Impact and Utilization of Emerging PHEV in Smart Power Systems
F. M. Rabiul Islam
A thesis submitted in fulfilment of the requirements of the degree of
Doctor of Philosophy
SCIENTIA
MANU E T MENTE
School of Engineering and Information Technology
The University of New South Wales
Canberra, Australia
May 2013
Abstract
Stability and quality are two important issues for the operation of power systems as an unsecured system faces a lot of unusual operating conditions due to which it can undergo blackouts and incur huge losses. In addition, the demand for electrical energy continues to grow steadily. Electric grid upgrades, and especially the con- struction of new transmission lines and loads, cannot keep pace with the demand for increasing power plant capacity and energy for various reasons. The changing characteristics and new loads of power systems continue to provide new challenges for system designers and operators. A plug-in hybrid electric vehicle (PHEV) is such a load, which can have huge impact on, and provide many opportunities for, power systems. In the planning and design of a distribution system, PHEVs are one of the most important factors as they can be a spinning reserve of energy, as well as a major load for distribution networks.
The first contribution of this work is to identify the charging effect of dynamic
PHEVs in a power system for which a single-machine infinite-bus (SMIB) system, solar power system and a distribution system are used. A dynamic load model of PHEVs based on a third-order battery model is introduced. To determine the adequacy of a system, it is necessary to conduct a micro-level analysis to determine the impact of the PHEV loads on the grid. The scope of such an analysis covers the performances of wind and solar generation with dynamic PHEV loads, and
ii Abstract iii small signal stability analysis of the power grid demonstrates that it is important to consider the dynamics of PHEV loads.
A second and unique contribution of this thesis is the design of virtual active
filter for power systems using V2G technology. Nonlinear loads, inverters and con- verters used for rectifying or inverting operations absorb reactive power from the connected bus, the compensation of which is essential in power system. Moreover, the nonlinear behavior of power electronic devices produce harmonics in power net- works which are filtered by passive and active filters. Due to the increasing interest in power electronics-based nonlinear loads, the reactive power compensation and harmonic reduction method should be improved but doing so would increase the total cost as well as hardware and control complexity. In this thesis, the V2G tech- nology is used to design a virtual active filter based on the instantaneous power theory (p-q theory). The potential of a low-cost solution to the power quality prob- lem that utilizes the reactive power and filtering capabilities of PHEVs parked in charging stations is investigated. Simulations are performed for various power sys- tem networks to demonstrate that the proposed virtual active filter improves power quality while meeting IEEE standards.
The other significant contribution of this research is the design of FACTS devices using V2G technology, which can fulfill multiple power flow control objectives, such as the needs of reactive shunt compensation, phase shifting and series compensation.
However, as current FACTS devices are quite expensive, they are not widely used. Abstract iv
Therefore, in this dissertation, the potential of PHEVs in a V2G mode of operation, which provides a low-cost solution for designs of virtual FACTS devices (UPFC,
UPQC, DVR) using a PHEV charging station, is explained. Simulations undertaken demonstrate that PHEVs have the potential to work as virtual FACTS devices to improve power quality.
Finally, a benchmark distribution network and microgrid are used to verify the performances of the proposed filter and FACTS devices. Acknowledgements
Foremost, I would like to express my sincere gratitude and appreciation to my su- pervisor, Associate Professor Hemanshu Roy Pota, for his support, patience, and encouragement throughout my PhD study. His technical and editorial advice has been essential for the completion of this dissertation. He is not only my supervi- sor but has also encouraged and challenged me throughout my academic research.
Despite his busy schedule, he always manages to squeeze in a meeting when there is a need. Thanks to his remarkable engineering intuition, open-mindedness, and endless enthusiasm, I have always left those meetings in a happy and illuminated mood.
My sincere appreciation goes to fellow group members for making my time en- joyable in Canberra. I also thank people who were not part of my group but helped me out.
Furthermore, I would like to thank all members of the institute for contributing to such an inspiring and pleasant atmosphere. I would like to thank Pam Giannakakis,
Joan Woodward, Elizabeth Carey, Denise Russell for the general administrative support they provided to me. Software support for computing issues from Jon
Lowrey, Mrs Eri Rigg, Michael Lanza and Mike Wilson was invaluable. I also thank
John Davis and the building officer, Ty Everett, who provided me with day-to-day infrastructure support.
v Acknowledgements vi
My appreciation goes to fellow postgraduate students for making my journey a memorable experience and sharing friendship during the past few years. My sincere thanks go to those who, either nearby or at a distance, were concerned about my studies and me. I also wish to extend my warmest thanks to my friends in the University and community who have made my life in the Australian Capital
Territory fruitful and enjoyable. Moreover, I am grateful for the enormous support and understanding of my loving wife, Farjana Afrin, my sons, Ahmed Abdullah
Ayan and Anuvob Abdullah Shayan.
