powertrains without external recharge possibility. It provides low fuel consumption over the whole operating range and is optimized for low and medium torque applications. However, it is also the most expensive variation of the gasoline engine, due to the amount of technology that is used.

3.1.2

Gasoline Engine (Budget Optimized Concept)

The base engine for the Budget Optimized Concept is a 1.6 l engine with 4 cylinders and 141 kW of maximum brake system power. The engine has a specific power of 88 kW / l, direct injection, a compression ratio of 14.2:1 and a stroke-to-bore ratio of 1.35. The maximum indicated efficiency of the engine is 42.6 %. The maximum steady state exhaust temperature is 900 °C. The maximum in-cylinder pressure during combustion is 130 bar and the maximum pressure rise during combustion is limited to 8 bar / °CA. To achieve maximum conversion rates of the three-way catalyst, the concept always works with air-to-fuel ratio of 1.0. The air and exhaust path as well as the engine map for the brake specific fuel consumption are shown in Figure 3.2. The engine uses a turbocharger with a variable nozzle turbine. For the exhaust gas aftertreatment an electrical heated three-way catalyst and a gasoline particle filter are intended. Furthermore, the engine concept uses a medium cost technology

3.1 Internal Combustion Engines

83

package to achieve low levels of fuel consumption over wide parts of the operating range. The impact of these technologies is further divided into two measures.

240

300 230 250

220 torque [Nm]

200 213 150 100

50

240

250

300 400

0 1000

Figure 3.2:

2000

600 3000 4000 speed [min-1]

5000

6000

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the budget optimized gasoline engine.

The first measure to obtain a low fuel consumption is the reduction of knock probability while keeping a high combustion efficiency. The engine uses a high turbulence concept with tumble flap and a pre-chamber spark plug to keep the burn duration on low levels over the whole operating range. On high loads a water injection is used, which lowers the temperature and improves the heat capacity of the cylinder charge. That leads a lower knock probability in those operating points. The second measure includes mechanical improvements of the engine. The improvement in engine friction and turbocharger efficiency from [13] is used for the baseline of the engine. The Budget Optimized Concept is used for P2 hybrid electric powertrains with external recharge possibility. It provides low fuel consumption and is optimized for medium and high torque applications. It is a good trade-off between fuel consumption and costs.

3 Development in Powertrain Technology

84 3.1.3

Gasoline Engine (Range Extender Concept)

The base engine for the Range Extender Concept is a 1.3 l engine with 3 cylinders and 80 kW of maximum brake system power. The engine has a specific power of 62 kW / l, direct injection, a compression ratio of 12.3:1 and a stroketo-bore ratio of 1.12. The maximum indicated efficiency of the engine is 41.0 %. The maximum steady state exhaust temperature is 765 °C. The maximum in-cylinder pressure during combustion is 120 bar and the maximum pressure rise during combustion is limited to 8 bar / °CA. To achieve maximum conversion rates of the three-way catalyst, the concept always works with air-tofuel ratio of 1.0. The air and exhaust path as well as the engine map for the brake specific fuel consumption are shown in Figure 3-3. The engine uses a turbocharger as well as an electrical heated three-way catalyst and a gasoline particle filter for exhaust gas aftertreatment.

170

230

153 136

222

torque [Nm]

119 102 85

230 250

240

68 260

51 280

34 400

17 0 1000

Figure 3.3:

2000

3000 4000 speed [min-1]

270 300

600 5000

6000

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the range extender gasoline engine.

The engine uses a high turbulence concept for lower knock probability. In addition, the concept includes improvements in friction and turbocharger efficiency from [13]. The Range-Extender Concept is used for the S- and S/P2hybrid electric powertrains.

3.1 Internal Combustion Engines 3.1.4

85

Natural Gas Engine

The base engine for the natural gas engine is a 2.0 l engine with 4 cylinders and 188 kW of maximum brake system power. The engine has a specific power of 94 kW / l, manifold injection, a variable compression ratio between 8:1 and 22:1 and a stroke-to-bore ratio of 1.15. The maximum indicated efficiency of the engine is 44.5 %. The maximum steady state exhaust temperature is 885 °C. The maximum in-cylinder pressure during combustion is 200 bar and the maximum pressure rise during combustion is limited to 8 bar / °CA. To achieve maximum conversion rates of the three-way catalyst, the concept always works with air-to-fuel ratio of 1.0. The air and exhaust path as well as the engine map for the brake specific fuel consumption are shown in Figure 34. The engine uses one turbocharger with a variable nozzle turbine and an additional electric supercharger to achieve good performance at low engine speeds. For the exhaust gas aftertreatment an electrical heated three-way catalyst is intended. A particle filter is not required because of the low particle emissions of natural gas engines. Furthermore, the engine concept uses an extensive technology package to achieve low levels of fuel consumption over wide parts of the operating range. The impact of these technologies is further divided into three measures.

400 210

350 200

torque [Nm]

300 250 200 20

180

150 185

190

100 50 0 1000

Figure 3.4:

300 2000

220 600

3000 4000 speed [min-1]

5000

400 6000

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the natural gas engine.

3 Development in Powertrain Technology

86

The first measure is the minimization of charge exchange losses. For this, the same means as for the High Efficiency Concept of the Otto engine (3.1.1) are adopted. The second measure to obtain a low fuel consumption is the reduction of knock probability while keeping a high combustion efficiency. That results in high pressures and high temperatures during combustion by keeping the optimal ignition timing. Natural gas engines have a lower knock probability than gasoline engines due to the higher equivalent octane number of the fuel. The engine uses a high turbulence concept with tumble flap and a pre-chamber spark plug to keep the burn duration on low levels over the whole operating range. The third measure includes mechanical improvements of the engine. A variable compression ratio is used, which allows maximum efficiency over the whole operating range while providing optimum conditions for the other two measures. In addition, the improvement of the super- and turbocharger efficiency from [13] is used as the baseline of the engine. Because of the high incylinder pressures of the natural gas engine a steel piston instead of an aluminium piston is used and the engine friction is assumed 10 % higher than for the reference engine from [13]. The natural gas engine is used for the P0- and P2 hybrid electric powertrains without external charge possibility. It provides low fuel consumption over the whole operating range and is optimized for low and medium torque applications. The costs are comparable to the High Efficiency Concept of the gasoline engine.

3.1.5

Diesel Engine

The base engine for diesel engine is a 2.4 l engine with 4 cylinders and 172 kW of maximum brake system power. The engine has a specific power of 72 kW / l, a 2800 bar Common Rail injection system, a compression ratio of 22:1 and a stroke-to-bore ratio of 1.20. The maximum steady state exhaust temperature is 845 °C. The maximum in-cylinder pressure during combustion is 235 bar and the maximum pressure rise during combustion is limited to 8 bar / °CA. The maximum indicated efficiency of the engine is 45.7 %. The NOx-emissions of the engine are limited to a maximum of 5 g / kWh. The maximum Filter Smoke Number is limited to 2.5. To achieve low emissions

3.1 Internal Combustion Engines

87

levels for future emission standards an extensive exhaust gas aftertreatment system is intended. In Figure 3.5 the air and exhaust path as well as the engine map for the brake specific fuel consumption are shown.

188

450 400

190

1 2 3 4 5

Figure 3.5:

torque [Nm]

350

200

300

230

250 200 150 100 50 0 1000

250

210 400

600 2000 3000 speed [min-1]

4000

Air and exhaust gas path (left) and brake specific fuel consumption [g / kWh] (right) of the diesel engine.

The engine uses a turbocharger with a variable nozzle turbine as well as a highand a low-pressure exhaust gas recirculation for nitrogen oxide reduction. The aftertreatment package consist of a diesel oxidation catalyst with lean nitrogen oxide trap (1) for carbon monoxide and hydrocarbon emissions as well as nitrogen oxide emissions during cold start. In addition, there is a diesel particle filter (2) for particle emissions and two selective catalytic reduction catalysts (3, 4) to convert nitrogen oxide emissions during “warm” operation. The first selective catalytic reduction catalyst (3) is electrically heated to minimize nitrogen oxide emissions during warm-up of the engine. To achieve maximum nitrogen oxide conversion in the selective catalytic reduction catalysts (3, 4), more urea than required is injected into the exhaust gas to reach the maximum nitrogen oxide conversion rate. To avoid ammonia slip, an ammonia slip catalyst (5) is added.

3 Development in Powertrain Technology

88

The engine uses a heat insulated piston to minimize wall heat loss during combustion [15]. The turbocharge efficiency from [13] is used for the baseline of the engine. The engine friction is kept constant on 2020-levels because the maximum pressure is raised from 185 bar to 235 bar and a more robust construction of the engine is needed. The diesel engine is used for the P0- and P2-hybrid electric powertrains with and without external charge possibility as well as for the S- and S/P2-hybrid electric powertrains with external recharge possibility. It provides low fuel consumption over the whole operating range and is optimized for low, medium and high torque applications.

3.2

Electric Drive Systems

The electric drive systems have different configurations to match the different powertrain architectures. To reduce the configurations, integrated modules for the electric drive systems are assumed. The different modules for the drive systems are: x x x x

P0-module integrated in the internal combustion engines belt drive P2-module integrated in automatic gearbox E-Axle with 1- or 2- speed gearbox Highly integrated designs for special S/P-hybrid–gearbox applications

For the P0- and P2-modules, a permanent magnet synchronous motor is used. For the e-axle-drive an induction motor is used. The highly integrated S/Pdesigns use both types of motors. The permanent magnet synchronous motor is used for medium and low drive power applications. It has a high specific power density and a high efficiency but it is also one of the most expensive motor types on the market. In addition, it has a bad environmental impact due to the use of neodymium magnets. The induction motor causes lower costs and lower environmental impact than the permanent magnet synchronous motor. The specific power density is lower and the efficiency is slightly worse. It is used for medium and high drive power applications in the e-axle-drive and the highly integrated drive systems.

3.2 Electric Drive Systems

89

Disregarding the gearboxes of the different powertrain architectures, each electric drive system consists of a: x x x x x

Electric motor: Permanent magnet synchronous or induction motor Power electronics module: A bi-directional DC/AC-converter Connector: For connecting the electric drive system to the intermediate circuit of the powertrains electric system Motor controller: For optimal control of the electric drive system Sensors: For control and monitoring of the electric drive system

In the following section, the two electric motor types as well as on the composition of the power electronics module are discussed more closely.

3.2.1

Permanent Magnet Synchronous Motor

The reference motor for the permanent magnet synchronous motor has a maximum power out- and input of 85 kW. The motor works with a B6c-Inverter, thus providing variable alternating current, using a constant inverter switching frequency of 10 kHz. The motor has three phases and is connected in star. The inverter for the reference motor works with an intermediate circuit voltage of 400 V. The motor has four pole pairs and a reluctance rotor with neodymium magnets. The d-axis to q-axis inductance ratio is 1.4 to generate additional reluctance torque. The stator windings are assumed as hairpin copper windings. Figure 3.6 shows the general configuration of the motor as well as the efficiency map for motor and generator operation. For a consistent motor map, the efficiency of the generator operating points is inverted. The motor works with Maximum Torque per Ampere operating strategy to minimize the energy consumption and slowly switches to Maximum Torque per Voltage while entering the field weakening area. The efficiency of the motor refers to the IE5Efficiency-Class after [25]. The design point for the IE5-Efficiency-Class is at 45 % of rated torque and 90 % of characteristic speed (corner speed), thus providing optimal conditions for vehicle applications. The permanent magnet synchronous motor is used in all hybrid electric powertrains that use low and medium drive power motors, thus including all P0, P2 and S/P2 applications.

3 Development in Powertrain Technology

90

ϭϮ͕Ϯ

Figure 3.6:

3.2.2

Schematic cross section of the permanent magnet synchronous motor with reluctance rotor (left) and efficiency map of reference motor without inverter (right).

Induction Motor

The reference motor for the induction motor has a maximum power out- and input of 150 kW. The motor works with a B6c-Inverter, thus providing variable alternating current, using a constant inverter switching frequency of 10 kHz. The motor has three phases and is connected in star. The inverter for the reference motor works with an intermediate circuit voltage of 400 V. The motor has two pole pairs and a squirrel cage rotor. The assumed material of the squirrel cage rotor is aluminium. The stator windings, are also assumed as hairpin aluminium windings, to lower the price and improve the environmental impact of the motor. Figure 3.7 shows the general configuration of the motor as well as the efficiency map for motor and generator operation. For a consistent motor map, the efficiency of the generator operating points is inverted. The motor works with Maximum Torque per Ampere operating strategy to minimize the energy consumption and slowly switches to Maximum Torque per Voltage while entering the field weakening area. The efficiency of the motor refers to the IE3-Efficiency-Class after [25] and is further reduced by adapting the electric resistance of the stator to aluminium. The design point for the IE3-Efficiency-Class is at 45 % of rated torque and 90 % of characteristic speed (corner speed), thus providing optimal conditions for vehicle applications. The induction motor is mainly used for e-axle drives using medium and high drive power motors.

3.2 Electric Drive Systems

Figure 3.7:

3.2.3

91

Schematic cross section of the induction motor with squirrel cage rotor (left) and efficiency map of reference motor without inverter (right).

Power Electronics

Each electric motor has a power electronics module consisting of a controlled six pulse bridge converter (B6c) and an intermediate circuit capacitor. The module can work as inverter for motor and as rectifier for generator operation. Each power electronics module uses six silicon carbide metal-oxide-semiconductor field-effect transistors to optimize the efficiency of the power electronics module. The drain voltage range on the intermediate circuit varies between 0 V and 1000 V depending on the configuration of battery and electric motor. The maximum root-mean-squared drain current of each power transistor is limited to 1000 A. The size of the power transistors scales with the maximum power of the motor, meeting thus electric current demands. In Figure 3.8 the layout of the power electronics module as well as the modules efficiency for the 85 kW permanent magnet synchronous motor can be seen. For a consistent map of the converter, the efficiency of the rectifier operating points is inverted.

3 Development in Powertrain Technology

92

1-phase DC

3-phase AC

Figure 3.8:

3.3

Layout of the power electronics B6c-module (left) and efficiency map of the power electronics module for the permanent magnet synchronous reference motor (right).

Battery Systems

The battery system of each powertrain consists of different components depending on its size and the voltage level of the system. The main components are: x

x x x x

Housing: With matching high voltage protection for systems with a system voltage of 60 V or higher. For big batteries the cells are split up in modules, each module having its own controller. Splitting up the cells in modules enables easier maintenance and change of defective cells. In addition, defective modules can be disconnected from the battery system. Battery Cells: The battery cells in cylindrical, prismatic or pouch design Cooling System: For the cells with water- or air-cooling Connector: For connecting the battery system to the intermediate circuit of the powertrains electric system Battery Management System: For control and monitoring of the battery system

3.3 Battery Systems x x

93

Thermal Monitoring: Each cell has a temperature sensor for monitoring its temperature for safety and cooling reasons Balancing System: For balancing differences in cell voltage; system can be active or passive

The main part of the battery system are the battery cells. To meet the requirements for the different powertrain architectures, respectively hybrids, plug-in hybrids and battery electric, three cell designs for 2040 are defined. The assumptions from [44] are used as a baseline for the different battery cell types. In addition, more precise information is collected to find appropriate cell characteristics for these cell types. In Table 3.2 the three cell designs are shown. In the following section the three cell designs are discussed more closely. Table 3.2:

Battery cell definitions and composition for the different powertrains with electric drive systems.

Cell Type

Cathode

Anode

Gravimetric energy density

Use case

High power

LFP

LTO

140 W h / kg

HEV

Medium power

LMFP

C

220 W h / kg

PHEV/FCEV

High energy

NMC9.5.5

SiC

300 W h / kg

BEV

3.3.1

High Power Battery Cells

The high power battery cells use a cathode whose active material consists of lithium iron phosphate nanoparticles on graphene sheets. The battery half-cell data is retrieved from [38]. The anode consists of lithium titanium oxide nanoparticles, which enable charge and discharge rates of the anode half-cell up to 300 C. The data for the anode half-cell is retrieved from [35]. The fit of the full cell is oriented on [43]. The cell is designed as a prismatic cell and has an estimated usable gravimetric energy density of 140 W h / kg. The estimated capacity loss over 8000 full cycles is 10 %, which ensures that the cell has

3 Development in Powertrain Technology

94

enough fatigue strength to endure a full automotive life cycle. The usable voltage range of the cell is between 1.0 V and 2.3 V [43]. The fitted cell losses for the high power battery cells are shown in Figure 3.9. The maximum charge and discharge rate for this cell type is 60 C respectively -60 C. Due to the high possible C-rates and high fatigue strength this cell type is used for P0- and P2hybrid electric powertrains without external recharge possibility, keeping the nominal capacity of the battery systems on low levels.

2.4 cell voltage [V]

2.2 2 1.8 1.6 1.4 -60 C -5 C 10 C

1.2 1 0 Figure 3.9:

3.3.2

0.2

0.4

SOC [-]

0.6

-30 C OCV 30 C

0.8

-10 C 5C 60 C

1

Voltage and Current characteristic of the high power battery cells in different States-of-Charge (SOC).

Medium Power Battery Cells

The medium power battery cells use a cathode whose active material consists of lithium manganese iron phosphate, with a manganese iron ratio of 50 % to 50 %. The battery half-cell data is retrieved from [39]. The anode consists of nano-graphite. The data for the anode half-cell is retrieved from [42]. The fit of the full cell is oriented on [39] [42].The cell is designed as a prismatic cell and has an estimated usable gravimetric energy density of 220 W h / kg. The lifespan comprises estimated 1000 full cycles. The end-of-life capacity is 80 % of the nominal capacity. The usable voltage range of the cell is between 2.7 V and 4.35 V. The fitted cell losses for the medium power battery cells are shown in Figure 3.10. The maximum charge and discharge rate for this cell type is

3.3 Battery Systems

95

15 C respectively -15 C. This cell type is used for all hybrid electric powertrains with external recharge possibility, providing a compromise between high power and high energy cells. 4.3

cell voltage [V]

4.1 3.9 3.7 3.5 3.3 3.1

-15 C -1 C 5C

2.9 2.7 0

0.2

0.4

0.6

-10 C OCV 10 C

0.8

-5 C 1C 15 C

1

SOC [-] Figure 3.10:

3.3.3

Voltage and Current characteristic of the medium energy battery cells in different States-of-Charge (SOC).

High Energy Battery Cells

The high energy battery cells use a cathode whose active material consists of 90 % nickel, 5 % manganese and 5 % cobalt. The battery half-cell data is retrieved from [40]. The anode consists of an 80 % graphite and 20 % silicone to reach high energy densities. The data for the anode half-cell is retrieved from [34]. The fit of the full cell is oriented on [41]. The cell is designed as a prismatic cell and has an estimated usable gravimetric energy density of 300 Wh / kg. The lifespan comprises estimated 500 full cycles. The end-oflife capacity is 80 % of the nominal capacity. The usable voltage range of the cell is between 2.7 V and 4.15 V. The fitted cell losses for the high energy battery cells are shown in Figure 3.11. The maximum charge and discharge rate for this cell type is 4 C respectively -4 C. Due to the low C-Rates this cell is only used for battery electric powertrains, that use high capacity batteries for high ranges. These batteries provide enough current for high electric drive powers.

3 Development in Powertrain Technology

96

4.1 cell voltage [V]

3.9 3.7 3.5 3.3 3.1 -4 C -0.1 C 0.5 C

2.9 2.7 0

0.2

0.4

-1 C OCV 1C

0.6

-0.5 C 0.1 C 4C

0.8

1

SOC [-] Figure 3.11:

3.4

Voltage and Current characteristic of the high energy battery in different States-of-Charge (SOC).

Fuel Cell Systems

The fuel cell system consists of different components to provide best operating conditions for the electrochemical energy conversion. The main components of the fuel cell system are: x x x x x x

Fuel Cell Stack: Containing the components for electrochemical conversion Intake Air Path: Pump or compressor for air supply Exhaust Gas Path: Disposal of H2O and used air Intake Air Conditioning: Humidification of the intake air if needed Fuel Conditioning: Fuel supply and control of fuel gas parameters Power Electronics: Converting the variable DC-voltage of the stack to requested DC-voltage

The fuel cell stack uses a 20 μm Proton Exchange Membrane as electrolyte with applied nano structured thin film electrodes. A platinum loading of 0.125 mg Platinum / cm² membrane is assumed. The rated power is 1.0 W / cm² of membrane. The voltage loss at 300 mA / cm² is 0.8 V. The membrane

3.4 Fuel Cell Systems

97

electrode assembly has a durability of 8000 h in an automotive drive cycle. With these assumptions retrieved from [47] and [48] the voltage characteristic of the fuel cell is fitted on a fuel cell model. The fitted cell design is shown in Figure 3.12. The curve is for an anode and cathode pressure of 1.5 bar using pure hydrogen with a stoichiometric flow rate of 2.0 on the anode side and using air with a stoichiometric flow rate of 2.5 on the cathode side. The cell temperature is 80 °C.

cell voltage [V]

1.0 0.8 0.6 0.4 0.2 0.0 0

Figure 3.12:

500

1000 1500 i [mA / cm²]

2000

2500

Fitted voltage characteristic of the fuel cell.

Figure 3.13 shows the air and exhaust path for the fuel cell system on the left and the system efficiency for a 60 kW and an 80 KW fuel cell system on the right. The fuel cell system has an electrically assisted turbocharger that delivers the air to the fuel cell stack and assures maximum energy recovery from the exhaust gas. After the compressor, a water injection (1) is installed to assure optimal humidity of the cathode air and as a result, optimal humidity of the membrane. The water needed for humidification is extracted from the exhaust air with a dehumidifier (2). Between humidifier and dehumidifier, a water tank is installed to always assure enough water for humidification. The hydrogen from the high-pressure tank (3) is decompressed and delivered to an ejector that recirculates the hydrogen in the anode circuit. To suppress to high concentrations of nitrogen caused by nitrogen diffusion from the cathode side, the anode side has a purge valve. When the nitrogen level reaches 3 % the anode circuit purges into the diluter behind the dehumidifier. The diluter ensures that the hydrogen concentration in the exhaust gas stays lower than 4 %

3 Development in Powertrain Technology

98

to keep the mixture non-flammable. On the electric side a buck-boost converter with two silicon carbide metal-oxide-semiconductor field-effect transistors are used for optimized efficiency. 70%

2 1

system efficiency

60% 50% 40% 30% 20% 80 kW 60 kW

10% 0% 0

3 Figure 3.13:

3.5

20 40 60 80 electrical power output [kW]

Air and fuel path (left) and overall system efficiency (right) of the fuel cell system.

Transmissions

For the transmission eight gearboxes and one all-wheel-drive module is assumed. The transmission components are: x x x x x x x x

9-speed automatic gearbox with / without torque converter for P0 and P2 powertrains 6-speed automatic gearbox with / without torque converter for P0 and P2 powertrains 4-speed spur gearbox with dual clutch for S/P2 powertrains 2-speed spur gearbox with clutch for S/P2 powertrains 2-speed spur gearbox with clutch for e-axle drives 1-speed spur gearbox for e-axle drives Bevel- and spur-differential gearbox Electronically controlled multi-plate-clutch

3.5 Transmissions

99

The efficiency of these components is assumed on 2020 levels. The gear ratios of the transmissions are listed in Table 3.3. Table 3.3:

Gear ratios of the different transmission types.

Gear 1

2

3

4

5

6

7

8

9

9-speed

4.70

2.85

1.90

1.38

1.00

0.80

0.70

0.58

0.48

6-speed

4.17

2.34

1.52

1.14

0.88

0.69

-

-

-

4-speed (PHEV)

1.52

0.85

0.58

0.48

-

-

-

-

-

2-speed (PHEV)

1.00

0.67

-

-

-

-

-

-

-

2-speed (E-AXLE)

3.19

1.00

-

-

-

-

-

-

-

1-speed (E-AXLE)

2.20

-

-

-

-

-

-

-

-

Transmission

The 9- and 6- speed automatic gearboxes are used for the P0- and P2- hybrid electric powertrains with and without external recharge. The hybrid electric powertrains with external recharge possibility do not have a torque converter because an electric run-up of the vehicle is assumed. The 9-speed gearbox is used for the sedan and the sport utility vehicle, the 6-speed gearbox for the light-duty vehicle. The 4- and 2-speed gearboxes are highly integrated for the S/P2-hybrid electric powertrain applications. The 4-speed transmission uses a dual clutch because of more frequent switching events compared to the 2speed transmission. The 2-speed transmission uses a standard clutch. The 2speed transmission for e-axle applications uses a multi-plate clutch for switching the gear. The differential gearboxes do not have a fixed gear ratio. The

3 Development in Powertrain Technology

100

gear ratio depends on the vehicle concept and the used powertrain. The electronically controlled multi-plate clutch is used for all-wheel drive on the P0and P2- hybrid electric powertrain architectures.

3.6

Tank and Charging Systems

For the gasoline and diesel engines a standard tank system for liquid fuels is assumed. The natural gas engines use a 200 bar pressure tank, the fuel cell system a 700 bar pressure tank. All pressure tanks are assumed as composite tanks. The pressure levels of the tanks are determined by the available space in the vehicle for fitting the pressure tank system. The schematic Figure 3.14 shows the transport of electric energy between the grid and the vehicle for external electric charging. The accounting limit for the electric energy consumption of the vehicle is set to position (1).

Figure 3.14:

Electric charging system between grid and vehicle battery with energy accounting limit (1).

3.6 Tank and Charging Systems

101

The charging system is assumed to be the CCS3 (combined charging system). For charging, two different types of charging scenarios are assumed based on data from [59]: x x

Fast charging: 500 kW maximum charging power with a high power ultra-fast direct current charging system Normal charging: 16.5 kW charging power with a normal alternating current charging system.

Based on these two scenarios a decision for the usage distribution of the scenarios for plug-in hybrid and battery electric vehicles is made. For the plug-in hybrid electric vehicles, a usage of 10 % fast charging and 90 % normal charging is assumed. For the battery electric vehicles, the distribution is 20 % fast charging and 80 % normal charging. For fast charging a charging range between 10 % and 90 % state of charge of the nominal capacity of the vehicle battery is stated. For the normal charging this range is between 5 % and 95 % state of charge of the nominal capacity of the vehicle battery. For the fast charging station, a high power ultra-fast direct current charging system is assumed. The maximum charging current is 500 A using a voltage range between 0 and 1000 V. The maximum charging power with these parameters is 500 kW. The charging stations AC/DC-converter uses silicon carbide metal-oxide-semiconductor field-effect transistors for power conversion. The charging efficiency is assumed to be 100 % between charging station and battery. Due to the chosen energy accounting limit, the efficiency of the charging column is not considered for the vehicle electric energy consumption. The charging efficiency (left) and charging time (right) over battery size for the high energy battery using fast charging are shown in Figure 3.15. Throttling effects of the charging power and cooling losses due to thermal effects of the battery are not considered.

3 Development in Powertrain Technology 98

charging time [min]

charging efficiency [%]

102

97 96 95 0

Figure 3.15:

30 25 20 15 10 5 0 0

50 100 150 200 battery size [kWh]

50 100 150 200 battery size [kWh]

Charging efficiency (left) and charging time (right) over battery size for the high energy battery using fast charging.

96 charging time [h]

charging efficiency [%]

The charging-efficiency of the on-board-charger for AC-charging is estimated at 96 % between charging column and vehicle battery. The on-board-charger, also uses silicon carbide metal-oxide-semiconductor field-effect transistors for power conversion. The maximum power output is limited to 16.5 kW. Due to the chosen energy accounting limit, the efficiency of the charging column is not considered for the electric energy consumption of the vehicle. The charging efficiency (left) and charging time (right) over battery size for the medium power battery using normal charging are shown in Figure 3.16. Throttling effects of the charging power and cooling losses due to thermal effects of the battery are not considered.

95 94 93 92 0

Figure 3.16:

10 20 30 40 battery size [kWh]

50

3 2.5 2 1.5 1 0.5 0 0

10 20 30 40 battery size [kWh]

50

Charging efficiency (left) and charging time (right) over battery size for the medium power battery using normal charging.

4 Powertrain Simulation For the powertrain simulation, a new simulation environment in MATLAB is developed. The simulation environment contains models for the different powertrain components, the vehicle models, as well as a testing environment for the different drive cycles. For calculating the optimized energy consumption, the general approach is to use a backward simulation with a dynamic programming algorithm. This algorithm has major disadvantages in calculation times for high numbers of input variables. In addition, the algorithm tends to have stability problems for complex simulations and the use case application can be difficult. The advantage of the dynamic programming algorithm is that the solution is seen as the global optimum of the problem. On the other hand, the simulation can be performed as a forward simulation using equivalent consummation minimization strategy. This algorithm is simpler for application purposes and offers much faster calculation times, but due to its application parameters the results do not show the optimal solution. Another advantage of the equivalent consumption minimization strategy is that a more complex operational strategy for the vehicle can be implemented including hysteresis, triggers and retardations. For these reasons, a new optimization algorithm is introduced combining the advantages of both algorithms by minimizing the disadvantages. The new algorithm is called “optimized Multi-Objective Equivalent Consumption Minimization Strategy”. The algorithm works by setting the forward simulation of the vehicle using Equivalent Consumption Minimization Strategy into an optimization loop using the MATLAB global optimization algorithm patternsearch. The parameters of the Equivalent Consumption Minimization Strategy in the simulation are then optimized until the global optimum of the solution is found. The setup parameters of the patternsearch-algorithm are set to optimized levels through testing for fast and reproducible solutions. Figure 4.1 shows the state of charge curve for an example hybrid electric vehicle driving the full Worldwide harmonized Light Duty Test Cycle.

© The Author(s), under exclusive license to Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2023 T. Stoll, A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040, Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart, https://doi.org/10.1007/978-3-658-42168-7_4

104

Figure 4.1:

4 Powertrain Simulation

State of charge over time comparison of dynamic programming (DP) and optimized Multi-Objective Equivalent Consumption Minimization Strategy (opt. MO-ECMS) algorithm on an example hybrid electric vehicle.

The solution created with the optimized Multi-Objective Equivalent Consumption Minimization Strategy (opt. MO-ECMS) shows the same curve as the solution created with the dynamic programming algorithm (DP). The fuel consumption calculated with the opt. MO-ECMS algorithm is 0.14 % lower than the one calculated with the DP-algorithm. This may be caused by the interpolation and discretization of the output parameters in the DP-algorithm. The calculation time can also be improved by using parallel computing with the opt. MO-ECMS algorithm. The DP-algorithm cannot be parallelized. For example, on a six-core computer, the calculation time of the opt. MO-ECMS algorithm is 34 % of the DPs calculation time. In addition, the opt. MO-ECMS algorithm shows a more stable calculation, easier application and more flexibility than the DP-Algorithm. One disadvantage, however, is that the algorithm is limited by the underlying control strategy of the vehicle. In the following chapter, the different models for the powertrain simulation are introduced and the boundary conditions for the simulation are defined. In the first section the three vehicle types, sedan, sport utility vehicle and lightduty vehicle are introduced. In the second section, the different powertrain designs are presented. After that, a quick overview of the function and scope of the used component models is given. In the fourth section, the nine different

4.2 Vehicle Models

105

drive cycles for different use cases of the vehicle are introduced. In the last section the operational strategy, respectively the multi-objective consumption minimization strategy for the different powertrain configurations is presented.

4.1

Vehicle Models

In this section, the parameters for the vehicle model are introduced. The vehicle parameters are retrieved from [13] and [60]. For each vehicle a weight and a payload are defined. For the sedan, the average payload is 240 kg. For the sport utility vehicle, the average payload is assumed 290 kg. For the light-duty vehicle, a utilization rate of 0.53 for the reference vehicle is stated, resulting in a payload of 2804 kg. After adding the average payload to the vehicle, the weight of the reference powertrain is subtracted. The reference powertrain is a 2020 state-of-the-art combustion engine powertrain containing the internal combustion engine, a 9-speed gearbox and further required components such as differential gearboxes and all-wheel drive gearboxes. The vehicle parameters for the different vehicles and the 2.0 t tarpaulin trailer are shown in Table 4.1. In addition, the electric ranges for the different powertrain architectures are defined as follows: x x x

Parallel plug-in hybrid powertrains: 80 km Serial/parallel plug-in hybrid powertrains: 80 km Serial plug-in hybrid powertrains (range extender): 120 km

The overall range of all powertrains stays constant with 800 km for the sedan and sport utility vehicle and 400 km for the light-duty vehicle.

4.2

Powertrain Component Models

The powertrain component models are mainly equations and characteristics imported into MATLAB. In this section, the most important dependencies for the input and output values are shown. The underlying component models are created with different simulation software programs. The component modelling can be retrieved from chapter 2 and the results of the modelling from

4 Powertrain Simulation

106

chapter 3. In the following, the model dependencies for the different component models are explained. Table 4.1:

Model parameters for the different vehicles used in the simulation. Sedan

Sport utility vehicle

Tarpaulin trailer

Light-duty vehicle

Overall system power

100 kW

200 kW

-

150 kW

Gross weight w/o powertrain

1376 kg

1634 kg

2000 kg

4636 kg

Overall Range

800 km

800 km

-

400 km

Drag coefficient

0.255

0.281

1.250

0.300

Frontal area

2.247

2.613

4.2

4.27

Tires

205/55 R16

225/50 R17

165/65 R14

225/75 R16

Rolling resistance coefficient

0.0062

0.0068

0.0068

0.0068

Dynamic rolling radius

0.307

0.319

0.277

0.361

Vehicle type

4.2 Powertrain Component Models

107

For the internal combustion engine model, different dependencies are modelled. The combustion engine main equation for the fuel consumption is modelled as:

݉ሶ௙௨௘௟ ൌ ݂ሺ݊ூ஼ா ǡ ‫݌‬௠௜ ǡ ܸ௛ ሻ

‡“ǤͶǤͳ

and is therefore dependent on the engine speed ݊ூ஼ா , the indicated mean effective pressure ‫݌‬௠௜ and the cylinder swept volume ܸ௛ . The indicated mean effective pressure having the following dependencies:

‫݌‬௠௜ ൌ ݂൫‫ܯ‬ூ஼ாǡ௜௡ ǡ ‫݌‬௠௥ ൯

‡“ǤͶǤʹ

with ‫ܯ‬ூ஼ாǡ௜௡ being the torque request on internal combustion engine input and ‫݌‬௠௥ being the frictional mean effective pressure. The dependencies for the friction mean effective pressure are:

‫݌‬௠௥ ൌ ݂൫݊ூ஼ா ǡ ‫ܯ‬ூ஼ாǡ௜௡ ǡ ܸ௛ ǡ ‫ݖ‬௖௬௟Ǥ ǡ ܶூ஼ா ൯

‡“ǤͶǤ͵

The friction mean effective pressure depends on these five parameters, with ܶூ஼ா being the mean temperature and ‫ݖ‬௖௬௟Ǥ being the number of cylinders of the internal combustion engine. In addition, more parameters are needed for the exhaust gas aftertreatment model. The raw emissions of the internal combustion engines are not modelled, except for the thermal nitrogen oxides (NO) of the diesel engine. These emissions result in the following function:

݉ሶேைǡ௧௛Ǥ ൌ ݂ሺ݊ூ஼ா ǡ ‫݌‬௠௜ ሻ

The input values for the thermal catalyst model are:

‡“ǤͶǤͶ

4 Powertrain Simulation

108

݉ሶ௘௫௛௔௨௦௧ǡସ ൌ ݂ሺ݊ூ஼ா ǡ ‫݌‬௠௜ ሻ

‡“ǤͶǤͷ

ܶ௘௫௛௔௨௦௧ǡସ ൌ ݂ሺ݊ூ஼ா ǡ ‫݌‬௠௜ ሻ

‡“ǤͶǤ͸

In addition, the internal combustion engine model comprises a thermal model for the engine and the catalysts as well as a model for the engines inertia and the engine weight. For the electric drive model, the efficiency of the motors is modelled with the energy transformation rate of the electric drive system defined as:

‫ݓ‬ா஽ ൌ

ܲ௠௘௖௛Ǥ  ܲ௘௟Ǥ

‡“ǤͶǤ͹

with ܲ௠௘௖௛Ǥ being the mechanical power and ܲ௘௟Ǥ being the electrical power of the electric drive system. The energy transformation rate has the following dependencies:

‫ݓ‬ா஽ ൌ ݂൫݊ா஽ ǡ ‫ܯ‬ா஽ǡ௜௡ ǡ ܲேǡா஽ ǡ ‫ܫ‬ேǡ௉ா ൯

‡“ǤͶǤͺ

with ݊ா஽ being the rotational speed of the electric drive system,‫ܯ‬ா஽ǡ௜௡ being the mechanical input/output torque of the electric drive system, ܲேǡா஽ being the nominal power of the electric drive and ‫ܫ‬ேǡ௉ா being the nominal current of the power electronics module. In addition, the model for the electric drive system contains a model for the motor inertia and the motor weight. For the battery model the cell energy transformation rate is defined as:

4.2 Powertrain Component Models

‫ݓ‬௖௘௟௟ ൌ

ܲ௖௘௟௟ǡ௜௡ǤȀ௢௨௧Ǥ  ܲ௖௘௟௟ǡை஼௏

109

‡“ǤͶǤͻ

with ܲ௖௘௟௟ǡ௜௡ǤȀ௢௨௧Ǥ being the power in- or output on the cell clamps and ܲ௖௘௟௟ǡை஼௏ defined as the inner stored or taken power of one battery cell. The inner cell power has the following dependencies:

ܲ௖௘௟௟ǡை஼௏ ൌ ݂ ቀܷ௖௘௟௟ǡை஼௏ ሺܱܵ‫ܥ‬ሻǡ ‫ܫ‬௖௘௟௟ ൫ܷ௖௘௟௟ ǡ ܲ௖௘௟௟ǡ௜௡ǤȀ௢௨௧Ǥ ൯ቁ

‡“ǤͶǤͳͲ

The open circuit voltage ܷ௖௘௟௟ǡை஼௏ is a function of the batteries state of charge (SOC). ‫ܫ‬௖௘௟௟ is the current in one battery cell dependent on the clamp cell voltage ܷ௖௘௟௟ and the cell in- or output power ܲ௖௘௟௟ǡ௜௡ǤȀ௢௨௧Ǥ . The clamp cell voltage has the following dependencies:

ܷ௖௘௟௟ ൌ ݂൫ܱܵ‫ܥ‬ǡ ‫ܥ‬ሺܲ௖௘௟௟ǡ௜௡ǤȀ௢௨௧Ǥ ǡ ‫ܧ‬ேǡ௖௘௟௟ ሻ൯

‡“ǤͶǤͳͳ

The cell voltage ܷ௖௘௟௟ is a function of the batteries state of charge and the CRate ‫ܥ‬, which is calculated with the in- or output power ܲ௖௘௟௟ǡ௜௡ǤȀ௢௨௧Ǥ and the nominal capacity of one battery cell ‫ܧ‬ேǡ௖௘௟௟ . In addition, the battery model contains a model for the battery weight. For the fuel cell system model, the efficiency is defined as:

ߟி஼ௌǡ௚௘௦Ǥ ൌ ݂൫ܲி஼ௌǡ௘௟ǡ௢௨௧Ǥ ൯

‡“ǤͶǤͳʹ

with ߟி஼ௌǡ௚௘௦Ǥ being the overall efficiency and ܲி஼ௌǡ௘௟ǡ௢௨௧Ǥ beeing the electric energy output of the fuel cell system. In addition, the fuel cell model contains a model for the weight of the fuel cell system. For all transmission models the following dependencies for the gearbox efficiency are used:

4 Powertrain Simulation

110

ߟீ஻ ൌ ݂൫݊ீ஻ǡ௜௡Ǥ ǡ ݅௚௘௔௥ ǡ ܶீ஻ ൯

‡“ǤͶǤͳ͵

with ݊ீ஻ǡ௜௡Ǥ as input speed of the transmission, ݅௚௘௔௥ as the current gear ratio of the gearbox and ܶீ஻ as current gearbox temperature. For the gearbox drag torque the following dependencies are used:

‫ீܯ‬஻ǡௗ௥௔௚ ൌ ݂൫݊ீ஻ǡ௜௡Ǥ ǡ ݅௚௘௔௥ ǡ ܶீ஻ ǡ ‫ீܯ‬஻ǡ௠௔௫Ǥ ൯

‡“ǤͶǤͳͶ

with ‫ீܯ‬஻ǡ௠௔௫Ǥ being the maximum input torque of the transmission. The electronically controlled multi plate clutch for the all-wheel drive has no thermal model. In addition, each transmission includes an inertia and component weight model.

4.3

Powertrain Design

In this section, the different powertrain architectures for the simulations are introduced. Each powertrain architecture is implemented into the vehicles from section 4.1. For each configuration the electric drive power, the battery size and the gear ratio of the differential gearbox is defined and optimized. Here, only a short overview of the simulation variations is given. More detailed information on the powertrain configurations can be found in the simulation data sheets described in Appendix A1, which are available for download.

4.3 Powertrain Design

111

The first powertrain architecture is a P0-hybrid electric powertrain with a 48 V battery system and no external recharge possibility. The P0-powertrain architecture is only examined for the sedan. The powertrain configuration can be seen in Figure 4.2.

Figure 4.2:

P0-hybrid electric powertrain architecture.

Three different combustion engines are examined: the high efficiency concept gasoline engine, the diesel engine and the natural gas engine. The combustion engine power matches the overall system power. The maximum speed of the vehicle is limited to 250 km / h. An exception are the light-duty vehicles for which the maximum speed is limited to 130 km / h. The powertrain is mounted in front transverse position and uses a 9-speed automated gearbox with torque converter. An electric permanent magnet synchronous motor with 12 kW of mechanical power is installed on the belt drive of the internal combustion engine. The voltage level of the battery is limited to 60 V resulting in a mean voltage level of 48 V. For the battery, the high power battery cells are used. The simulation variations for the P0-hybrid without external recharge possibility are shown in Table 4.2.

4 Powertrain Simulation

112 Table 4.2:

Simulation variations of the P0-hybrid electric powertrain architecture.