I would like to thank very much my mother, brother as well as sister for their never-ending support and inspiration during my long stay in Australian Capital
Territory.
Last, but certainly not least, I am indebted to the School of Engineering and
Information Technology at The University of New South Wales and the Australian
Research Council for their financial support and the opportunity to embark on this
PhD journey. Dedicated to
My parents, my siblings
My wife, Farjana Afrin
And
My sons, Ahmed Abdullah Ayan and Anuvob Abdullah Shayan List of Publications
Refereed Journal Papers
1. F. R. Islam and H. R. Pota, “Impact of Dynamic PHEV Load on Photo-
voltaic System,” International Journal of Electrical and Computer Engineering
(IJECE), vol. 2, no. 5, pp. 644–654, October 2012.
2. F. R. Islam and H. R. Pota, “PHEVs Park as Virtual UPFC” TELKOMNIKA
Indonesian Journal of Electrical Engineering, vol. 10 no.8, pp. 1701–1708,
December 2012.
3. F. R. Islam and H. R. Pota, “Plug in Hybrid Electric Vehicles Park as Virtual
DVR” IET Electronics Letters , vol. 49 no.3, January 2013.
4. F. R. Islam and H. R. Pota, “V2G Technology to Design Smart Active Filter
for Solar Power System” International Journal of Power Electronics and Drive
System (IJPEDS), vol. 3 no.1, March 2013.
5. F. R. Islam and H. R. Pota, “Virtual Active Filter for HVDC Networks using
V2G Technology” International Journal of Electrical Power and Energy Sys-
tems (IJEPES), Under review, Submitted on July 2012, Manuscript Number:
IJEPES-D-12-00650
Refereed Conference Papers
6. F.R. Islam, H.R. Pota, M.A. Mahmud and M.J. Hossain, “Impact of PHEV
viii ix
Loads on the Dynamic Performance of Power System,” 20th Australasian Uni-
versities Power Engineering Conference (AUPEC), New Zealand , pp.1-5, 5-8
Dec. 2010
7. F.R. Islam, H.R. Pota and M.S. Ali, “V2G Technology for Designing Active
Filter System to Improve Wind Power Quality,” 21st Australasian Universities
Power Engineering Conference (AUPEC), Brisbane, Australia , pp.1-6, 25-28
Sept. 2011
8. F. R. Islam and H. R. Pota, “V2G Technology to Improve Power Quality”
European Electric Vehicle Congress, Brussels, Belgium. 25-18 Oct. 2011
9. F. R. Islam and H. R. Pota, “Smart Operation of Microgrid with PHEV.”
European Electric Vehicle Congress, Brussels, Belgium. 25-28 Oct. 2011
10. F. R. Islam and H. R. Pota, “Smart Microgrid with Renewable Energy and
PHEV.” International Conference on Energy and Meteorology Weather and
Climate for the Energy Industry, Gold coast, Australia 8-11 Nov. 2011
11. F.R. Islam and H.R. Pota, “Design a PV-AF System using V2G Technology
to Improve Power Quality,” IECON 2011 - 37th Annual Conference on IEEE
Industrial Electronics Society, 2011 21st Australasian , pp.861-866, 7-10 Nov.
2011
12. F.R. Islam and H.R. Pota, “Impact of Dynamic PHEVs Load on Renewable
Sources Based Distribution System,” IECON 2011 - 37th Annual Conference
on IEEE Industrial Electronics Society, 2011 21st Australasian , pp.4698-4703, x
7-10 Nov. 2011
13. F.R. Islam and H.R. Pota, “V2G Technology to Improve Wind Power Quality
and Stability,” Australian Control Conference (AUCC) 2011, pp.452-457, 10-
11 Nov. 2011
14. F.R. Islam, H.R. Pota and A.B.M. Nasiruzzaman, “PHEV’s Park as a Virtual
Active Filter for HVDC Networks,” 11th International Conference on Envi-
ronment and Electrical Engineering (EEEIC), 2012, pp.885-890, 18-25 May
2012
15. F.R. Islam, H.R. Pota and M.S. Ali, “V2G Technology to Design a Virtual
UPFC,” 11th International Conference on Environment and Electrical Engi-
neering (EEEIC), 2012, pp.568-573, 18-25 May 2012
16. F.R. Islam, H.R. Pota and A.B.M. Nasiruzzaman, “Design a Unified Power
Quality Conditioner using V2G Technology,” IEEE International Power Engi-
neering and Optimization Conference (PEOCO), 2012, pp.521-526, 6-7 June
2012
17. F.R. Islam, H.R. Pota M.S. Rahman and M.S. Ali, “Performance Analysis of
Photovoltaic Cell with Dynamic PHEV Loads,” 22nd Australasian Universities
Power Engineering Conference (AUPEC), Bali, Indonesia , pp.1-6, 25-28 Sept.