Engine Type

ID

Internal combustion engine power

Electric drive power

Battery size

Gears

Sedan Gasoline – High Efficiency Concept

01

100 kW

12 kW

0.5 kWh

9

Diesel

02

100 kW

12 kW

0.5 kWh

9

Natural Gas

03

100 kW

12 kW

0.5 kWh

9

The second powertrain architecture is a P2-hybrid electric powertrain with a 48 V battery system and no external recharge possibility. This powertrain configuration is examined in all three vehicle types. The P2-powertrain architecture for the 48 V-system can be seen in Figure 4.3, with front-wheel drive on the top and all-wheel drive on the bottom. Three different engine types are also being investigated for this powertrain architecture: the high efficiency gasoline engine, the diesel engine and the natural gas engine. The overall system power matches the sum of the internal combustion engine power and the electric drive power. The maximum speed of the vehicle is limited to 250 km / h. An exception are the light-duty vehicles for which the maximum speed is limited to 130 km / h. The powertrain is mounted in front transverse position and uses a 9-speed automated gearbox with torque converter for the sedan and sport utility vehicle and a 6-speed automated gearbox with torque converter for the light-duty vehicle. The allwheel drive is realized with an electronically controlled multi plate clutch unit, driving the rear axle of the vehicle. An electric permanent magnet synchronous motor with a power of 18 kW is mounted between engine output and gearbox input. For this architecture, the maximum output power of the electric drive system is limited due to current limitations caused by the low voltage level of 48 V. The voltage level of the battery is limited to 60 V resulting in a mean

4.3 Powertrain Design

113

voltage level of 48 V. For the battery, the high power battery cells are used. The simulation variations for the P2-hybrid with 48V battery and without external recharge possibility are listed in Table 4.3.

ϭϬͬϳͬE'

 Ε с 

Figure 4.3:

ϰϴsͲ 

P2-hybrid electric powertrain architecture for a 48 V battery system, for front-wheel drive (top) and all-wheel drive (bottom).

4 Powertrain Simulation

114

Table 4.3: Simulation variations of the P2-hybrid electric powertrain architecture for a 48 V battery system.

Engine Type

ID

Internal combustion engine power

Electric drive power

Battery size

Gears

Sedan Gasoline – High Efficiency Concept

04

82 kW

18 kW

0.8 kWh

9

Diesel

07

82 kW

18 kW

0.8 kWh

9

Natural Gas

10

82 kW

18 kW

0.8 kWh

9

Sport utility vehicle Gasoline – High Efficiency Concept

05

182 kW

18 kW

0.8 kWh

9

Diesel

08

182 kW

18 kW

0.8 kWh

9

Natural Gas

11

182 kW

18 kW

0.8 kWh

9

Light-duty vehicle Gasoline – High Efficiency Concept

06

132 kW

18 kW

0.8 kWh

6

Diesel

09

132 kW

18 kW

0.8 kWh

6

Natural Gas

12

132 kW

18 kW

0.8 kWh

6

The third powertrain architecture is a P2-hybrid electric powertrain with a 400 V battery system and no external recharge possibility. This powertrain architecture is examined in all three vehicle types. The P2-powertrain architecture for the 400 V-system without external recharge possibility can be seen in Figure 4.4, with front-wheel drive on the top and all-wheel drive on the bot-

4.3 Powertrain Design

115

tom. For this powertrain architecture, the general configuration of the powertrain is similar to the P2-hybrid electric powertrain with 48 V, except for the electric parts of the powertrain. An electric permanent magnet synchronous motor with a power between 25 kW and 60 kW is mounted between engine output and gearbox input. The mean voltage level of the battery is 400 V. For the battery, the high power battery cells are used. The simulation variations for the P2-hybrid electric powertrain with 400V battery and without external recharge possibility are summarized in Table 4.4.

ϭϬͬϳͬE'

 Ε с 

ϰϬϬsͲ 

ϭϬͬϳͬE'

 Ε с 

Figure 4.4:

ϰϬϬsͲ 

P2-hybrid electric powertrain architecture for a 400 V battery system, for front-wheel drive (top) and all-wheel drive (bottom).

4 Powertrain Simulation

116 Table 4.4:

Simulation variations of the P2-hybrid electric powertrain architecture for a 400 V battery system.

Engine Type

ID

Internal combustion engine power

Electric drive power

Battery size

Gears

Sedan Gasoline – High Efficiency Concept

13

75 kW

25 kW

1.8 kWh

9

Diesel

16

75 kW

25 kW

1.8 kWh

9

Natural Gas

19

75 kW

25 kW

1.8 kWh

9

Sport utility vehicle Gasoline – High Efficiency Concept

14

140 kW

60 kW

4.3 kWh

9

Diesel

17

140 kW

60 kW

4.3 kWh

9

Natural Gas

20

140 kW

60 kW

4.3 kWh

9

Light-duty vehicle Gasoline – High Efficiency Concept

15

110 kW

40 kW

2.9 kWh

6

Diesel

18

110 kW

40 kW

2.9 kWh

6

Natural Gas

21

110 kW

40 kW

2.9 kWh

6

The fourth powertrain architecture is again a P2-hybrid electric powertrain with a 400 V battery system, but with an external recharge possibility. This powertrain configuration is examined in all three vehicle types. The P2-powertrain architecture for the 400 V-system with external recharge possibility is

4.3 Powertrain Design

117

shown in Figure 4.5, with front-wheel drive on the top and all-wheel drive on the bottom.

ϭϬͬϳͬE'

 Ε с 

ϰϬϬsͲ   с Ε 

ϭϬͬϳͬE'

 Ε с 

Figure 4.5:

ϰϬϬsͲ 

 с Ε 

P2-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility, for front-wheel drive (top) and all-wheel drive (bottom).

For this powertrain architecture, two different engine types are examined: the budget optimized gasoline engine and the diesel engine. The overall system power matches the sum of the internal combustion engine power and the electric drive power. The maximum speed of the vehicle is limited to 250 km / h. An exception are the light-duty vehicles for which the maximum speed is limited to 130 km / h. The powertrain is mounted in front transverse position and

4 Powertrain Simulation

118

uses a 9-speed automated gearbox for the sedan and sport utility vehicle and a 6-speed automated gearbox for the light-duty vehicle. The all-wheel drive is realized with an electronically controlled multi plate clutch unit, driving the rear axle of the vehicle. An electric permanent magnet synchronous motor with a power between 40 kW and 75 kW is mounted between engine output and gearbox input. The mean voltage level of the battery is 400 V. For the battery, the medium power battery cells are chosen. The simulation variations for the P2-hybrid with 400 V battery and external recharge possibility are summarized in Table 4.5. Table 4.5:

Simulation variations of the P2-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility.

Engine Type

ID

Internal combustion engine power

Electric drive power

Battery size

Gears

Sedan Gasoline – Budget Optimized C.

22

60 kW

40 kW

16 kWh

9

Diesel

25

60 kW

40 kW

18 kWh

9

Sport utility vehicle Gasoline – Budget Optimized C.

23

125 kW

75 kW

24 kWh

9

Diesel

26

125 kW

75 kW

28 kWh

9

Light-duty vehicle Gasoline – Budget Optimized C.

24

75 kW

75 kW

30 kWh

6

Diesel

27

75 kW

75 kW

33 kWh

6

4.3 Powertrain Design

119

The fifth powertrain architecture is a serial-hybrid electric powertrain (range extender) with a 400 V battery system and external recharge possibility. Again, this powertrain configuration is examined in all three vehicle types. Figure 4.6 shows the serial-powertrain architecture for the 400 V-system with external recharge possibility, with two-wheel drive on the top and all-wheel drive on the bottom.

Figure 4.6:

Serial-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility, for two-wheel drive (top) and all-wheel drive (bottom).

For this powertrain architecture, only the range extender concept of the gasoline engine is used. The full overall system power is provided by one induction

120

4 Powertrain Simulation

motor axle drive (front- or rear-wheel drive) or two induction motor axle drives (all-wheel drive). For the sedans the axle drive is installed on the rear and for the light-duty vehicles on the front axle. Two different gearboxes for the electric axle drives are examined: a 1-speed gearbox and a 2-speed gearbox. The maximum speed of the vehicle is limited to 160 km / h for the 1speed gearbox, due to the bad low-end torque, and 250 km / h for the 2-speed gearbox. An exception are the light-duty vehicles for which the maximum speed is limited to 130 km / h. A permanent magnet synchronous motor matches the internal combustion engine power and is used as a generator for extending the range of the vehicle. The mean voltage level of the battery is 400 V. For the battery, the medium power battery cells are used. The simulation variations for the serial-hybrid electric powertrain with 400 V battery and external recharge possibility are listed in Table 4.6. The sixth powertrain architecture is a serial-parallel-hybrid powertrain with a 400 V battery system and external recharge possibility. This powertrain architecture is also examined in all three vehicle types. The serial-parallel-powertrain configuration for the 400 V-system with external recharge possibility are shown in Figure 4.7, with front-wheel drive on the top and all-wheel drive on the bottom. For this powertrain architecture, the range extender concept of the gasoline engine and the diesel engine is used. The front-wheel-drive uses an induction motor, matching the overall system power. In addition, a permanent magnet synchronous motor matching the power of the combustion engine is installed. Both motors are mounted between engine output and gearbox input. For the all-wheel drive the overall system power is provided half by electric axle drive with induction motor on the rear axle and the other half by a permanent magnet synchronous motor on the front axle, mounted between engine output and gearbox input. The permanent magnet synchronous motor can either been used in serial operating mode as generator or in parallel operating mode as motor- / generator-unit. For the front axle, two different gearboxes are provided: a 2-speed gearbox and a 4-speed gearbox. The rear axle always uses a 2-speed gearbox. The maximum speed of the vehicle is limited to 250 km / h. An exception are the light-duty vehicles for which the maximum speed is limited to 130 km / h. The mean voltage level of the battery is 400 V. For the battery, the medium power battery cells are used. The simulation variations for the serial-parallel-hybrid with 400 V battery and external recharge possibility are summarized in Table 4.7 and Table 4.8. For front-wheel drive the permanent magnet synchronous motor is not part of the electric drive power

4.3 Powertrain Design

121

but can be used as such. The internal combustion engine is not part of the overall drive power, but can be used as such. Table 4.6:

Simulation variations of the serial-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility.

Engine Type

ID

Internal combustion engine power

Electric drive power

Battery size

Gears (front /rear)

Sedan Gasoline – Range Extender C.

28

60 kW

100 kW

32 kWh

1/-

Gasoline – Range Extender C.

31

60 kW

100 kW

32 kWh

2/-

Sport utility vehicle Gasoline – Range Extender C.

29

80 kW

200 kW

40 kWh

1/1

Gasoline – Range Extender C.

32

80 kW

200 kW

40 kWh

2/2

Light-duty vehicle Gasoline – Range Extender C.

30

60 kW

150 kW

48 kWh

1/-

Gasoline – Range Extender C.

33

60 kW

150 kW

48 kWh

2/-

122

Figure 4.7:

4 Powertrain Simulation

Serial-parallel-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility, for frontwheel drive (top) and all-wheel drive (bottom).

4.3 Powertrain Design Table 4.7:

123

Simulation variations of the serial-parallel-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility for sedan and SUV.

Engine Type

ID

Internal combustion engine power

Gears

Electric drive power

Battery size

(front/ rear)

Sedan Gasoline – Range Extender C.

40

60 kW

100 kW

17 kWh

2/-

Gasoline – Range Extender C.

43

60 kW

100 kW

17 kWh

4/-

Diesel

46

60 kW

100 kW

17 kWh

2/-

Diesel

49

60 kW

100 kW

16 kWh

4/-

Sport utility vehicle Gasoline – Range Extender C.

41

80 kW

200 kW

22 kWh

2/2

Gasoline – Range Extender C.

44

80 kW

200 kW

22 kWh

4/2

Diesel

47

80 kW

200 kW

22 kWh

2/2

Diesel

50

80 kW

200 kW

22 kWh

4/2

4 Powertrain Simulation

124 Table 4.8:

Simulation variations of the serial-parallel-hybrid electric powertrain architecture for a 400 V battery system with external recharge possibility for the light-duty-vehicle.

Engine Type

ID

Internal combustion engine power

Electric drive power

Gears Battery size

(front/ rear)

Light-duty vehicle Gasoline – Range Extender C.

42

60 kW

150 kW

30 kWh

2/-

Gasoline – Range Extender C.

45

60 kW

150 kW

28 kWh

4/-

Diesel

48

60 kW

150 kW

28 kWh

2/-

Diesel

51

60 kW

150 kW

28 kWh

4/-

The seventh powertrain architecture is a battery electric powertrain with a fuel cell system as range extender and a 400 V battery system with external recharge possibility. This powertrain configuration, once more, is examined in all three vehicle types. The fuel cell electric powertrain architecture for the 400 V-system with external recharge possibility is seen in Figure 4.8, with two-wheel drive on the top and all-wheel drive on the bottom. For this powertrain architecture, the fuel cell system is used as range extender. For twowheel-drive, an induction motor axle drive is used on the front or rear axle, matching the overall system power. For the sedans the axle drive is installed on the rear and for the light-duty vehicles on the front axle. For all-wheel drive the overall system power is provided half by electric axle drive with induction motor on the rear axle and for the other half by an induction motor axle drive on the front axle. Two different gearboxes for the electric axle drives are examined: a 1-speed gearbox and a 2-speed gearbox. The maximum speed of the vehicle is limited to 160 km / h for the 1-speed gearbox due to the bad lowend torque and 250 km / h for the 2-speed gearbox. An exception are the light-

4.3 Powertrain Design

125

duty vehicles for which the maximum speed is limited to 130 km / h. The mean voltage level of the battery is 400 V. For the battery, the medium power battery cells are applied. The simulation variations for the fuel cell electric powertrain with 400 V battery and external recharge possibility are summarized in Table 4.9.

Figure 4.8:

Fuel cell electric powertrain architecture for a 400 V battery system with external recharge possibility, for two-wheel drive (top) and all-wheel drive (bottom).

4 Powertrain Simulation

126 Table 4.9:

Simulation variations of the fuel cell electric powertrain architecture for a 400 V battery system with external recharge possibility.

Fuel Cell Type

ID

Fuel cell power

Electric drive power

Battery size

Gears (front/ rear)

Sedan Proton-exchange membrane

34

60 kW

100 kW

30 kWh

1/-

Proton-exchange membrane

37

60 kW

100 kW

30 kWh

2/-

Sport utility vehicle Proton-exchange membrane

35

80 kW

200 kW

39 kWh

1/1

Proton-exchange membrane

38

80 kW

200 kW

39 kWh

2/2

Light-duty vehicle Proton-exchange membrane

36

60 kW

150 kW

46 kWh

1/-

Proton-exchange membrane

39

60 kW

150 kW

46 kWh

2/-

The final powertrain architecture is a battery electric powertrain with an 800 V battery system. This powertrain configuration is examined in all three vehicle types. Figure 4.9 shows the electric powertrain architecture for the 800 V-, with front-wheel drive on the top and all-wheel drive on the bottom. For the two-wheel-drive, an induction motor axle drive is used on the front axle, matching the overall system power. For the all-wheel drive the overall system

4.3 Powertrain Design

127

power is provided half by electric axle drive with induction motor on the rear axle and for the other half by an induction motor axle drive on the front axle. Two different gearboxes for the electric axle drives are examined, including a 1-speed gearbox and a 2-speed gearbox. The maximum speed of the vehicle is limited to 160 km / h for the 1-speed gearbox due to the bad low-end torque and 250 km / h for the 2-speed gearbox. For the light-duty vehicles the maximum speed is limited to 130 km / h. The mean voltage level of the battery is 800 V. For the battery, the high energy battery cells are used. The simulation variations for the battery electric powertrain with 800 V battery are summarized in Table 4.10.

Figure 4.9:

Full electric powertrain architecture for an 800 V battery system, for front-wheel drive (top) and all-wheel drive (bottom).

4 Powertrain Simulation

128 Table 4.10:

ID

Simulation variations of the full electric powertrain architecture for an 800 V battery system.

Electric drive power

Battery size

Gears (front/rear)

Sedan 52

100 kW

140 kWh

1/-

55

100 kW

140 kWh

2/-

Sport utility vehicle 53

200 kW

180 kWh

1/1

56

200 kW

180 kWh

2/2

Light-duty vehicle 54

150 kW

115 kWh

1/-

57

150 kW

115 kWh

2/-

4.4 Drive Cycles

4.4

129

Drive Cycles

In this section, a variety of drive cycles are introduced to match the different use cases for the vehicles. The basic data for the driving cycles is retrieved from [61]. Figure 4.10 shows the testing environment with the associated driving cycles.

Primary Driving Cycle (Charge-sustaining) (+ Trailer)

City Cycle Testing under different conditions

Commuter Cycle

Maximum Range Cycle (+ Trailer)

Primary Driving Cycle (Charge-sustaining)

City Cycle Testing under different conditions

Urban Delivery Cycle

Maximum Range Cycle

Figure 4.10:

Driving cycles and test conditions for sedan, sports utility vehicle (SUV) and light-duty vehicle (LDV).

130

4 Powertrain Simulation

The Primary Driving Cycle represents the reference cycle as the reference use case and is used for the calibration of the operating strategy of the vehicle. The City Cycle represents the use case for urban driving. The Commuter Cycle represents the use case for a daily work commuter drive. The Urban Delivery Cycle replaces the Commuter Cycle for the light-duty vehicles and represents the use case for an urban delivery vehicle day tour. The Maximum Range Cycle is used to determine the tank and battery size and represents a long distance travelling scenario. All drive cycles are simulated in charge-depleting mode, except for the Primary Driving Cycle, which is always simulated in chargesustaining mode. A precise RDE-analysis [62] [63] [64] is described in Appendix A2 and available for all drive cycles as downloadable content. In the following, a short overview of these drive cycles is given. The Primary Driving Cycle is a synthetic RDE-Cycle that uses modified measurement data from [61]. The cycle has no altitude difference between start and end of the cycle. The cycle is a high dynamic cycle that is close to the upper dynamic limit of RDE-conformity. The cycle has an overall travelling distance of 97 km with a distance distribution of urban/rural/motorway of 1/1/1. The cycle shows full RDE-conformity. Simulations of the Primary Driving Cycle always use the charge-sustaining mode of the battery to show the impact of high dynamic driving to the powertrain components. The Primary Driving Cycle also includes additional modes with speed limitations of 100 km / h and 80 km / h, respectively for driving with a trailer and for a light-duty vehicles over 3.5 t gross weight. All modes of the Primary Driving Cycle can be seen in Figure 4.11, with the reference cycle on the top, the cycle for 100 km / h speed limitation in the middle and the cycle for 80 km / h speed limitation at the bottom. The City Cycle consists of twice the worldwide harmonized light vehicles test cycle - Class 3 low and medium parts with an 1800 s break in between. Each of the two identical sections has a length of 7.85 km, simulating an urban drive to a destination with approach, 30 min stay and following departure. The cycle does not show full RDE-conformity due to missing motorway shares, overall trip duration and the 30 min break. Simulations with the City Cycle are always performed in the charge-depleting mode of the battery to show a maximum electric urban driving scenario. The City Cycle is shown in Figure 4.12.

4.4 Drive Cycles

600 speed [km / h]

altitude [m]

500

150

400

100

300 200

50

altitude [m]

200 speed [km / h]

131

100

0

0 0

2000

4000

6000

time [s] 600 altitude [m]

500

100

400 300 50

200

altitude [m]

speed [km / h]

speed [km / h]

100 0

0 0

2000

4000

6000

time [s] 600 speed [km / h]

altitude [m]

500

80

400

60

300

40

200

20

100

0

altitude [m]

speed [km / h]

100

0 0

2000

4000

6000

time [s] Figure 4.11:

Speed and altitude of the Primary Driving Cycle with no speed limitation (top), 100 km / h speed limitation (mid) and 80 km / h speed limitation (bottom).

4 Powertrain Simulation

132

The Commuter Cycle is a synthetic RDE-Cycle that uses modified measurement data from [61]. The cycle has no altitude difference between start and end of the cycle and is a medium to high dynamic cycle regarding RDE-conformity. The cycle has an overall travelling distance of 40 km with a distance distribution of urban/rural of 3/2. The cycle does not show full RDE-conformity due to missing motorway shares and overall trip duration. Simulations of the Commuter Cycle always use the charge-depleting mode of the battery to show the impact of electric driving on the fuel consumption of the vehicle. The Commuter Cycle is seen in Figure 4.13. 100

600 altitude [m]

500

80

400

60

300

40

200

20

altitude [m]

speed [km / h]

speed [km / h]

100

0

0 0

500

1000

1500

2000

2500

time [s] Figure 4.12:

Speed and altitude of the City Cycle. 600 speed [km / h]

100

altitude [m]

500

80

400

60

300

40

200

20

100

0

0 0

1000

2000

3000

time [s] Figure 4.13:

Speed and altitude of the Commuter Cycle.

altitude [m]

speed [km / h]

120

4.4 Drive Cycles

133

The Urban Delivery Cycle is a synthetic RDE-Cycle that uses modified measurement data from [61]. The cycle has no altitude difference between start and end of the cycle and is a medium dynamic cycle regarding RDE-conformity. The cycle has an overall travelling distance of 150 km with a distance distribution urban/rural of 3/1. In addition, the cycle includes four 1600 s breaks to simulate loading and unloading scenarios. The cycle does not show full RDEconformity due to missing motorway shares, overall trip duration and the four 1600 s breaks. Simulations of the Urban Delivery Cycle always use the chargedepleting mode of the battery to show the maximum use of the electric driving range. The Urban Delivery Cycle is shown in Figure 4.14. 600 speed [km / h]

altitude [m]

400 50 200 0

altitude [m]

speed [km / h]

100

0 0

Figure 4.14:

5000

10000

15000 time [s]

20000

Speed and altitude of the Urban Delivery Cycle.

The Maximum Range Cycle is a synthetic RDE-Cycle that uses modified measurement data from [61]. The cycle has no altitude difference between start and end of the cycle. The cycle is a high dynamic cycle that is close to the upper dynamic limit of RDE-conformity. The cycle has an overall travelling distance of 400 km with a distance distribution of urban/rural/motorway of 1/2/15. The cycle does not show full RDE-conformity due to overweight motorway shares, high vehicle speed and overall trip duration. Simulations of the Maximum Range Cycle always use the charge-depleting mode of the battery to show the impact of all energy storages to the overall energy consumption. The Maximum Range Cycle also has additional modes with speed limitations of 100 km / h and 80 km / h, respectively for driving with a trailer and for a light-duty vehicle over 3.5 t gross weight. All modes of the Maximum Range Cycle are illustrated in Figure 4.15, with the reference cycle on the top, the cycle for 100 km / h speed limitation in the middle and the cycle for 80 km / h speed limitation at the bottom.

4 Powertrain Simulation

134

600 speed [km / h]

altitude [m]

500

150

400

100

300 200

50

altitude [m]

speed [km / h]

200

100

0

0 0

5000

10000

15000

time [s] 600 altitude [m]

500

100

400 300 50

200

altitude [m]

speed [km / h]

speed [km / h]

100 0

0 0

5000

10000

15000

20000

time [s] 600 speed [km / h]

altitude [m]

500

80

400

60

300

40

200

20

100

0

altitude [m]

speed [km / h]

100

0 0

5000

10000

15000

20000

time [s] Figure 4.15:

Speed and altitude of the Maximum Range Cycle with no speed limitation (top), 100 km / h speed limitation (mid) and 80 km / h speed limitation (bottom).

4.5 Operating Strategies

4.5

135

Operating Strategies

In this section, the operating strategies applied to the different powertrain configurations are introduced using [65] as a baseline. For that, the Equivalent Consumption Minimization Strategy (ECMS) first introduced by Paganelli [66] is modified and adapted to fit the different powertrain architectures. The ECMS is a heuristic method to solve an optimal control problem, using partial optimizations to reach to global optimum. The approach is supported by Bellman’s principal of optimality which says in [67] that: “An optimal policy has the property that whatever the initial state and initial decision are, the remaining decisions must constitute an optimal policy with regard to the state resulting from the first decision.” Regarding the ECMS, that does not necessarily mean that the solution is always the optimal solution, due to further application parameters. But by finding the optimal application parameters the solution can reach its optimal solution for the underlying control problem. The base function for the ECMS is defined as:

‫ ܥ‬ൌ  ܲ௙௨௘௟ ൅ ߣ ‫ܲ כ‬௘௟Ǥ 

‡“ǤͶǤͳͷ

with ‫ ܥ‬being the overall cost for distributing or retrieving power, ܲ௙௨௘௟ being the used chemical power referring to the lower heating value, ܲ௘௟Ǥ being the distributed or retrieved power from the battery and ɉ being the equivalence factor between the two energy storages. The overall cost ‫ ܥ‬is then calculated for each possible operating point and thus delivers the minimum overall energy consumption for a pre-defined problem by choosing the minimum value of the overall costs ‫ܥ‬. The first problem addressed is the control of the battery electric powertrain. In this application, only one energy storage is used, which means that the equivalence factor drops out of the equation so that the operating strategy is defined as Consumption Minimization Strategy. For more than one energy storage, the equivalence factor must be considered as an optimization parameter to reach the optimal solution. In the following subsections, the different operating strategies used in the simulations are defined in more detail.

4 Powertrain Simulation

136 4.5.1

Consumption Minimization Strategy

In this subsection, the Consumption Minimization Strategy is explained. This strategy functions as a base for the other operating strategies and is also the operating strategy for the battery electric vehicles with only one energy storage. For that, the equation (eq. 4.15) for only one energy storage, being the battery, simplifies to:

‫ ܥ‬ൌ ܲ௘௟Ǥ 

‡“ǤͶǤͳ͸

Now the cost  equals the current electric energy consumption of the vehicle. The electric energy consumption in a current time step is now calculated over the whole range of each control variable. A scheme of this process can be seen in Figure 4.16. In this example two e-axle drives are assumed, one on the front axle and one on the rear axle of the vehicle. Each e-axle has a 2-speed gearbox that can be switched separately and has no fixed switching curve. The third control variable is the torque distribution between the two drive units, henceforth referred to as torque split. For the hybrid powertrains using Equivalent Consumption Minimization Strategy the torque split can also be between a combustion engine and the electric drive systems. In this case, the equivalence factor ɉ is used and optimized to reach optimal energy balance between the two energy storages. In each time step of the drive cycle a matrix is calculated including the electric energy effort in each gear for each e-axle and each torque split. The overall dimension of the cost matrix is determined by the number of drive units and the number of gearboxes. After calculating the cost matrix , the minimum value of the cost matrix is chosen, henceforth referred to as ‫ܥ‬௠௜௡ . The matrix address of the value ‫ܥ‬௠௜௡ now gives the optimal setup of control variables for a current time step. This calculation is repeated for each time step of the drive cycle resulting in the optimal solution for the underlying control problem.

137

Gear Front Axle

4.5 Operating Strategies

Gear Rear Axle

Figure 4.16:

4.5.2

Scheme of the Consumption Minimization Strategy for three control variables.

Equivalent Consumption Minimization Strategy for parallel hybrid powertrains

In this subsection the Equivalent Consumption Minimization Strategy for parallel hybrid powertrains is introduced. For calculating the minimum value of the cost matrix ‫ܥ‬௠௜௡ the same strategy as in subsection 4.5.1 is used. Due to the setup with two energy storages the equivalence factor ɉ is needed to optimize the energy balance between the used fuel and the electric energy effort. Thus, ɉ can not be a constant value but has to be a function over the battery state of charge. This is because for a high state of charge the powertrain is intended to use the electric energy for driving to save fuel, while for a low state of charge the combustion engine is intended to be used to save electric energy and preserve the battery from discharging completely. The general approach for the equivalence factor can be seen in Figure 4.17. In this example the equivalence factor for a hybrid electric vehicle with no external recharge possibility is shown. The function for the equivalence factor ߣ shows a constant plateau between 45 % and 55 % state of charge. For higher states of charge value of the equivalence factor decreases, resulting in more

4 Powertrain Simulation

138

electric driving; for lower states of charge the value of the equivalence factor increases, which results in a higher use of the internal combustion engine. The first optimization parameter is the base equivalence factor (OPT(1)), the second optimization parameter (OPT(2)) is the gradient of the equivalence factor function. By optimizing these two parameters, an optimized operating strategy is found that leads to an optimal control of the vehicles powertrain. For different powertrain architectures different functions for the equivalence factor are possible. The most important function characteristics of the equivalence factor for the hybrid electric vehicle without external recharge possibility (HEV) and with external recharge possibility (PHEV) are shown in Figure 4.18. Figure 4.17:

General approach for the equivalence factor in dependency of

OPT(2)

λ [-]

OPT(1)

0

20

40

60

80

100

SOC [%] the state of charge.

HEV (catalyst heating) PHEV (high load)

λ [-]

HEV PHEV

0

20

40

60

80

100

SOC [%] Figure 4.18:

Equivalence factor functions for different conditions and different powertrains.

4.5 Operating Strategies

139

For the hybrid electric vehicle without external recharge possibility (HEV), the two functions of the equivalence factor are shown. The first one is the standard curve that is used for electric driving and hybrid driving with heated exhaust gas aftertreatment system. In this context heated means that the exhaust gas aftertreatment works with at least 95 % of its temperature dependent conversion rate. If this conversion rate is not reached, the “HEV (catalyst heating)” curve is used. Due to the higher values for the equivalence factor, the engine aims for higher operating points and by that increases the exhaust gas temperature which leads to faster heating of the catalysts. For the hybrid electric vehicle with external recharge possibility (PHEV), two possible equivalence factor curves can also be seen. In general, the equivalence factor for the PHEVs can reach almost zero, which means that the internal combustion engine does not start up at any time, except the maximum traction torque of the electric drive systems is exceeded. The hybrid mode only starts when the state-of-charge of the battery drops below 30 % or a traction power exceeding-event occurs. For high vehicle loads, such as the tarpaulin trailer, a complete discharge of the battery and a malfunction of the powertrain is possible, due to the higher amount of energy circulating in the system. Hence, an equivalence factor curve for high loads “PHEV (high load)” is used. This curve, due to higher states-of-charge-operation, prevents the battery from fully discharging and thereby preventing a powertrain malfunction. Also, for the hybrid electric vehicle with external recharge possibility, a function for catalyst heating exists, but as the behaviour of this function is similar to the one for the hybrid electric vehicle without external recharge possibility, it is not presented in more detail.

4.5.3

Equivalent Consumption Minimization Strategy for serial hybrid powertrains

The equivalent consumption minimization strategy for serial hybrid powertrains is similar to the one for the parallel powertrains, except that the torque split from section 4.5.1 is replaced by a power split between the range extender unit respectively the fuel cell system and the battery. Therefor an optimal consumption curve for the range extender unit or the fuel cell system is calculated to match different possible power demands. After that, the overall cost function  for the different possible power flows between traction machines, battery and range extender unit are calculated. The serial hybrid powertrains are

4 Powertrain Simulation

140

all PHEV powertrains and use the same characteristics for the equivalence factor ߣ as the PHEV powertrains in Figure 4.18, but with different optimization parameter values.

4.5.4

Selective Equivalent Consumption Minimization Strategy for serial/parallel hybrid powertrains

Gear Front Axle

Gear Front Axle

For the serial parallel hybrid powertrains a simple approach is chosen, as shown in Figure 4.19. For each time step, two cost functions ௣ and ௦ are calculated, one for the parallel operating mode and one for the serial operating mode. Then, the two cost functions in the current time step are minimized for each operating mode. The last step is selecting the minimum cost value from the two functions, and by that determining the current most efficient operating mode.

Gear Rear Axle

Gear Rear Axle

‫ܥ‬௠௜௡ǡ௚௘௦ ൌ ‹ሺ‫ܥ‬௣ǡ௠௜௡ ǡ ‫ܥ‬௦ǡ௠௜௡ Ϳ

‫ܥ‬௠௜௡ǡ௚௘௦

Figure 4.19:

4.5.5

Scheme of Equivalent Consumption Minimization Strategy for serial/parallel hybrid powertrains.

Additional Control Strategy Parameters

Additional control strategy parameters and functions are used to avoid undesired behaviour of the powertrains. These parameters are implemented in the

4.5 Operating Strategies

141

control strategy and are part of the optimal solution that is found for each powertrain. These parameters may increase the energy consumption but can also be seen as part of the optimal solution, as they suppress undesired behaviour. For the following behaviours, parameters are implemented in the powertrains: x x

x

x

The internal combustion engine must be “on” for at least 40 s if the vehicle is moving. The internal combustion engine should normally work at 3500 min-1 or less for gasoline and natural gas engines and at 2500 min-1 or less for diesel engines. The electrically heated catalysts in the exhaust gas aftertreatment systems of the internal combustion engines are heated up to 95 % of their temperature depended conversion rate when the engine is on. A speed independent gear switching hysteresis is implemented to avoid constant gear switching.

All these measures will cause around 10 % increase in fuel consumption for internal combustion engine powertrains. For electric powertrains, this value will decrease significantly, due to the fact that only the gear switching hysteresis is used.

5 Results of Powertrain Simulation In this chapter, the results of the simulation are summarized and discussed. The classification is based on the vehicle types. For each vehicle type the overall consumption of primary energy is presented, with a further distinction between chemical and electric energy. The chemical energy refers to the lower heating value of the used chemical energy source. The second criterion taken into consideration is the overall local CO2-emissions of each powertrain. Each discussion is made for all drive cycles to emphasize the differences between different driving conditions. For the CO2-emission factors of the fuels (as shown in Table 5.1), the values are either retrieved from [68] or calculated. Table 5.1:

CO2-emisson factors for the different used fuels. E10 [kg / l]

B7 [kg / l]

Methane [kg / kg]

2.21

2.65

2.74

CO2-emissions factor

5.1

Sedan

In this section, the energy consumption and local CO2-emissions of the sedan for the different powertrain architectures are discussed. The performed driving cycles are the open Primary Driving Cycle, the City Cycle, the Commuter Cycle and the open Maximum Range Cycle. The results for the sedan driving the Primary Driving Cycle with no speed limitation are shown in Figure 5.1. Regarding the overall energy consumption, it can be observed that for the hybrid powertrains with no external recharge possibility (01 – 19) the energy consumption decreases with the degree of electrification, respectively with the degree of hybridization. For the hybrid powertrains with external recharge possibility (22 – 49), the P2-powertrain architectures (22, 25) show the lowest chemical energy consumption of all internal combustion engine powertrains. The S/P2-powertrain architectures (40 – 49) © The Author(s), under exclusive license to Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2023 T. Stoll, A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040, Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart, https://doi.org/10.1007/978-3-658-42168-7_5

5 Results of Powertrain Simulation

144

55 - Sedan - Battery - ED - 800 V - 2SP 52 - Sedan - Battery - ED - 800 V - 1SP 49 - Sedan - Diesel - S/P2 - 400 V - 4SP 46 - Sedan - Diesel - S/P2 - 400 V - 2SP 43 - Sedan - Gasoline - S/P2 - 400 V - 4SP 40 - Sedan - Gasoline - S/P2 - 400 V - 2SP 37 - Sedan - Fuel Cell - S - 400 V - 2SP 34 - Sedan - Fuel Cell - S - 400 V - 1SP 31 - Sedan - Gasoline - S - 400 V - 2SP 28 - Sedan - Gasoline - S - 400 V - 1SP 25 - Sedan - Diesel - P2 - 400 V - 9SP 22 - Sedan - Gasoline - P2 - 400 V - 9SP 19 - Sedan - Diesel - P2 - 400 V - 9SP 16 - Sedan - CNG - P2 - 400 V - 9SP 13 - Sedan - Gasoline - P2 - 400 V - 9SP 10 - Sedan - Diesel - P2 - 48 V - 9SP 07 - Sedan - CNG - P2 - 48 V - 9SP 04 - Sedan - Gasoline - P2 - 48 V - 9SP 03 - Sedan - Diesel - P0 - 48 V - 9SP 02 - Sedan - CNG - P0 - 48 V - 9SP 01 - Sedan - Gasoline - P0 - 48 V - 9SP 0 electric energy consumption [kWh / 100 km]

10

20

30

40

50

chemical energy consumption [kWh / 100 km]

55 - Sedan - Battery - ED - 800 V - 2SP 52 - Sedan - Battery - ED - 800 V - 1SP 49 - Sedan - Diesel - S/P2 - 400 V - 4SP 46 - Sedan - Diesel - S/P2 - 400 V - 2SP 43 - Sedan - Gasoline - S/P2 - 400 V - 4SP 40 - Sedan - Gasoline - S/P2 - 400 V - 2SP 37 - Sedan - Fuel Cell - S - 400 V - 2SP 34 - Sedan - Fuel Cell - S - 400 V - 1SP 31 - Sedan - Gasoline - S - 400 V - 2SP 28 - Sedan - Gasoline - S - 400 V - 1SP 25 - Sedan - Diesel - P2 - 400 V - 9SP 22 - Sedan - Gasoline - P2 - 400 V - 9SP 19 - Sedan - Diesel - P2 - 400 V - 9SP 16 - Sedan - CNG - P2 - 400 V - 9SP 13 - Sedan - Gasoline - P2 - 400 V - 9SP 10 - Sedan - Diesel - P2 - 48 V - 9SP 07 - Sedan - CNG - P2 - 48 V - 9SP 04 - Sedan - Gasoline - P2 - 48 V - 9SP 03 - Sedan - Diesel - P0 - 48 V - 9SP 02 - Sedan - CNG - P0 - 48 V - 9SP 01 - Sedan - Gasoline - P0 - 48 V - 9SP 0

20

40

60

80

100

120

CO2-emissions [g / km]

Figure 5.1:

Energy consumption (top) and CO2-emissions (bottom) for the sedan driving the Primary Driving Cycle (open).

5.1 Sedan

145

show a higher chemical energy consumption than the P2-powertrains. The Spowertrains with internal combustion engine (28, 31) show a higher chemical energy consumption than the P2- and S/P2-powertrain architectures, due to the efficiency decrease caused by the used generator unit inducing an additional conversion step to electrical energy. In contrast, the S-powertrains with the fuel cell system (34, 37) show the lowest chemical energy consumption of all powertrains, due to the high peak efficiency of the fuel cell system of around 60 %. The battery electric powertrains (52, 55) show the lowest overall energy consumption due to the high efficiency of the battery and absence of chemical energy converters. For the different internal combustion engine concepts, the gasoline engines show the lowest chemical energy consumption followed by the diesel and the natural gas engines (CNG). The higher energy consumption of the diesel engines is caused by the electrical heating system for the selective catalytic reduction catalyst and the low exhaust gas temperatures of these engines, resulting in a high use of the electrical catalyst heating system. The natural gas engines in general show a lower efficiency due to a higher air-to-fuel ratio, resulting in higher charge exchange losses for these engine concepts. For the CO2-emissions, it can be said that the battery electric powertrains and the S-powertrains with fuel cell system are locally emission free. The CO2emissions of the internal combustion engines show close correlation with the chemical energy consumption. The natural gas engines show the lowest CO2emissions of all engine concepts, due to the high hydrogen-to-carbon ratio of the fuel. Regarding the CO2-emissions, the natural gas engines are followed by the gasoline engines, due to their lower overall energy consumption and better hydrogen-to-carbon ratio compared to the diesel engines. The diesel engines show the highest CO2-emissions of all engines. The results for the Maximum Range Cycle with no speed limitation are shown in Figure 5.2. The results are similar to the results for the Primary Driving Cycle and will therefore not be discussed separately. The results for the sedan driving the City Cycle are summarized in Figure 5.3. Regarding the overall energy consumption, it can be observed that for the hybrid powertrains with no external recharge possibility (01 – 19), the overall energy consumption decreases with the degree of electrification or respectively with the degree of hybridization. For the hybrid powertrains with external recharge possibility (22 – 49) all powertrains have a low overall energy consumption due to full electric driving in the City Cycle. The S/P2-powertrain architectures (40 – 49) show the lowest overall energy consumption of

5 Results of Powertrain Simulation

146

55 - Sedan - Battery - ED - 800 V - 2SP 52 - Sedan - Battery - ED - 800 V - 1SP 49 - Sedan - Diesel - S/P2 - 400 V - 4SP 46 - Sedan - Diesel - S/P2 - 400 V - 2SP 43 - Sedan - Gasoline - S/P2 - 400 V - 4SP 40 - Sedan - Gasoline - S/P2 - 400 V - 2SP 37 - Sedan - Fuel Cell - S - 400 V - 2SP 34 - Sedan - Fuel Cell - S - 400 V - 1SP 31 - Sedan - Gasoline - S - 400 V - 2SP 28 - Sedan - Gasoline - S - 400 V - 1SP 25 - Sedan - Diesel - P2 - 400 V - 9SP 22 - Sedan - Gasoline - P2 - 400 V - 9SP 19 - Sedan - Diesel - P2 - 400 V - 9SP 16 - Sedan - CNG - P2 - 400 V - 9SP 13 - Sedan - Gasoline - P2 - 400 V - 9SP 10 - Sedan - Diesel - P2 - 48 V - 9SP 07 - Sedan - CNG - P2 - 48 V - 9SP 04 - Sedan - Gasoline - P2 - 48 V - 9SP 03 - Sedan - Diesel - P0 - 48 V - 9SP 02 - Sedan - CNG - P0 - 48 V - 9SP 01 - Sedan - Gasoline - P0 - 48 V - 9SP 0

10

electric energy consumption [kWh / 100 km]

20

30

40

50

chemical energy consumption [kWh / 100 km]

55 - Sedan - Battery - ED - 800 V - 2SP 52 - Sedan - Battery - ED - 800 V - 1SP 49 - Sedan - Diesel - S/P2 - 400 V - 4SP 46 - Sedan - Diesel - S/P2 - 400 V - 2SP 43 - Sedan - Gasoline - S/P2 - 400 V - 4SP 40 - Sedan - Gasoline - S/P2 - 400 V - 2SP 37 - Sedan - Fuel Cell - S - 400 V - 2SP 34 - Sedan - Fuel Cell - S - 400 V - 1SP 31 - Sedan - Gasoline - S - 400 V - 2SP 28 - Sedan - Gasoline - S - 400 V - 1SP 25 - Sedan - Diesel - P2 - 400 V - 9SP 22 - Sedan - Gasoline - P2 - 400 V - 9SP 19 - Sedan - Diesel - P2 - 400 V - 9SP 16 - Sedan - CNG - P2 - 400 V - 9SP 13 - Sedan - Gasoline - P2 - 400 V - 9SP 10 - Sedan - Diesel - P2 - 48 V - 9SP 07 - Sedan - CNG - P2 - 48 V - 9SP 04 - Sedan - Gasoline - P2 - 48 V - 9SP 03 - Sedan - Diesel - P0 - 48 V - 9SP 02 - Sedan - CNG - P0 - 48 V - 9SP 01 - Sedan - Gasoline - P0 - 48 V - 9SP 0

20

40

60

80

100

120

CO2-emissions [g / km]

Figure 5.2:

Energy consumption (top) and CO2-emissions (bottom) for the sedan driving the Maximum Range Cycle (open).

5.1 Sedan

147

all vehicles compared to the battery electric vehicles (52, 55), due to low powertrain weight, high gear switching capability, high electric driving power and the use of a permanent magnet synchronous motor. The electric energy consumption of the S/P2- powertrain architectures is followed by the S-powertrains (38 – 37), also with high electric driving power and the P2-architectures (22, 25), with lower electric driving power. The P2-architectures show a significantly high electric energy consumption for the diesel powertrain (25), due to the heating strategy for the electrically heated selective catalytic reduction catalyst. It must be assumed that the engine is started, even when it does not deliver any torque. For correcting this value, a more complex heuristic operating strategy needs to be implemented, but that is not part of this project. In conclusion, the electric energy consumption of (25) is not used for further investigations. Besides (25) the battery electric vehicles show the highest electric energy consumption due to the high powertrain weight of these concepts caused by the battery. For the different internal combustion engine concepts, the gasoline engines show the lowest chemical energy consumption followed by the diesel and the natural gas engines (CNG). The higher overall energy consumption of the diesel engines is caused by the electrical heating system for the selective catalytic reduction catalyst and the low exhaust gas temperatures of these engine concepts, resulting in a high use of the electrical catalyst heating system. The natural gas engines generally have lower efficiency because they have a higher air-fuel ratio, which leads to higher charge exchange losses in this engine concept. Regarding the CO2-emissions, all hybrid powertrains with external recharge possibility as well as the battery electric powertrains are locally emission free. The CO2-emissions of the internal combustion engines show close correlation with the chemical energy consumption. The natural gas engines show the lowest CO2-emissions of all engine concepts, due to the high hydrogen-to-carbon ratio of the fuel. The CO2-emissions of the natural engines are followed by the gasoline engines, due to their lower chemical energy consumption and better hydrogen-to-carbon ratio compared to the diesel engines. The diesel engines show the highest CO2-emissions of all engines. The results for the Commuter Cycle are shown in Figure 5.4. The results are similar to those for the City Cycle and will therefore not be discussed separately.

5 Results of Powertrain Simulation

148

55 - Sedan - Battery - ED - 800 V - 2SP 52 - Sedan - Battery - ED - 800 V - 1SP 49 - Sedan - Diesel - S/P2 - 400 V - 4SP 46 - Sedan - Diesel - S/P2 - 400 V - 2SP 43 - Sedan - Gasoline - S/P2 - 400 V - 4SP 40 - Sedan - Gasoline - S/P2 - 400 V - 2SP 37 - Sedan - Fuel Cell - S - 400 V - 2SP 34 - Sedan - Fuel Cell - S - 400 V - 1SP 31 - Sedan - Gasoline - S - 400 V - 2SP 28 - Sedan - Gasoline - S - 400 V - 1SP 25 - Sedan - Diesel - P2 - 400 V - 9SP 22 - Sedan - Gasoline - P2 - 400 V - 9SP 19 - Sedan - Diesel - P2 - 400 V - 9SP 16 - Sedan - CNG - P2 - 400 V - 9SP 13 - Sedan - Gasoline - P2 - 400 V - 9SP 10 - Sedan - Diesel - P2 - 48 V - 9SP 07 - Sedan - CNG - P2 - 48 V - 9SP 04 - Sedan - Gasoline - P2 - 48 V - 9SP 03 - Sedan - Diesel - P0 - 48 V - 9SP 02 - Sedan - CNG - P0 - 48 V - 9SP 01 - Sedan - Gasoline - P0 - 48 V - 9SP 0 electric energy consumption [kWh / 100 km]

10

20

30

40

50

chemical energy consumption [kWh / 100 km]

55 - Sedan - Battery - ED - 800 V - 2SP 52 - Sedan - Battery - ED - 800 V - 1SP 49 - Sedan - Diesel - S/P2 - 400 V - 4SP 46 - Sedan - Diesel - S/P2 - 400 V - 2SP 43 - Sedan - Gasoline - S/P2 - 400 V - 4SP 40 - Sedan - Gasoline - S/P2 - 400 V - 2SP 37 - Sedan - Fuel Cell - S - 400 V - 2SP 34 - Sedan - Fuel Cell - S - 400 V - 1SP 31 - Sedan - Gasoline - S - 400 V - 2SP 28 - Sedan - Gasoline - S - 400 V - 1SP 25 - Sedan - Diesel - P2 - 400 V - 9SP 22 - Sedan - Gasoline - P2 - 400 V - 9SP 19 - Sedan - Diesel - P2 - 400 V - 9SP 16 - Sedan - CNG - P2 - 400 V - 9SP 13 - Sedan - Gasoline - P2 - 400 V - 9SP 10 - Sedan - Diesel - P2 - 48 V - 9SP 07 - Sedan - CNG - P2 - 48 V - 9SP 04 - Sedan - Gasoline - P2 - 48 V - 9SP 03 - Sedan - Diesel - P0 - 48 V - 9SP 02 - Sedan - CNG - P0 - 48 V - 9SP 01 - Sedan - Gasoline - P0 - 48 V - 9SP 0

20

40

60

80

100

120

CO2-emissions [g / km]

Figure 5.3:

Energy consumption (top) and CO2-emissions (bottom) for the sedan driving the City Cycle.

5.1 Sedan

149

55 - Sedan - Battery - ED - 800 V - 2SP 52 - Sedan - Battery - ED - 800 V - 1SP 49 - Sedan - Diesel - S/P2 - 400 V - 4SP 46 - Sedan - Diesel - S/P2 - 400 V - 2SP 43 - Sedan - Gasoline - S/P2 - 400 V - 4SP 40 - Sedan - Gasoline - S/P2 - 400 V - 2SP 37 - Sedan - Fuel Cell - S - 400 V - 2SP 34 - Sedan - Fuel Cell - S - 400 V - 1SP 31 - Sedan - Gasoline - S - 400 V - 2SP 28 - Sedan - Gasoline - S - 400 V - 1SP 25 - Sedan - Diesel - P2 - 400 V - 9SP 22 - Sedan - Gasoline - P2 - 400 V - 9SP 19 - Sedan - Diesel - P2 - 400 V - 9SP 16 - Sedan - CNG - P2 - 400 V - 9SP 13 - Sedan - Gasoline - P2 - 400 V - 9SP 10 - Sedan - Diesel - P2 - 48 V - 9SP 07 - Sedan - CNG - P2 - 48 V - 9SP 04 - Sedan - Gasoline - P2 - 48 V - 9SP 03 - Sedan - Diesel - P0 - 48 V - 9SP 02 - Sedan - CNG - P0 - 48 V - 9SP 01 - Sedan - Gasoline - P0 - 48 V - 9SP 0 electric energy consumption [kWh / 100 km]

10

20

30

40

50

chemical energy consumption [kWh / 100 km]

55 - Sedan - Battery - ED - 800 V - 2SP 52 - Sedan - Battery - ED - 800 V - 1SP 49 - Sedan - Diesel - S/P2 - 400 V - 4SP 46 - Sedan - Diesel - S/P2 - 400 V - 2SP 43 - Sedan - Gasoline - S/P2 - 400 V - 4SP 40 - Sedan - Gasoline - S/P2 - 400 V - 2SP 37 - Sedan - Fuel Cell - S - 400 V - 2SP 34 - Sedan - Fuel Cell - S - 400 V - 1SP 31 - Sedan - Gasoline - S - 400 V - 2SP 28 - Sedan - Gasoline - S - 400 V - 1SP 25 - Sedan - Diesel - P2 - 400 V - 9SP 22 - Sedan - Gasoline - P2 - 400 V - 9SP 19 - Sedan - Diesel - P2 - 400 V - 9SP 16 - Sedan - CNG - P2 - 400 V - 9SP 13 - Sedan - Gasoline - P2 - 400 V - 9SP 10 - Sedan - Diesel - P2 - 48 V - 9SP 07 - Sedan - CNG - P2 - 48 V - 9SP 04 - Sedan - Gasoline - P2 - 48 V - 9SP 03 - Sedan - Diesel - P0 - 48 V - 9SP 02 - Sedan - CNG - P0 - 48 V - 9SP 01 - Sedan - Gasoline - P0 - 48 V - 9SP 0

20

40

60

80

100

120

CO2-emissions [g / km]

Figure 5.4:

Energy consumption (top) and CO2-emissions (bottom) for the sedan driving the Commuter Cycle.

150

5 Results of Powertrain Simulation

A more general analysis of the powertrain configurations with a relative comparison for the different cases reveals the impact of the different measures for improving powertrain efficiency. Figure 5.5 shows the impact of the different internal combustion engine powertrain technologies. For the natural gas engines (CNG) and the diesel engines, the gasoline engines are used as a reference. For the hybrid powertrain configurations, the P2-HEV-48 V is chosen as a reference. The values given in the two diagrams are mean values over the different configurations. The fuel cell vehicle and the battery electric vehicle are not part of this comparison. On the top, the mean values of the measures taken for the Primary Driving Cycle can be seen. Of all internal combustion engines, the gasoline engines are the most efficient ones, referring to the overall energy consumption, followed by the diesel and the natural gas engines. Regarding the hybridization rate, the P2-PHEV-400 V is the most efficient variation, followed by the S/P2-PHEV-400 V and the P2-HEV-400 V. For the S-PHEV-400 V and the P0-HEV-48 V an increase in overall energy consumption in comparison to the P2-HEV-48 V can be observed. In the bottom diagram, the values of the measures taken for the City Cycle are shown. Of all internal combustion engines, the gasoline engines have the lowest overall energy consumption, followed by the natural gas and the diesel engines. All PHEV-powertrains show a strong improvement in overall energy consumption due to full electric driving in the City Cycle. The same comparative analysis is carried out for the local CO2-emissions of the different powertrain configurations, as shown in Figure 5.6. One illustrates the Primary Driving Cycle (top) and one the City Cycle (bottom). Of all internal combustion engines, the natural gas engines show the lowest local CO2emissions, followed by the gasoline and the diesel engines. With regard to the local CO2-emissions driving the Primary Driving Cycle, the P2-PHEV-400V shows the highest decrease in local CO2-emissions, followed by the S/P2PHEV-400 V and the P2-HEV-400 V. For the S-PHEV-400 V and the P0HEV-48 V an increase in local CO2-emission in comparison to the P2-HEV48 V can be seen. For the City Cycle all PHEVs are almost emission free, due to full electric driving in this cycle. Figure 5.7 shows the impact of the measures of improvement on the electric energy consumption, referring to the data of driving the City Cycle for all PHEV and BEV-configurations. All gearbox comparisons are based on the simulated variations with differing gearbox configurations. The hybrid powertrains are referenced to the battery electric vehicles with the corresponding

5.1 Sedan

151

Diesel 103%

S/P2 - PHEV - 400 V

CNG 114%

91%

S - PHEV - 400 V

114%

98%

P0 - HEV - 48 V 81%

93%

P2 - PHEV - 400 V

P2 - HEV - 400 V Diesel 114%

S/P2 - PHEV - 400 V

CNG 109% 29% 135% P0 - HEV - 48 V

37% S - PHEV - 400 V 38% 92% P2 - PHEV - 400 V

P2 - HEV - 400 V

Referenced to gasoline engines Referenced to P2 – HEV – 48 V

Figure 5.5:

Impact of different internal combustion engine powertrain architectures on the overall energy consumption of the sedan driving the Primary Driving Cycle (open) (top) and driving the City Cycle (bottom).

5 Results of Powertrain Simulation

152

Diesel S/P2 - PHEV - 400 V

S - PHEV - 400 V

117% CNG 90%

88%

113% P0 - HEV - 48 V

98% 79%

93%

P2 - PHEV - 400 V

P2 - HEV - 400 V Diesel 125%

S/P2 - PHEV - 400 V

CNG 84% 0% 0%

S - PHEV - 400 V

135% P0 - HEV - 48 V

1% 92%

P2 - PHEV - 400 V

P2 - HEV - 400 V

Referenced to gasoline engines Referenced to P2 – HEV – 48 V

Figure 5.6: Impact of different internal combustion engine powertrain architectures on the local CO2-emissions of the sedan driving the Primary Driving Cycle (open) (top) and driving the City Cycle (bottom). gearboxes. Again, the values given in these diagrams are mean values over the different configurations. It becomes evident that the change from a 1-speed to a 2-speed gearbox has only a small impact of -3 % on the electric energy consumption of the sedan but the maximum speed of the vehicle is increased from 160 km / h to 250 km / h. The change from a 2-speed to a 4-speed gearbox

5.2 Sport Utility Vehicle

153

results in a mean decrease of 14 % in electric energy consumption. Of all hybrid powertrains, the S/P2-PHEV-400 V shows the highest decrease in electric energy consumption compared to the battery electric vehicle. The S/P2PHEV-400 V is followed by the P2-PHEV–400 V and after that by the SPHEV–400 V. 1SP to 2SP gearbox

97% S/P2 - PHEV - 400 V

86%

80%

95% S - PHEV - 400 V

2SP to 4SP gearbox

86% P2 - PHEV - 400 V

Referenced to 1 speed gearbox Referenced to 2 speed gearbox Referenced to BEV – 800 V

Figure 5.7:

5.2

Impact of different measures and powertrain architectures on the electric energy consumption of the sedan driving the City Cycle.

Sport Utility Vehicle

In this section, the energy consumption and local CO2-emissions of the sport utility vehicle (SUV), with and without tarpaulin trailer, for the different powertrain architectures are discussed. The following driving cycles are considered: the open Primary Driving Cycle, the Primary Driving Cycle with 100

154

5 Results of Powertrain Simulation

km / h speed limitation, the City Cycle, the Commuter Cycle, the open Maximum Range Cycle and the Maximum Range Cycle with 100 km / h speed limitation. The cycles with 100 km / h speed limitation are used for the SUV towing the tarpaulin trailer. The results for the SUV driving the Primary Driving Cycle with no speed limitation are shown in Figure 5.8. Regarding the overall energy consumption, it can be observed that for the hybrid powertrains with no external recharge possibility (05 – 20) the energy consumption decreases with the degree of electrification, respectively with the degree of hybridization. For the hybrid powertrains with external recharge possibility (23 – 50) the S/P2-powertrain architecture (41 - 50) shows the lowest chemical energy consumption of all internal combustion engine powertrains due to their small internal combustion engines, resulting in high efficiency operating points. The P2-powertrain architectures (23, 26) show a higher chemical energy consumption than the S/P2-powertrains due to their bigger combustion engines, resulting in more inefficient operating points. The S-powertrains with internal combustion engine (29, 32) show the highest chemical energy consumption of all PHEV-architectures, due to the efficiency decrease caused by the applied generator unit and, by that, the resulting additional conversion step to electrical energy. In contrast, the Spowertrains with the fuel cell system (35, 38) show the lowest chemical energy consumption of all powertrains, due to the high peak efficiency of the fuel cell system of around 60 %. The battery electric powertrains (53, 56) show the lowest overall energy consumption due to the high efficiency of the battery and absence of chemical energy converters. Of all internal combustion engine concepts, the gasoline engines show the lowest chemical energy consumption, followed by the diesel and the natural gas engines (CNG). The higher energy consumption of the diesel engines is caused by the electrical heating system for the selective catalytic reduction catalyst and the low exhaust gas temperatures of these engines, resulting in a high use of the electrical catalyst heating system. Exceptions are the S/P2-powertrains, where the diesel engines have a lower chemical energy consumptions than the gasoline engines. Due to high load operating points when using the small diesel engines in the S/P2-powertrains, the effect of catalyst heating decreases, making the diesel engine concepts more efficient than the gasoline engine concepts. The natural gas engines in general show a lower efficiency due to their higher air-to-fuel ratio resulting in higher charge exchange losses for this engine concept.

5.2 Sport Utility Vehicle

155

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

10

20

30

40

50

60

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

20

40

60

80

100

120 140

160

CO2-emissions [g / km]

Figure 5.8:

Energy consumption (top) and CO2-emissions (bottom) for the SUV driving the Primary Driving Cycle (open).

156

5 Results of Powertrain Simulation

Regarding the CO2-emissions, the battery electric powertrains and the Spowertrains with fuel cell system are locally emission free. The CO2-emissions of the internal combustion engines show close correlation with the chemical energy consumption. The natural gas engine shows the lowest CO2-emissions of all engine concepts, due to the high hydrogen-to-carbon ratio of the fuel. The CO2-emissions of the natural gas engines are followed by the gasoline engines, due to their lower overall energy consumption and better hydrogento-carbon ratio compared to the diesel engines. The diesel engines show the highest CO2-emissions of all engines. The results for the Maximum Range Cycle with no speed limitation are shown in Figure 5.9. The hybrid powertrains with internal combustion engines and external recharge possibility (23 – 50) show almost the same overall energy consumption for all concepts. An explanation for this is that for a certain degree of electrification the internal combustion engines work with maximum efficiency and therefore the effect of electrification reaches its optimum. The diesel engines show a lower overall energy consumption than the gasoline engines. This gives evidence that for high load applications the diesel engines are still more efficient than the gasoline engines, due to higher load operating points of the diesel engines, resulting in lower heating powers for the electrically heated selective catalytic reduction catalyst. It is as well evident that for (29, 35, 41, 47) and (32, 38, 44, 50) the variations with the higher number of switchable gears show a higher overall energy consumption. This may be caused by a bad configuration of the powertrain components for this specific driving cycle. All other results are similar to those of the Primary Driving Cycle. The results for the SUV driving the City Cycle are shown in Figure 5.10. Regarding the overall energy consumption, it can be observed that for the hybrid powertrains with no external recharge possibility (05 – 20) the overall energy consumption decreases with the degree of electrification, respectively with the degree of hybridization. For the hybrid powertrains with external recharge possibility (23 – 50) all powertrains have a low overall energy consumption, due to full electric driving in the City Cycle. The S/P2-powertrain architectures (41 – 50) show the lowest overall energy consumption compared to the battery electric vehicles (53, 56), due to low powertrain weight, high gear switching capability, high electric driving power and the use of a permanent magnet synchronous motor. The electric energy consumption of the S/P2- powertrain architectures is followed by the S-powertrains (29 - 38), also with high electric

5.2 Sport Utility Vehicle

157

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

10

20

30

40

50

60

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V -… 41 - SUV - Gasoline - S/P2 - 400 V -… 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

20

40

60

80

100 120 140 160

CO2-emissions [g / km]

Figure 5.9:

Energy consumption (top) and CO2-emissions (bottom) for the SUV driving the Maximum Range Cycle (open).

5 Results of Powertrain Simulation

158

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

10

20

30

40

50

60

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

20

40

60

80

100

120

140

160

CO2-emissions [g / km]

Figure 5.10:

Energy consumption (top) and CO2-emissions (bottom) for the SUV driving the City Cycle.

5.2 Sport Utility Vehicle

159

driving power and the P2-architectures (23, 26), with lower electric driving power. The P2-architectures show a significant high electric energy consumption for the diesel powertrain (26), due to the heating strategy for the electrically heated selective catalytic reduction catalyst. It must be assumed that the engine is started, even when it does not deliver any torque. For correcting this value, a more complex heuristic operating strategy needs to be implemented, but this is not part of this study. In conclusion, the electric energy consumption of (26) is not used for further investigations. Besides (26), the battery electric vehicles show the highest electric energy consumption, due to the high powertrain weight of these concepts, caused by the battery. Of all internal combustion engine concepts, the gasoline engines show the lowest chemical energy consumption followed by the diesel and the natural gas engines (CNG). The higher overall energy consumption of the diesel engines is caused by the electrical heating system for the selective catalytic reduction catalyst and the low exhaust gas temperatures of these engines, resulting in a high use of the electrical catalyst heating system. The natural gas engines in general show a lower efficiency due to their higher air-to-fuel ratio, resulting in higher charge exchange losses for this engine concept. With regard to the CO2-emissions, it can be said that all hybrid powertrains with external recharge possibility as well as the battery electric powertrains are locally emission free. The CO2-emissions of the internal combustion engines show close correlation with the chemical energy consumption. The natural gas engines show the lowest CO2-emissions of all engine concepts due to the high hydrogen-to-carbon ratio of the fuel. The CO2-emissions of the natural gas engines are followed by the gasoline engines due to their lower chemical energy consumption and better hydrogen-to-carbon ratio compared to the diesel engines. The diesel engines show the highest CO2-emissions of all engines. The results for the Commuter Cycle are summarized in Figure 5.11. The results are similar to the results for the City Cycle and will therefore not be discussed separately. The results for the SUV towing the tarpaulin trailer and driving the Primary Driving Cycle with 100 km / h speed limitation are shown in Figure 5.12. Regarding the overall energy consumption, it can be observed that for the hybrid powertrains with no external recharge possibility (05 – 20) the overall energy consumption decreases slightly with the degree of electrification, respectively with the degree of hybridization. The overall energy consumption of these

5 Results of Powertrain Simulation

160

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

10

20

30

40

50

60

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

50

100

150

CO2-emissions [g / km]

Figure 5.11:

Energy consumption (top) and CO2-emissions (bottom) for the SUV driving the Commuter Cycle.

5.2 Sport Utility Vehicle

161

powertrains is lower than for the S/P2-powertrains (41 - 50) and the S-powertrain architectures with internal combustion engine (29, 32). An explanation for this is that the small engines in these powertrains have to work over maximum efficiency to serve the high power demands for the operation with the trailer. For all hybrid powertrains with external recharge possibility (23 – 50), the P2-powertrain architectures (23, 26) show the lowest chemical energy consumption of all internal combustion engine powertrains, due to their bigger internal combustion engine compared to the S- and S/P2-powertrain architectures. The S-powertrain architectures with internal combustion engine show a higher chemical energy consumption than the P2-powertrains. The S/P2powertrains show the highest chemical energy consumption of all PHEV-architectures. The S-powertrains with the fuel cell system (35, 38) show the lowest chemical energy consumption of all powertrains, due to the high peak efficiency of the fuel cell system of around 60 %. The battery electric powertrains (53, 56) show the lowest overall energy consumption, due to the high efficiency of the battery and the absence of chemical energy converters. For (29, 35, 41, 47) and (32, 38, 44, 50) the variations with the higher number of switchable gears show a higher overall energy consumption. This may be caused by a bad configuration of the powertrain components for this specific driving cycle. Regarding the different internal combustion engine concepts, the diesel engines show the lowest chemical energy consumption, followed by the gasoline engines and the natural gas engines (CNG). The natural gas engines in general show a lower efficiency due to their higher air-to-fuel ratio, resulting in higher charge exchange losses for these engine concepts. For the CO2-emissions, it can be observed that the battery electric powertrains and the S-powertrains with fuel cell system are locally emission free. The CO2emissions of the internal combustion engines show close correlation with the chemical energy consumption. The natural gas engines show the lowest CO2emissions of all engine concepts due to the high hydrogen-to-carbon ratio of the fuel. The CO2-emissions of the natural gas engines are followed by the gasoline engines for all P2-architectures, due to better hydrogen-to-carbon ratio compared to the diesel engines. Regarding the S/P2-powertrains, the CO2emissions of the diesel and the gasoline engines are similar. The results for the SUV towing the tarpaulin trailer and driving the Maximum Range Cycle with 100 km / h speed limitation are shown Figure 5.13. The results are similar to the results for the Primary Driving Cycle and will therefore not be discussed separately.

5 Results of Powertrain Simulation

162

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

50

100

150

200

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

100

200

300

400

500

CO2-emissions [g / km]

Figure 5.12:

Energy consumption (top) and CO2-emissions (bottom) for the SUV + trailer driving the Primary Driving Cycle (100 km / h).

5.2 Sport Utility Vehicle

163

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

50

100

150

200

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

56 - SUV - Battery - ED - 800 V - 2SP 53 - SUV - Battery - ED - 800 V - 1SP 50 - SUV - Diesel - S/P2 - 400 V - 4SP/2SP 47 - SUV - Diesel - S/P2 - 400 V - 2SP/2SP 44 - SUV - Gasoline - S/P2 - 400 V - 4SP/2SP 41 - SUV - Gasoline - S/P2 - 400 V - 2SP/2SP 38 - SUV - Fuel Cell - S - 400 V - 2SP 35 - SUV - Fuel Cell - S - 400 V - 1SP 32 - SUV - Gasoline - S - 400 V - 2SP 29 - SUV - Gasoline - S - 400 V - 1SP 26 - SUV - Diesel - P2 - 400 V - 9SP 23 - SUV - Gasoline - P2 - 400 V - 9SP 20 - SUV - Diesel - P2 - 400 V - 9SP 17 - SUV - CNG - P2 - 400 V - 9SP 14 - SUV - Gasoline - P2 - 400 V - 9SP 11 - SUV - Diesel - P2 - 48 V - 9SP 08 - SUV - CNG - P2 - 48 V - 9SP 05 - SUV - Gasoline - P2 - 48 V - 9SP 0

100

200

300

400

500

CO2-emissions [g / km]

Figure 5.13:

Energy consumption (top) and CO2-emissions (bottom) for the SUV + trailer driving the Max. Range Cycle (100 km / h).

164

5 Results of Powertrain Simulation

A more general analysis of the powertrain configurations with a relative comparison of the different cases evaluates the impact of the different measures for improving powertrain efficiency. In Figure 5.14 the impact of the different internal combustion engine powertrain technologies can be seen. For the natural gas engines (CNG) and the diesel engines, the gasoline engines are used as a reference. For the hybrid powertrain configurations, the P2-HEV-48 V is chosen as a reference. The values given in the three diagrams are mean values over the different configurations. The fuel cell vehicle and the battery electric vehicle are not included in this comparison. Top, the mean values of the measures taken for the SUV driving the Primary Driving Cycle can be seen. Of all internal combustion engines, the gasoline and the diesel engines are the most efficient ones, referring to the overall energy consumption, followed by the natural gas engines (CNG). Regarding the hybridization rate, the S/P2PHEV-400 V is the most efficient variation, due to the small internal combustion engine working in more efficient operating points than the variations with big internal combustion engines. The S/P2-PHEV-400 V is followed by the P2-PHEV-400 V, the P2-HEV-400 V and the S-PHEV-400 V. Mid, the values of the measures taken for the City Cycle are shown. Of all internal combustion engines, the gasoline engines have the lowest overall energy consumption, followed by the natural gas and the diesel engines. All PHEV-powertrains show a strong decrease in overall energy consumption due to full electric driving in the City Cycle. At the bottom, the values of the measures taken for SUV + trailer driving the Primary Driving Cycle with 100 km / h speed limitation are shown. It shows that the diesel engines turn out to be more efficient than the gasoline engines in this case. The natural gas engines still have a higher overall energy consumption than the gasoline engines. Regarding the hybrid powertrains, the P2-PHEV-400 V and the P2-HEV-400 V show a decrease in overall energy consumption. In contrast, the S -PHEV-400 V and the S/P2-PHEV-400 V show a slight increase, due to their small engines operating over maximum efficiency. The same comparable analysis is carried out for the local CO2-emissions of the different powertrain configurations, illustrated in three diagrams in Figure 5.15. One is for the Primary Driving Cycle (top), one for the City Cycle (mid) and one for the Primary Driving Cycle with 100 km / h speed-limitation representing the SUV towing the tarpaulin trailer (bottom). Of all internal combustion engines, the natural gas engines show the lowest local CO2-emissions, followed by the gasoline and the diesel engines. Regarding the local CO2-

5.2 Sport Utility Vehicle

165

Diesel 100%

S/P2 - PHEV 400 V

111% CNG 66%

79%

85% S - PHEV - 400 V

P2 - HEV - 400 V

77%

P2 - PHEV - 400 V Diesel 124% S/P2 - PHEV 400 V

111% CNG 17% 27%

S - PHEV - 400 V

31%

69% P2 - HEV - 400 V

P2 - PHEV - 400 V Diesel

S/P2 - PHEV 400 V 103%

92% 109%

103% S - PHEV - 400 V

CNG

97% P2 - HEV - 400 V 91% P2 - PHEV - 400 V

Referenced to gasoline engines Referenced to P2 – HEV – 48 V

Figure 5.14:

Impact of different internal combustion engine powertrain architectures on the overall energy consumption of the SUV driving the Primary Driving Cycle (open) (top), the City Cycle (mid) and SUV + trailer driving the Primary Driving Cycle (100 km / h) (bottom).

5 Results of Powertrain Simulation

166

Diesel 114% S/P2 - PHEV 400 V

CNG 85%

66%

79%

85% S - PHEV - 400 V

P2 - HEV - 400 V

77%

P2 - PHEV - 400 V Diesel 142% S/P2 - PHEV 400 V

CNG 86% 0% 0% 0%

69%

S - PHEV - 400 V

P2 - HEV - 400 V

P2 - PHEV - 400 V Diesel

S/P2 - PHEV 400 V 107%

S - PHEV - 400 V

106% CNG 84%

97%

103%

P2 - HEV - 400 V 90%

P2 - PHEV - 400 V

Referenced to gasoline engines Referenced to P2 – HEV – 48 V

Figure 5.15:

Impact of different internal combustion engine powertrain architectures on the local CO2-emissions of the SUV driving the Primary Driving Cycle (open) (top), the City Cycle (mid) and SUV + trailer driving the Primary Driving Cycle (100 km / h) (bottom).

5.2 Sport Utility Vehicle

167

emissions, when driving the Primary Driving Cycle (open), the S/P2-PHEV400V shows the highest decrease in local CO2-emissions, followed by the P2PHEV-400 V, the P2-HEV-400 V and the S-PHEV-400 V. Regarding the City Cycle, all PHEVs are almost emission free, due to full electric driving in this cycle. For the Primary Driving Cycle (100 km / h), the P2-PHEV-400 V shows the biggest decrease in local CO2-emissions, followed the P2-HEV-400 V. The S/P2-PHEV-400V and the S-PHEV-400 V show an increase in local CO2eimissions for driving the SUV + trailer. Regarding local CO2-emissions, the S/P2-PHEV-400V shows slightly worse results than the S-PHEV-400 V, even if the overall energy consumption is almost the same. This may be caused by the increase of the mean SOC-level in the operating strategy in order to avoid battery failure due to deep discharge. Figure 5.16 shows the impact of the measures of improvement for the electric energy consumption, referring to the data for the City Cycle for all PHEV and BEV-configurations. 1SP to 2SP gearbox

94% S/P2 - PHEV - 400 V

2SP to 4SP gearbox 82%

57%

81% S - PHEV - 400 V

88% P2 - PHEV - 400 V

Referenced to 1 speed gearbox Referenced to 2 speed gearbox Referenced to BEV – 800 V

Figure 5.16:

Impact of different measures and powertrain architectures on the electric energy consumption of the SUV driving the City Cycle

5 Results of Powertrain Simulation

168

All gearbox variations are based on the simulated variations with differing gearbox configurations. The hybrid powertrains are referenced to the battery electric vehicles with the corresponding gearboxes. It can be observed that the change from a 1-speed to a 2-speed gearbox results in a mean decrease of 6 % in the electric energy consumption for the SUV. In addition, the maximum speed of the vehicle is increased from 160 km / h to 250 km / h. The change from a 2-speed to a 4-speed gearbox results in a mean decrease of 18 % in electric energy consumption. Of all hybrid powertrains, the S/P2-PHEV-400 V shows the highest decrease in electric energy consumption. The S/P2PHEV-400 V is followed by the S-PHEV–400 V and finally by the P2-PHEV– 400 V.

5.3

Light-Duty Vehicle

In this section, the energy consumption and local CO2-emissions of the lightduty vehicle (LDV) for the different powertrain architectures are discussed. The driving cycles are the Primary Driving Cycle with 80 km / h speed limitation, the City Cycle, the Urban Delivery Cycle and the Maximum Range Cycle with 80 km / h speed limitation. The results for the LDV driving the Primary Driving Cycle with 80 km / h speed limitation are shown in Figure 5.17. Regarding the overall energy consumption, the hybrid powertrains with no external recharge possibility (06 – 21) the energy consumption decreases with the degree of electrification, respectively with the degree of hybridization. Of all hybrid powertrains with external recharge possibility (24 – 51), the P2-powertrain architectures (24, 27) show the lowest chemical energy consumption of all internal combustion engine powertrains. The S/P2-powertrain architectures (42 – 51) show a higher chemical energy consumption than the P2-powertrains. The S-powertrains with internal combustion engine (30, 33) show a higher chemical energy consumption than the P2- and S/P2-powertrain architectures due to the efficiency decrease caused by the used generator unit and by that, the additional resulting conversion step to electrical energy. In contrast, the S-powertrains with the fuel cell system (36, 39) show the lowest chemical energy consumption of all powertrains due to the high peak efficiency of the fuel cell system of around 60 %. The battery electric powertrains (54, 57) show the lowest

5.3 Light-Duty Vehicle

169

57 - LDV - Battery - ED - 800 V - 2SP 54 - LDV - Battery - ED - 800 V - 1SP 51 - LDV - Diesel - S/P2 - 400 V - 4SP 48 - LDV - Diesel - S/P2 - 400 V - 2SP 45 - LDV - Gasoline - S/P2 - 400 V - 4SP 42 - LDV - Gasoline - S/P2 - 400 V - 2SP 39 - LDV - Fuel Cell - S - 400 V - 2SP 36 - LDV - Fuel Cell - S - 400 V - 1SP 33 - LDV - Gasoline - S - 400 V - 2SP 30 - LDV - Gasoline - S - 400 V - 1SP 27 - LDV - Diesel - P2 - 400 V - 6SP 24 - LDV - Gasoline - P2 - 400 V - 6SP 21 - LDV - Diesel - P2 - 400 V - 6SP 18 - LDV - CNG - P2 - 400 V - 6SP 15 - LDV - Gasoline - P2 - 400 V - 6SP 12 - LDV - Diesel - P2 - 48 V - 6SP 09 - LDV - CNG - P2 - 48 V - 6SP 06 - LDV - Gasoline - P2 - 48 V - 6SP 0

20

40

60

80

100

120

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

57 - LDV - Battery - ED - 800 V - 2SP 54 - LDV - Battery - ED - 800 V - 1SP 51 - LDV - Diesel - S/P2 - 400 V - 4SP 48 - LDV - Diesel - S/P2 - 400 V - 2SP 45 - LDV - Gasoline - S/P2 - 400 V - 4SP 42 - LDV - Gasoline - S/P2 - 400 V - 2SP 39 - LDV - Fuel Cell - S - 400 V - 2SP 36 - LDV - Fuel Cell - S - 400 V - 1SP 33 - LDV - Gasoline - S - 400 V - 2SP 30 - LDV - Gasoline - S - 400 V - 1SP 27 - LDV - Diesel - P2 - 400 V - 6SP 24 - LDV - Gasoline - P2 - 400 V - 6SP 21 - LDV - Diesel - P2 - 400 V - 6SP 18 - LDV - CNG - P2 - 400 V - 6SP 15 - LDV - Gasoline - P2 - 400 V - 6SP 12 - LDV - Diesel - P2 - 48 V - 6SP 09 - LDV - CNG - P2 - 48 V - 6SP 06 - LDV - Gasoline - P2 - 48 V - 6SP 0

50

100

150

200

250

CO2-emissions [g / km]

Figure 5.17:

Energy consumption (top) and CO2-emissions (bottom) for the LDV driving the Primary Driving Cycle (80 km / h).

170

5 Results of Powertrain Simulation

overall energy consumption due to the high efficiency of the battery and absence of chemical energy converters. Of all internal combustion engine concepts, the diesel engines show the lowest chemical energy consumption, followed by the gasoline and the natural gas engines (CNG). The natural gas engines in general show a lower efficiency, due to its higher air-to-fuel ratio, resulting in higher charge exchange losses for these engine concepts. Regarding the CO2-emissions, the battery electric powertrains and the Spowertrains with fuel cell system are locally emission free. The CO2-emissions of the internal combustion engines show close correlation with the chemical energy consumption. The natural gas engines show the lowest CO2-emissions of all engine concepts, due to the high hydrogen-to-carbon ratio of the fuel. The CO2-emissions of the natural engines are followed by the gasoline engines, due to their better hydrogen-to-carbon ratio compared to the diesel engines. The diesel engines show the highest CO2-emissions of all engines. The results for the Maximum Range Cycle with 80 km / h speed limitation are shown in Figure 5.18. The results are similar to the results for the Primary Driving Cycle and will therefore not be discussed separately. For both driving cycles it can be said that for (36, 42) and (39, 45) the variations with the higher number of switchable gears show a higher overall energy consumption. This may be caused by a bad configuration of the powertrain components for this specific driving cycle. All other results are similar to the Primary Driving Cycle. The results for the LDV driving the City Cycle are shown in Figure 5.19. Regarding the overall energy consumption, it can be observed that for the hybrid powertrains with no external recharge possibility (06 – 21), the overall energy consumption decreases with the degree of electrification, respectively with the degree of hybridization. For the hybrid powertrains with external recharge possibility (24 – 51) all powertrains have a low overall energy consumption due to almost full electric driving in the City Cycle. The S/P2-powertrain architectures (41 – 51) show the lowest overall energy consumption of all vehicles compared to the battery electric vehicles (52, 55). The electric energy consumption of the S/P2-powertrain architectures is followed by the S-powertrains (30 - 39) and the P2-architectures (24, 27). The P2-architectures show a significant high electric energy consumption for the diesel powertrain (27), due to the heating strategy for the electrically heated selective catalytic reduction catalyst. Besides (27), the battery electric vehicles show the highest electric energy consumption due to the high powertrain weight of these concepts

5.3 Light-Duty Vehicle

171

57 - LDV - Battery - ED - 800 V - 2SP 54 - LDV - Battery - ED - 800 V - 1SP 51 - LDV - Diesel - S/P2 - 400 V - 4SP 48 - LDV - Diesel - S/P2 - 400 V - 2SP 45 - LDV - Gasoline - S/P2 - 400 V - 4SP 42 - LDV - Gasoline - S/P2 - 400 V - 2SP 39 - LDV - Fuel Cell - S - 400 V - 2SP 36 - LDV - Fuel Cell - S - 400 V - 1SP 33 - LDV - Gasoline - S - 400 V - 2SP 30 - LDV - Gasoline - S - 400 V - 1SP 27 - LDV - Diesel - P2 - 400 V - 6SP 24 - LDV - Gasoline - P2 - 400 V - 6SP 21 - LDV - Diesel - P2 - 400 V - 6SP 18 - LDV - CNG - P2 - 400 V - 6SP 15 - LDV - Gasoline - P2 - 400 V - 6SP 12 - LDV - Diesel - P2 - 48 V - 6SP 09 - LDV - CNG - P2 - 48 V - 6SP 06 - LDV - Gasoline - P2 - 48 V - 6SP 0

20

40

60

80

100

120

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

57 - LDV - Battery - ED - 800 V - 2SP 54 - LDV - Battery - ED - 800 V - 1SP 51 - LDV - Diesel - S/P2 - 400 V - 4SP 48 - LDV - Diesel - S/P2 - 400 V - 2SP 45 - LDV - Gasoline - S/P2 - 400 V - 4SP 42 - LDV - Gasoline - S/P2 - 400 V - 2SP 39 - LDV - Fuel Cell - S - 400 V - 2SP 36 - LDV - Fuel Cell - S - 400 V - 1SP 33 - LDV - Gasoline - S - 400 V - 2SP 30 - LDV - Gasoline - S - 400 V - 1SP 27 - LDV - Diesel - P2 - 400 V - 6SP 24 - LDV - Gasoline - P2 - 400 V - 6SP 21 - LDV - Diesel - P2 - 400 V - 6SP 18 - LDV - CNG - P2 - 400 V - 6SP 15 - LDV - Gasoline - P2 - 400 V - 6SP 12 - LDV - Diesel - P2 - 48 V - 6SP 09 - LDV - CNG - P2 - 48 V - 6SP 06 - LDV - Gasoline - P2 - 48 V - 6SP 0

50

100

150

200

250

CO2-emissions [g / km]

Figure 5.18:

Energy consumption (top) and CO2-emissions (bottom) for the LDV driving the Maximum Range Cycle (80 km / h).

5 Results of Powertrain Simulation

172

57 - LDV - Battery - ED - 800 V - 2SP 54 - LDV - Battery - ED - 800 V - 1SP 51 - LDV - Diesel - S/P2 - 400 V - 4SP 48 - LDV - Diesel - S/P2 - 400 V - 2SP 45 - LDV - Gasoline - S/P2 - 400 V - 4SP 42 - LDV - Gasoline - S/P2 - 400 V - 2SP 39 - LDV - Fuel Cell - S - 400 V - 2SP 36 - LDV - Fuel Cell - S - 400 V - 1SP 33 - LDV - Gasoline - S - 400 V - 2SP 30 - LDV - Gasoline - S - 400 V - 1SP 27 - LDV - Diesel - P2 - 400 V - 6SP 24 - LDV - Gasoline - P2 - 400 V - 6SP 21 - LDV - Diesel - P2 - 400 V - 6SP 18 - LDV - CNG - P2 - 400 V - 6SP 15 - LDV - Gasoline - P2 - 400 V - 6SP 12 - LDV - Diesel - P2 - 48 V - 6SP 09 - LDV - CNG - P2 - 48 V - 6SP 06 - LDV - Gasoline - P2 - 48 V - 6SP 0

20

40

60

80

100

120

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

57 - LDV - Battery - ED - 800 V - 2SP 54 - LDV - Battery - ED - 800 V - 1SP 51 - LDV - Diesel - S/P2 - 400 V - 4SP 48 - LDV - Diesel - S/P2 - 400 V - 2SP 45 - LDV - Gasoline - S/P2 - 400 V - 4SP 42 - LDV - Gasoline - S/P2 - 400 V - 2SP 39 - LDV - Fuel Cell - S - 400 V - 2SP 36 - LDV - Fuel Cell - S - 400 V - 1SP 33 - LDV - Gasoline - S - 400 V - 2SP 30 - LDV - Gasoline - S - 400 V - 1SP 27 - LDV - Diesel - P2 - 400 V - 6SP 24 - LDV - Gasoline - P2 - 400 V - 6SP 21 - LDV - Diesel - P2 - 400 V - 6SP 18 - LDV - CNG - P2 - 400 V - 6SP 15 - LDV - Gasoline - P2 - 400 V - 6SP 12 - LDV - Diesel - P2 - 48 V - 6SP 09 - LDV - CNG - P2 - 48 V - 6SP 06 - LDV - Gasoline - P2 - 48 V - 6SP 0

50

100

150

200

250

CO2-emissions [g / km]

Figure 5.19:

Energy consumption (top) and CO2-emissions (bottom) for the LDV driving the City Cycle.

5.3 Light-Duty Vehicle

173

caused by the battery. Of all internal combustion engine concepts, the diesel engines show the lowest chemical energy consumption followed by the gasoline and the natural gas engines (CNG). The natural gas engines in general show a lower efficiency due to its higher air-to-fuel ratio, resulting in higher charge exchange losses for this engine concept. Regarding the CO2-emissions, all hybrid powertrains with external recharge possibility as well as the battery electric powertrains are almost locally emission free. The CO2-emissions of the internal combustion engines show close correlation with the chemical energy consumption. The natural gas engines show the lowest CO2-emissions of all engine concepts, due to the high hydrogen-to-carbon ratio of the fuel. The CO2-emissions of the natural engines are followed by the gasoline engines, due to its lower chemical energy consumption and better hydrogen-to-carbon ratio compared to the diesel engines. The diesel engines show the highest CO2emissions of all engines. The results for the Urban Delivery Cycle are shown in Figure 5.20. The results for the hybrid powertrains with no external recharge possibility (06 – 21) are similar to those of the City Cycle. Considering the hybrid powertrains with external recharge possibility (24 – 51), the S-powertrains (30 - 39) are driving the Urban Delivery Cycle with almost no use of chemical energy, due to their higher full electric range compared to the P2- and S/P2-powertrains (24, 27, 42 – 51). The S/P2-powertrains are more efficient than the P2-powertrains with regard to overall and chemical energy consumption. All other results are similar to the City Cycle. A more general analysis of the powertrain configurations with a relative comparison of the different cases evaluates the impact of the different measures for improving powertrain efficiency. Figure 5.21 shows the impact of the different internal combustion engine powertrain technologies. For the natural gas engines (CNG) and the diesel engines, the gasoline engines are used as a reference. For the hybrid powertrain configurations, the P2-HEV-48 V is chosen as a reference. The values given in these diagrams are mean values over the different configurations. The fuel cell vehicle and the battery electric vehicle are not included in this comparison. On the top, the mean values of the measures taken for the LDV driving the Primary Driving Cycle with 80 km / h speed limitation can be seen. Of all internal combustion engines, the diesel engines are the most efficient ones referring to the overall energy consumption, followed by the gasoline and the natural gas engines (CNG). Regarding

5 Results of Powertrain Simulation

174

57 - LDV - Battery - ED - 800 V - 2SP 54 - LDV - Battery - ED - 800 V - 1SP 51 - LDV - Diesel - S/P2 - 400 V - 4SP 48 - LDV - Diesel - S/P2 - 400 V - 2SP 45 - LDV - Gasoline - S/P2 - 400 V - 4SP 42 - LDV - Gasoline - S/P2 - 400 V - 2SP 39 - LDV - Fuel Cell - S - 400 V - 2SP 36 - LDV - Fuel Cell - S - 400 V - 1SP 33 - LDV - Gasoline - S - 400 V - 2SP 30 - LDV - Gasoline - S - 400 V - 1SP 27 - LDV - Diesel - P2 - 400 V - 6SP 24 - LDV - Gasoline - P2 - 400 V - 6SP 21 - LDV - Diesel - P2 - 400 V - 6SP 18 - LDV - CNG - P2 - 400 V - 6SP 15 - LDV - Gasoline - P2 - 400 V - 6SP 12 - LDV - Diesel - P2 - 48 V - 6SP 09 - LDV - CNG - P2 - 48 V - 6SP 06 - LDV - Gasoline - P2 - 48 V - 6SP 0

20

40

60

80

100

120

electric energy consumption [kWh / 100 km] chemical energy consumption [kWh / 100 km]

57 - LDV - Battery - ED - 800 V - 2SP 54 - LDV - Battery - ED - 800 V - 1SP 51 - LDV - Diesel - S/P2 - 400 V - 4SP 48 - LDV - Diesel - S/P2 - 400 V - 2SP 45 - LDV - Gasoline - S/P2 - 400 V - 4SP 42 - LDV - Gasoline - S/P2 - 400 V - 2SP 39 - LDV - Fuel Cell - S - 400 V - 2SP 36 - LDV - Fuel Cell - S - 400 V - 1SP 33 - LDV - Gasoline - S - 400 V - 2SP 30 - LDV - Gasoline - S - 400 V - 1SP 27 - LDV - Diesel - P2 - 400 V - 6SP 24 - LDV - Gasoline - P2 - 400 V - 6SP 21 - LDV - Diesel - P2 - 400 V - 6SP 18 - LDV - CNG - P2 - 400 V - 6SP 15 - LDV - Gasoline - P2 - 400 V - 6SP 12 - LDV - Diesel - P2 - 48 V - 6SP 09 - LDV - CNG - P2 - 48 V - 6SP 06 - LDV - Gasoline - P2 - 48 V - 6SP 0

50

100

150

200

250

CO2-emissions [g / km]

Figure 5.20:

Energy consumption (top) and CO2-emissions (bottom) for the LDV driving the Urban Delivery Cycle.

5.3 Light-Duty Vehicle

175

Diesel

98% S/P2 - PHEV - 400 V

CNG 107% 76%

89%

82% S - PHEV - 400 V

P2 - HEV - 400 V

70%

P2 - PHEV - 400 V Diesel

102% S/P2 - PHEV - 400 V

117%

CNG

25% 28% 32%

79%

S - PHEV - 400 V

P2 - HEV - 400 V

P2 - PHEV - 400 V Referenced to gasoline engines Referenced to P2 – HEV – 48 V

Figure 5.21:

Impact of different internal combustion engine powertrain architectures on the overall energy consumption of the LDV driving the Primary Driving Cycle (80 km / h) (top) and the City Cycle (bottom).

176

5 Results of Powertrain Simulation

the hybridization rate, the P2-PHEV-400 V is the most efficient variation, followed by the S/P2-PHEV-400 V, the S-HEV-400 V and the P2-HEV-400 V. In the bottom diagram, the values of the measures taken for the City Cycle are shown. Of all internal combustion engines, the gasoline engines have the lowest overall energy consumption, followed by the diesel and the natural gas engines. All PHEV-powertrains show a high decrease in overall energy consumption, due to full electric driving in the City Cycle. The same comparative analysis is executed for the local CO2-emissions of the different powertrain configurations. The results are summarized in the two diagrams in Figure 5.22. One diagram is for the Primary Driving Cycle with 80 km /h speed limitation (top) and one for the City Cycle (bottom). Of all internal combustion engines, the natural gas engines show the lowest local CO2-emissions followed by the gasoline and the diesel engines. Regarding the local CO2-emissions, driving the Primary Driving Cycle (80 km / h), the P2-PHEV400V shows the biggest decrease in local CO2-emissions, followed by the S/P2-PHEV-400 V, the S-PHEV-400 V and the P2-HEV-400 V. For the City Cycle all PHEVs are almost emission free, due to almost full electric driving in this cycle. Figure 5.23 illustrates the impact of the measures of improvement for the electric energy consumption, referring to the data for the City Cycle for all PHEV and BEV-configurations. All gearbox variations are based on the simulated variations with differing gearbox configurations. The hybrid powertrains are referenced to the battery electric vehicles with the corresponding gearboxes. It is evident that the change from a 1-speed to a 2-speed gearbox has an increasing impact on the electric energy consumption that may be caused by bad configuration of the powertrain components for this specific driving cycle. The change from a 2-speed to a 4-speed gearbox results in a mean decrease of 5 % in electric energy consumption. Of all hybrid powertrains, the S/P2-PHEV400 V shows the highest decrease in electric energy consumption. The SPHEV–400 V and the P2-PHEV–400 V show almost no decrease in electric energy consumption compared to the BEV. This is due to the high overall weight of the vehicle, which means that the weight of the battery has less impact on the overall energy consumption.

5.3 Light-Duty Vehicle

177

Diesel 112% S/P2 - PHEV - 400 V

CNG 83%

76%

89%

82% S - PHEV - 400 V

P2 - HEV - 400 V

70%

P2 - PHEV - 400 V Diesel

102% CNG

S/P2 - PHEV - 400 V 90% 0% 0% 2%

78%

S - PHEV - 400 V

P2 - HEV - 400 V

P2 - PHEV - 400 V Referenced to gasoline engines Referenced to P2 – HEV – 48 V

Figure 5.22:

Impact of different internal combustion engine powertrain architectures on the local CO2-emissions of the LDV driving the Primary Driving Cycle (80 km / h) (top) and the City Cycle (bottom).

5 Results of Powertrain Simulation

178

1SP to 2SP gearbox

102% S/P2 - PHEV - 400 V

2SP to 4SP gearbox

81%

95%

96% S - PHEV - 400 V

99% P2 - PHEV - 400 V

Referenced to 1 speed gearbox Referenced to 2 speed gearbox Referenced to BEV – 800 V

Figure 5.23:

5.4

Impact of different measures and powertrain architectures on the electric energy consumption of the LDV driving the City Cycle.

Conclusion

In this final section, a brief comparative overview of the different powertrain architectures, particularly with regard to their efficiency on one hand, and their pollutant emissions on the other hand, is presented. First, the effect of different internal combustion engine technologies on the overall energy consumption and the local CO2-emissions is discussed and evaluated. After that, the impact of the varying forms of hybridization as well as the use of fuel cell and battery electric vehicles is considered for the same requirements. Finally, the advantages and disadvantages of the different gearbox types are explained. Gasoline engines: For the gasoline engines, the results show that for low load driving, such as driving the City or the Commuter Cycle, the overall energy consumption is lowest for all internal combustion engine concepts. If the load

5.4 Conclusion

179

increases for higher load driving cycles, such as the Primary Driving Cycle and the Maximum Range Cycle, the gasoline engines show almost the same efficiency than the diesel engines. For high load applications, such as towing the 2.0 t tarpaulin trailer, the gasoline engines turn out to be more inefficient compared to the diesel engine. Regarding the local CO2-emissions, the gasoline engine reaches always lower emissions than the diesel engine, but higher emission than the natural gas engine (CNG). Diesel engines: The results show that for low load driving, such as the City or the Commuter Cycle, the diesel engines are the internal combustion engines with the highest overall energy consumption of all internal combustion engine concepts. This is caused by the high electric power used to heat up the electrically heated selective catalytic reduction catalyst to keep nitrogen oxide emissions on low levels. High heating power is needed because of the low exhaust gas temperatures for driving the diesel engine in low load operating points. For medium load applications such as the Primary Driving Cycle and the Maximum Range cycle the gap in overall energy consumption between the diesel and the gasoline engine shrinks due to lower electric catalyst heating power and higher load operation. For high load application such as towing the 2.0 t tarpaulin trailer the diesel engines turns out to be the most efficient engines in terms of overall energy consumption. Regarding the local CO2-emissions, the diesel engines always have the highest emissions of all internal combustion engine concepts, due to lower hydrogen-to-carbon ratio of the fuel compared to the gasoline and the natural gas engines. Natural gas engines: For the natural gas engines, the results show that for low load applications, such as driving the City or Commuter Cycle, the natural gas engines (CNG) are more efficient than the diesel engines but less efficient than the gasoline engines in terms of overall energy consumption. For medium load applications, such as the Primary Driving Cycle and the Maximum Range Cycle as well as high load applications, such as towing the 2.0 t tarpaulin trailer, the natural gas engines are the least efficient engines in terms of overall energy consumption. This is due to the higher air-to-fuel ratio for combusting methane, resulting in higher charge exchange losses compared to the other internal combustion engine concepts. Regarding the local CO2-emissions, the natural gas engines always show the lowest emissions of all engine concepts. However, it must be considered that the real impact on GHG emissions by the natural gas engines may be higher than calculated, due to methane leakage

180

5 Results of Powertrain Simulation

during operation of the natural gas engines. This might be caused by incomplete combustion of methane and blow-by leakage during combustion. P0/P2–HEV: For the P0 and P2-powertrain architecture, the overall energy consumption decreases for higher hybridization rates and these concepts become more efficient in terms of overall energy consumption. With high hybridization rates and medium and high load applications, such as driving the Primary Driving Cycle or towing the 2.0 t tarpaulin trailer, these concepts are more efficient than the serial-powertrain concepts or even the serial/parallelpowertrain concepts. One reason may be a higher number of switchable gears in combination with more optimal operation of the internal combustion engines. Second, compared to the serial-powertrains, the absence of an extra energy conversion step for the generator unit may also have a positive effect. Finally, for high load operation the bigger combustion engines of the P0/P2HEV-concepts, in contrast to the serial and serial/parallel powertrains, can still be operated with good efficiency on high loads. The smaller internal combustion engines in the serial and serial/parallel powertrains must work over their optimal operating points to meet the high power demands when driving in high load applications. Full electric driving is not possible, due to the missing external recharge possibility. P2–PHEV: The P2-powertrain architectures are almost the most efficient ones in terms of overall energy consumption for medium and high load operation, such as driving the Primary Driving Cycle or towing the 2.0 t tarpaulin trailer. One exception is the SUV for medium load applications, such as driving the Primary Driving Cycle or the Maximum Range Cycle. Here, the serial/parallel architecture turns out to be more efficient. An explanation is the smaller internal combustion engines in the serial/parallel concepts, compared to the P2PHEV-concepts caused by the high overall system power of the SUV. This results in more optimal operating points of the smaller engines. For low load applications, such as driving the City or Commuter Cycle, these powertrain concepts drive fully electric. S–PHEV: The S-powertrain architectures show a good efficiency in manner of overall energy consumption. For high loads such as towing the 2.0 t tarpaulin trailer the overall energy consumption increases due to high load operating points of the internal combustion engines. For low load applications, such as driving the City or Commuter Cycle these powertrain concepts drive fully electric.

5.4 Conclusion

181

S/P2–PHEV: The S/P2-powertrain architectures show a good efficiency in terms of overall energy consumption. For the SUV driving medium load applications, such as driving the Primary Driving Cycle or the Maximum Range Cycle, this powertrain architecture is the most efficient one. This is due to the small internal combustion engines in relation to the overall system power, resulting in optimal operating points of these small engines. For high loads, such as towing the 2.0 t tarpaulin trailer, the overall energy consumption increases due to high load operating points, resulting in operation over optimal efficiency of the internal combustion engines. For low load applications, such as driving the City or Commuter Cycle, these powertrain concepts drive fully electric. Regarding fully electric driving, these concepts always show the lowest electric energy consumption of all powertrain concepts. FC–PHEV: The fuel cell electric vehicles with external recharge possibility show the second lowest overall energy consumption of all powertrains for medium and high load applications, such as driving the Primary Driving Cycle or towing the 2.0 t tarpaulin trailer. This is caused by the good efficiency of the used fuel cell system. The concept always drives locally CO2-emission free. For low load applications, such as driving the City or Commuter Cycle, these powertrain concepts drive fully electric. BEV: The battery electric vehicles show the lowest overall energy consumption of all powertrains for medium and high load applications, such as driving the Primary Driving Cycle or towing the 2.0 t tarpaulin trailer. This is caused by the high efficiency of the battery storage and absence of secondary chemical energy converters. The concept always drives locally CO2-emission free. For low load applications, such as driving the City or Commuter Cycle, these powertrain concepts are the most inefficient ones, due to their high weight caused by the high capacity battery. 1SP to 2SP gearbox: Switching from a 1-speed to a 2-speed gearbox not necessarily leads to a lower fuel consumption. It also depends on how the electric drive systems work together with the gear ratio at a certain speed in the driving cycle. An advantage of the 2-speed gearbox is that for the sedan and the SUV the maximum speed is increased from 160 km / h to 250 km / h. Reaching 250 km / h with the 1-speed gearbox would result in bad low speed torque and certain drive cycles could not be driven.

182

5 Results of Powertrain Simulation

2SP to 4SP gearbox: Switching from a 2-speed to a 4-speed gearbox not necessarily leads to a lower fuel consumption for medium and high load applications, such as driving the Primary Driving Cycle or towing the 2.0 t tarpaulin trailer. It also depends on how the drive systems, such as electric drives and the internal combustion engine, work together with the gear ratio at a certain speed in the driving cycle. Regarding low load driving respectively fully electric driving, the 4-speed gearbox always shows a lower electric energy consumption than the 2-speed gearbox.

6 Summary and Conclusion In this research study, 57 different vehicles are simulated and evaluated with regard to their energy consumption and local CO2-emissions, based on three vehicle types, respectively a sedan, a sport utility vehicle and a converted lightduty vehicle with 7.5 t of maximum gross weight. Each vehicle type is simulated with different powertrain architectures, which will be suitable for the market in 2040. For each powertrain architecture, the technical specifications for the used components are estimated and extrapolated to the year 2040. In total, there are five combustion engine concepts, two electric drive concept, three battery concepts, one fuel cell system concept and nine gearbox variations. These components are afterwards combined to fit the different powertrain architectures. Each vehicle is tested under various conditions to give an overview of the different use cases of the vehicle. The simulation applies an optimized version of the Equivalent Consumption Minimization Strategy to predict the optimal minimized energy consumption for each vehicle. All simulation results are presented in detailed data sheets, which are explained in Appendix A1 and are available for download. For the driving cycles, the data is shown accordingly in Appendix A2 and is also available as a download. For the internal combustion engines, the results show that the gasoline engines are the most efficient engines for low and medium load application, such as normal or urban driving. Solely for high load applications, such as the sport utility vehicle (SUV) towing a trailer or the light-duty vehicle (LDV), the diesel engines turn out to be more efficient. Two effects are responsible for the high overall energy consumption of the diesel engines, compared to the gasoline engines. The first effect is that the gasoline engines have a good efficiency at low torques because intensive measures are taken for dethrotteling the gas exchange process. The second effect is that the diesel engines need a high amount of electric energy to pre-heat the electrically heated selective catalytic reaction catalyst, due to high exhaust gas mass flows and low exhaust gas temperatures of this engine concept. The natural gas engines are always the worst concept in terms of overall energy consumption. Regarding the local CO2-emission, the natural gas engines are always the most efficient ones, followed by the gasoline engines and after that by the diesel engines. The natural gas engines may worsen in terms of local GHG emissions if methane leakage © The Author(s), under exclusive license to Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2023 T. Stoll, A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040, Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart, https://doi.org/10.1007/978-3-658-42168-7_6

184

6 Summary and Conclusion

due to incomplete combustion and blow-by is taken into account. For the hybrid electric vehicles without external recharge possibility it shows that in general the fuel consumption decreases with an increasing degree of electrification regarding the electric driving power. In high load applications, these concepts may become more efficient than concepts with smaller internal combustion engines, due to the possibility of maintaining the operation in optimal efficiency. The hybrid electric vehicles with external recharge possibility can drive fully electric for low load application, such as urban driving, which results in zero local CO2-emissions. For fully electric driving, the S/P2-plug-in hybrid electric vehicles are the most efficient concept, due to high electric drive powers compared to the other hybrids and low overall weight compared to the battery electric vehicles. For hybrid driving the P2-plug-in hybrid electric vehicles are the most efficient ones, unless the engine is oversized for the current application; then the S/P2-plug-in hybrids become more efficient, due to their smaller engines and the resulting operation in more efficient operating points. The fuel cell electric vehicles show the lowest chemical energy consumption of all hybrid electric vehicles, due to the high efficiency of the fuel cell system. This results in the lowest chemical energy consumption for medium and high load applications. For low load applications, this powertrain architecture can drive fully electric. The electric energy consumption for low load driving is similar to that of the S-hybrid electric vehicles with external recharge possibility, due to the similarity of the powertrain architecture. The fuel cell electric vehicle always drives with zero local CO2-emissions. The battery electric vehicles show the lowest overall energy consumption for medium and high load driving due to the high efficiency of the battery. For low load driving, the hybrid electric vehicles with external recharge possibility turn out to be more efficient than the battery electric vehicles in terms of overall energy consumption. This is due to the high extra weight of the big batteries used in this powertrain architectures. The battery electric vehicles always drive with zero local CO2-emissions. This study compares different powertrain technologies with extrapolated components for the year 2040. Not only battery and fuel cell electric vehicles are examined, but also different hybrid electric vehicles with different combustion engine technologies. Fuel cell and battery electric vehicles are the only way to reduce fleet CO2-emissions towards zero under current legislation and without the market introduction of renewable fuels (e-fuels). Hybrid electric vehicles with combustion engine also offer a great opportunity due to their good local

6 Summary and Conclusion

185

energy efficiency, low production emissions, low overall weight and, last but not least, low production costs. Furthermore, the introduction of e-fuels can lead to a rapid decarbonisation of the existing fleet. However, the use of efuels requires the introduction of a legal mechanism to compensate for local CO2-emissions. To gain more detail knowledge further investigations into specific powertrain variations would be desirable. Further optimizations of gear ratios and more specific operating strategies may lead to a more optimized operation of the powertrain and may result in a further but slight decrease in energy consumption respectively in a decrease in local CO2-emissions. Another promising investigation could be the simulation of heavy-duty vehicles between 7.5 t and 40.0 t of gross weight, to investigate the impact of the introduced powertrain technologies on additional use cases. This might help to evaluate and react to forthcoming legal provisions and regulations concerning pollutant reduction for heavy-duty vehicles.

Bibliography [1]

United Nations, "Paris Agreement," United Nations, Paris, 2015.

[2]

European Comission, "REGULATION OF THE EU PAR. AND OF THE COUNCIL amending Regulation (EU) 2019/631 as regards strengthening the CO2 emission performance standards for new passenger cars and new light commercial vehicles in line with the Union’s increased climate ambition," European Comission, Brussels, 2022.

[3]

European Parliament, "2019/631: setting CO2 emission performance standards for new passenger cars and for new light commercial vehicles," European Parliament, Brussels, 2019.

[4]

European Commission, "Road transport: Reducing CO₂ emissions from vehicles," European Commission, https://ec.europa.eu/clima/euaction/transport-emissions/road-transport-reducing-co2-emissionsvehicles_en, 2021.

[5]

eurostat, Average CO2 emissions per km from new passenger cars, http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=sdg_12_30& lang=en: eurostat, 2021.

[6]

European Parliament, "Sustainable maritime fuels: 'Fit for 55' package: The FuelEU Maritime proposal"," European Parliament, Brussels, 2022.

[7]

European Parliament, "ReFuelEU Aviation initiative: Sustainable aviation fuels and the fit for 55 package," European Parliament, Burssels, 2022.

[8]

A. Fandakov, A Phenomenological Knock Model for the Development of Future Engine Concepts, Wiesbaden: Springer Vieweg, 2019.

[9]

L. Urban, Modellierung der klopfenden Verbrennung methanbasierter Kraftstoffe, Wiesbaden: Springer Vieweg, 2021.

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2023 T. Stoll, A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040, Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart, https://doi.org/10.1007/978-3-658-42168-7

188

Bibliography

[10] A. Vacca, Potential of Water Injection for Gasoline Engines by Means of a 3D-CFD Virtual Test Bench, Wiesbaden: Springer Vieweg, 2021. [11] W. Attard and P. Parsons, A Normally Aspirated Spark Initiated Combustion System Capable of High Load, High Efficiency and Near Zero NOx Emissions in a Modern Vehicle Powertrain, Warrendale: SAE International, 2010. [12] P. Richardson, J. Gu, G. D. Liberto, K. Kimoto, K. Aoyagi and H. Shibata, Knock Prevention by Retarded Injection with Ultra-High Pressure and Fuel Injector Nozzle Development, Aachen: 28th Aachen Colloquium Automobile and Engine Technology, 2019. [13] A. Kuznik, F. Negüs, V. Kelich, M. Thielen, E. Wenz and C. Wouters, "ICE2025+: Ultimate System Efficiency," Forschungvereinigung Verbrennungskraftmaschinen, Frankfurt am Main, 2020. [14] M. Huß, Übertragung von Motoreigenschaften mit Hilfe charakteristischer Skalierfunktionen zur Simulatiion verschiedener Varianten von Ottomotoren, München: Technischen Universität München, 2013. [15] A. Kawaguchi, H. Iguma, H. Yamashita and T. N. e. al, "ThermoSwing Wall Insulation Technology; - A Novel Heat Loss Reduction Approach on Engine Combustion Chamber," SAE, SAE 2016 International Powertrains, Fuels & Lubricants Meeting, 2016. [16] A. Balazs, Optimierte Auslegung von ottomotorischen Hybridantriebssträngen unter realen Fahrbedingungen, Aachen: Rheinisch-Westfälischen Technischen Hochschule Aachen, 2015. [17] A. Roberts, R. Brooks and P. Shipway, Internal combustion engine cold-start efficiency: A review of the problem, causes and potential solutions, Amsterdam: ELSEVIER, 2014. [18] S. Fuchs, Verfahren zur parameterbasierten Gewichtsabschätzung neuer Fahrzeugkonzepte, München: Technische Universität München, 2014.

Bibliography

189

[19] G. Hepke, Direkte Nutzung von Abgasenthalpie zur Effzienzsteigerung von Kraftfahrzeugen, München: Technischen Universität München, 2010. [20] "Der Leichtbaukatalysator ist flexibler, widerstandsfähiger und leichter," Continental Emitec GmbH, [Online]. Available: https:// www.emitec.com/technik/katalysatortraeger/leichtbaukatalysator/. [Accessed 15 01 2021]. [21] M. Presti and L. Pace, Optimisation Development of Advanced Exhaust Gas Aftertreatment Systems for Automotive Applications, Lohmar: Emitec GmbH, 2005. [22] D. Schröder, Elektrische Antriebe - Grundlagen, Berlin Heidelberg: Springer-Verlag, 1994, 2000, 2007, 2009. [23] G. Bertotti, "General properties of power losses in soft ferromagnetic materials," IEEE, IEEE Access, 1988. [24] D. Reitz, W. Ruder, M. Dilzer and C. Blaes, "P2 HV-Antriebe: Die effiziente Hybridisierung für alle Getriebe," Schaeffler Kolloquium, Bühl, 2018. [25] 2019-02 DIN VDE 0530-30-2 (VDE0530-30-2):, "Rotating electrical machines – Part 30-2: Efficiency classes of variable speed AC motors (IE-code) (IEC/TS 60034-30-2:2016)," Beuth-Verlag, Berlin, 2019. [26] T. Eckenfels, F. Kolb, S. Lehmann, W. Neugebauer and M. Calero, "48-Volt-Hybridisierung: Eine intelligente Aufwertung des Antriebsstrangs," 11. Schaeffler Kolloquium, Baden-Baden, 2018. [27] IEEE, "Air Gap Power in Induction Motor," IEEE, www.eee guide.com, 2016. [28] N. Beeli and F. Jenni, "Effizienter Betrieb von Asynchronmotoren," Bundesamt für Energie Schweizerische Eidgenossenschaft, Bern, 2010.

190

Bibliography

[29] N. Rivière, G. Volpe, M. Villani, G. Fabri, L. D. Leonardo and M. Popescu, "Design Analysis of a High Speed Copper Rotor Induction Motor for a Traction Application," IEEE, 2019 IEEE International Electric Machines & Drives Conference (IEMDC), 2019. [30] T. Pfund, "Die Schaeffler eDrive Plattform," Schaeffler Kolloquium, Bühl, 2018. [31] P. Seipel, "automotive.at," Österreichischer Wirtschaftsverlag GmbH, [Online]. Available: https://www.automotive.at/automotive/elektro achse-der-neue-antriebsstrang-44630. [Accessed 08 06 2022]. [32] U. Baumann, "auto-motor-und-sport.de," Motor Presse Stuttgart, 16 08 2018. [Online]. Available: ttps://www.auto-motor-und-sport.de/news/ e-achse-bosch-elektrisches-antriebsmodul/. [Accessed 08 06 2022]. [33] DIN EN 61800-9-2 (VDE 0160-109-2), 2018-01:, Drehzahlveränderbare elektrische Antriebe: Teil 9-2: Ökodesign für Antriebssysteme, Motorstarter, Leistungselektronik und deren angetriebene Einrichtungen – Indikatoren für die Energieeffizienz von Antriebssystemen und Motorstartern, Berlin: Beuth-Verlag, 2018. [34] F. Jeschull and S. Trabesinger, "Fast Charge Limitations for Graphite Anodes with Si as Capacity Enhancing Additive.," 2020. [35] B. Gangaja, S. Nair and D. Santhanagopalan, "Surface Engineered Li4Ti5O12 Nanostructures for High Power Li Ion Batteries.," Ion Batteries. Nano Micro Lett. 12, 30 (2020)., 2020. [36] M. Rashid, A. Sahoo, Y. Sharma and A. Gupta, Electrochemical Assessment of Li4Ti5O12 Li-Ion Battery Anodes, ECS Meeting Abstracts, Volume MA2017-01, A05-Lithium-Ion Batteries and Beyond: ECS - The Electrochemical Society, 2017. [37] A. Yamada, H. Koizumi, S.-I. Nishimura, N. Sonoyama, R. Kanno, M. Yonemura, T. Nakamura and K. Yo, Room-temperaturemiscibility gap in LixFePO4, nature materials | Vol. 5: nature materials, 2006.

Bibliography

191

[38] A. Mauger and C. Julien, "Olivine Positive Electrodes for Li Ion Batteries: Status and Perspectives.," Batteries 4(3):39, 2018. [39] H. Shen, W. Xiang and X. Shi, "Hierarchical LiMn0.5Fe0.5PO4/C nanorods with excellent electrochemical performance synthesized by rheological phase method as cathode for lithium ion battery.," Ionics 22, 193 200, 2016. [40] H. J. Noh, "Comparison of the structural and electrochemical properties of layered Li[ NixCoyMnz ]O2 cathode material for lithium ion batteries," Journal of Power Sources 233, 2013. [41] SDI Samsung, Introduction of Samsung SDI’s 94Ah cells, unpublished, 2015. [42] H. Buqa, D. Goers, M. Holzapfel and M. Spahr, "Rate Capability of Graphite Negative Electrodes for Lithium Ion Batteries.," Journal of The Electrochemical Society., 2005. [43] D. D. Lecce, R. Verrelli and J. Hassoun, "Lithium ion batteries for sustainable energy storage recent advances towards new cell configurations," Green Chem., 2017,19, 3442 3467, 2017. [44] A. Thielmann and C. Neef, "Energiespeicher Roadmap (Update 2017) Hochenergie Batterien 2030+ und Perspektiven zukünftiger Batterietechnologien," Fraunhofer Institut ISI, Karlsruhe, 2017. [45] H. Entenmann and M. Reichenbacher, Modeling and Control of a Fuel Cell System for Vehicle Applications, Stuttgart: unpublished, 2020. [46] N. S. Kamgang Nzengang, Development and Modeling of a Dynamic Cell Model for a PEM Fuel Cell, Stuttgart: unpublished, 2020. [47] "ENERGY.GOV," U.S. Department of Energy (DOE), 30 09 2021. [Online]. Available: https://www.energy.gov/eere/fuelcells/doe-tech nical-targets-polymer-electrolyte-membrane-fuel-cell-components. [Accessed 30 09 2021].

192

Bibliography

[48] M. K. Debe, "Effect of Electrode Surface Area Distribution on High Current," Journal of The Electrochemical Society, J. Electrochem. Soc. 159 B53, 2012. [49] Hexagon Purus ASA, "Hydrogen high-pressure Type 4 cylinders," Hexagon Purus ASA, [Online]. Available: https://hex-agonpurus.com/ our-solutions/hydrogen-systems/hydrogen-type-4-cylinders. [Accessed 03 02 2022]. [50] H. Ding, C. Cai, Z. Chen, T. Ke and B. Mao, Configuration Synthesis and Performance Analysis of 9-Speed Automatic Transmissions, Chin. J. Mech. Eng. 33, 50 (2020): Chinese Journal of Mechanical Engineering, 2020. [51] H. Scherer, ZF 6-Speed Automatic Transmission for Passenger Cars, SAE Transactions , 2003, Vol. 112, Section 6: JOURNAL OF PASSENGER CAR: SAE International, 2003. [52] L. A. M. van Dongen, Efficiency Characteristics of Manual and Automatic Passenger Car Transaxles, SAE Transactions , 1982, Vol. 91, Section 3: 820689–821157 (1982), pp. 2498-2509: SAE International, 1982. [53] ZF Friedrichshafen AG, "Efficient and dynamic: Transmission technology from ZF," ZF Friedrichshafen AG, Friedrichshafen, 2017. [54] ZF Friedrichshafen AG, "ZF Friedrichshafen AG," ZF Friedrichshafen AG, [Online]. Available: https://www.zf.com/products/de/cars/ products_29289.html. [Accessed 2021]. [55] M. J. d. S. G. P. Correia, Power Loss Modelling of a Rear Axle Transmission with Experimental Study of No-load Losses, Porto: Faculdade de Engenharia da Universidade do Porto, 2017. [56] T. Biermann, Light, Compact and Efficient: Schaeffler differential systems set the pace, Schaeffler Symposium: Schaeffler Technologies AG & Co. KG, 2018.

Bibliography

193

[57] J. Ando, H. Ando, T. Tsuda, K. Suzuki and Y. Niikawa, Development of Third-Generation Electronically Controlles AWD Coupling with New High-Performance Electromagnetic Clutch, Jtekt Engineering Journal English Edition No. 1011E: JTEKT, 2014. [58] M. Wan, "4WD with Electronic (Active) torque split," AutoZine Technical School, http://autozine.org, 2016. [59] A. Windt and O. Arnhold, "Szenarien für den Markthochlauf, Nationale Leitstelle Ladeinfrastruktur," Nationale Leitstelle Ladeinfrastruktur, Berlin, 2020. [60] D. Berjoza, "Aerodynamic Characteristics of Cars with Trailers in Road Experiments," Engineering for Rural Development, Jelgava, 2015. [61] M. Brotz and M. Maul, "Variable Ventiltriebe für die Regeneration von NOx-Speicherkatalysatoren an Dieselmotoren," Institut für Fahrzeugtechnik Stuttgart, Stuttgart, 2021. [62] European Comission, "COMMISSION REGULATION (EU) 2016/ 646 of 20 April 2016 amending Regulation (EC) No 692/2008 as regards emissions from light passenger and commercial vehicles (Euro 6)," European Comission, Brussels, 2016. [63] European Comission, "COMMISSION REGULATION (EU) 2017/ 1154 of 7 June 2017 amending Regulation (EU) 2017/1151 supplementing Regulation (EC) No 715/2007 of the European Parliament and of the Council on type-approval of motor vehicles with respect to emissions from light ...," European Comission, Brussels, 2017. [64] European Comission, "ERORDNUNG (EU) 2016/427 DER KOMMISSION vom 10. März 2016 zur Änderung der Verordnung (EG) Nr. 692/2008 hinsichtlich der Emissionen von leichten Personenkraftwagen und Nutzfahrzeugen (Euro 6)," European Comission, Brussels, 2016.

194

Bibliography

[65] S. Onori, L. Serrao and G. Rizzoni, Hybrid Electric Vehicles: Energy Management Strategies, London: Springer, 2016. [66] G. Paganelli, Conception et commande d’une chaîne de traction pour véhicule hybride parallèle thermique et électrique, Valenciennes: Université de Valenciennes, 1999. [67] R. Bellman, Dynamic Programming, Princeton : Princeton University Press, 1957. [68] Natural Resources Canada, "Learn the facts: Fuel consumption and CO2," Natural Resources Canada, Canada, 2014. [69] J. Adolf and S. Rommerskirchen, "Shell PKW-Szenarien bis 2040," Shell Deutschland, 2014. [70] J. Brokate, E. D. Özdemir and U. Kugler, "Der Pkw-Markt bis 2040: Was das Auto von morgen antreibt," Deutsches Zentrum für Luft-und Raumfahrt e.V., 2013. [71] P. Summerton, P. Mock and P. Harrison, "Klimafreundliche Autos in Deutschland: Ein Überblick der sozioökonomischen," European Climate Foundation, 2017.

Appendix

A1. Simulation Data Sheets For each of the 57 simulated vehicles in this book, a data sheet with information about the powertrain configuration and the results is provided. The data sheets can be downloaded at the SpringerLink-Homepage of this book. The evaluation of the simulations in this section is exemplarily shown for the “01 - Sedan - Gasoline - P0 - 48 V - 9SP” powertrain configuration. In the data sheets, first, the powertrain configuration is illustrated. The used illustrations for the different components can be seen in Table A.1. Bellow the powertrain configuration illustration the operating points for the different powertrain components driving the Primary Driving Cycle in charge-sustaining mode can be seen. Following the illustrations first the data for the vehicle configuration is given. This data is used to parameterize the longitudinal dynamics simulation. Second, the powertrain data is displayed with the corresponding sizing for the different powertrain components. Third the simulation results for the fuel consumption simulating the different drive cycles is summarized. The CO2-emissions can be calculated using the emission factors in Table 5.1.

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2023 T. Stoll, A Simulative Approach to Predict Energy Consumption of Future Powertrain Configurations for the Year 2040, Wissenschaftliche Reihe Fahrzeugtechnik Universität Stuttgart, https://doi.org/10.1007/978-3-658-42168-7

Appendix

196 Table A.1:

Internal Engine

Illustration of powertrain components in simulation data sheets. Combustion

Electric Motor (permanent magnet / induction)

Fuel Cell System

Power Electronics Converter Torque Converter / Dual Clutch / Clutch / MultiPlate-Clutch

Gearbox

CCS3 - Charger

Tank / Battery

A1 Simulation Data Sheets

01 - Sedan - Gasoline - P0 - 48 V - 9SP

197

Appendix

198 Vehicle Data Vehicle Weight + Payload Drive Type Drag Coefficient cw Rolling Resistance Coefficient fR Powertrain Data Internal Combustion Engine - Type Electric Drive Front - Type Gearbox - Type System Power Electric Power Internal Combustion Engine Power Battery Size Tank Size Powertrain Weight (with full tank) Average gear switches per minute Simulation Results – Fuel Consumption Fuel Consumption Primary Driving Cycle Fuel Consumption Maximum Range Cycle Fuel Consumption City Cycle Fuel Consumption Commuter Cycle

1651 0.255 0.0062

kg 2WD -

High Efficiency Concept Permanent Magnet 9 – speed AT – w TQ 112 kW 12 kW 100 kW 0.5 kWh 35 l 275 kg 6.0 1 / min 4.23 4.27 3.84 3.70

l / 100 km l / 100 km l / 100 km l / 100 km

A2 Drive Cycle Data Sheets

199

A2. Drive Cycle Data Sheets For each of the 9 drive cycles used for the simulations in this book, a data sheet with information about its RDE-conformity is provided. The data sheets can be downloaded at the SpringerLink-Homepage of this book. The evaluation of the dive cycle is explained in this section, referring to the Primary Driving Cycle. In the first diagram the runtime of the corresponding drive cycle over the vehicle speed (black) and the current altitude level (grey) can be seen. In the diagrams below the (v*apos)95 and the RPA value over the average vehicle velocity for each route type within the drive cycle, respectively urban, rural and motorway are plotted. Both values are a measure for the dynamics of the drive cycle. While the grey curves represent the values for the current drive cycle, the black curves show the maximum permitted limits of RDE. For the table values, first the analysis of the RDE-speed conformity is given. The first values are the vehicle speed resolution, which has to be 1 Hz, and the minimum acceleration resulting form the vehicle speed measurement. These two parameters define that the drive cycle data has the correct resolution and sufficient accuracy. Following, the (v*apos)95 and the RPA values are given, as already explained before. The next values check the conformity of urban driving, especially the shares of pauses for valid stop-and-go traffic conformity. The trip shares for urban, rural and motorway are evaluated next. The values tend to achieve a trip distance composition for the RDE cycle of 1 to 1 to 1. Deviations form this values are allowed. For motorway driving, the maximum vehicle velocity is limited to 160 km / h. In addition, the share of overall driving time over 145 km / h has to be below 3 %, while the share over 100 km / h has to be over 5 %. The overall trip duration has to be between 90 min and 120 min, while each share of the cycle (urban, rural, motorway) has to be at least 16 km in distance. The RDE speed conformity is followed by the RDE altitude conformity. First the maximum altitude difference between start and end of the cycle has to be below 100 m. The maximum positive altitude difference per 100 km is limited for the whole driving cycle as well as for the urban trip shares. The cycle must comply with the maximum altitude boundary. It has to either comply for moderate altitude conditions up to 700 m or with extended altitude conditions up to 1200 m. Further specifications for the ambient conditions or respectively the allowed ranges for the ambient condition can be found in [62] [63] and [64].

Appendix

200

Primary Driving Cycle 600

200

speed [km / h]

altitude [m]

150

400 300

100

200

altitude [m]

speed [km / h]

500

50 100 0

0 0

2000

4000

6000

time [s]

RDE speed conformity Speed resolution Sampling rate [Hz]

1

=

1

valid

A2 Drive Cycle Data Sheets 2016/646_ Appendix_7a_3. 1.1

Min. acceleration a > 0 [m / s²]

201

0.0003

≤

0.01

valid

[95]-percentile – (v*apos)95 2016/646_ Appendix_7a_4. 1.1.1

Urban [W / kg]

17.22

≤

18.19

valid

Rural [W / kg]

24.08

≤

24.53

valid

Motorway [W / kg]

20.82

≤

27.53

valid

RPA – Relative Positive Acceleration 2016/646_ Appendix_7a_4. 1.2

Urban [m / s²]

0.2466

≥

0.1313

valid

Rural [m / s²]

0.1382

≥

0.0556

valid

Motorway [m / s²]

0.0857

≥

0.025

valid

Mean velocity [km / h]

27.61

‫א‬

15