2011 Contents
Declaration i
Abstract ii
Acknowledgements v
List of Publications viii
List of Symbols xxvi
Chapter 1 Introduction 1
1.1 Background ...... 2
1.1.1 Smart grid and distribution network ...... 5
1.1.2 Overview of integration of renewable energy into grid . . . . . 6
1.1.3 Putting two and two together ...... 8
1.1.4 Requirements to make the dream a reality ...... 10
1.2 Motivation for the Current Research ...... 10
1.3 Contributions of this Research ...... 12
1.4 Thesis Outline ...... 15
Chapter 2 V2G Technology in Future Smart Grid 18
2.1 Introduction ...... 20
2.2 Impact of G2V on Grid ...... 21
2.3 V2G Technology ...... 23
xi Contents xii
2.4 A simple Structure of V2G System ...... 25
2.5 PHEV as Source of Stored Energy ...... 28
2.6 Benefits of V2G System ...... 30
2.6.1 Renewable energy supporting ...... 31
2.6.2 Environmental benefits ...... 32
2.6.3 Auxiliary services ...... 34
2.7 Challenges to V2G Concept ...... 37
2.8 Scope of Research ...... 38
2.9 Chapter Summary ...... 39
Chapter 3 Impact of PHEV Load on Power System 41
3.1 Introduction ...... 43
3.2 Stability Analysis by Linearization ...... 46
3.2.1 Linearization method ...... 46
3.2.2 Modal analysis of power systems ...... 51
3.3 Load Modeling ...... 54
3.4 PHEV’s Impact on SMIB System ...... 58
3.4.1 Mathematical model of SMIB system ...... 58
3.4.2 Linearization of SMIB with PHEV load ...... 61
3.4.3 Small signal stability analysis ...... 62
3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System . . . . . 67
3.5.1 PV generator ...... 68 Contents xiii
3.5.2 Mathematical model of solar generator ...... 72
3.5.3 Linearization of the system model ...... 73
3.5.4 Simulation results for PV system with PHEV load ...... 73
3.5.5 Small-signal stability analysis ...... 77
3.6 Impact of Dynamic PHEV Load on Distribution System ...... 79
3.6.1 Distribution system design ...... 80
3.6.2 Wind generator model ...... 80
3.6.3 PHEV’s interface with network ...... 85
3.6.4 Stability analysis of distribution system ...... 88
3.7 Chapter Summary ...... 90
Chapter 4 Design of Virtual Active Filter for Power System using PHEVs 91
4.1 Introduction ...... 92
4.2 Power Quality ...... 93
4.3 Power System Harmonics ...... 94
4.4 Filters in Power Systems ...... 95
4.5 V2G Technology for Filter Design ...... 96
4.6 PHEV Battery Modeling for Filter Design ...... 97
4.6.1 Bidirectional charger ...... 101
4.7 Controller Design ...... 108
4.7.1 General control principle ...... 108
4.7.2 Instantaneous power calculation ...... 108 Contents xiv
4.7.3 Power compensation ...... 112
4.7.4 DC voltage regulator and current controller ...... 115
4.8 Virtual Filter for HVDC Test System ...... 117
4.9 HVDC Case Study ...... 119
4.9.1 Case 1: Virtual filter at rectifier side ...... 120
4.9.2 Case 2 : Virtual filter on both converter sides ...... 123
4.10 Virtual Active Filter System for Improving Wind Power Quality . . 126
4.11 Modeling of Wind Farm ...... 127
4.11.1 Dynamic models of wind generators ...... 130
4.11.2 Rotor model ...... 130
4.11.3 Shaft model ...... 132
4.11.4 Induction generator model ...... 134
4.11.5 Aggregated model of wind turbine ...... 136
4.12 Harmonics in Wind Power ...... 138
4.13 Network Interfacing of Wind Generator ...... 138
4.13.1 Simulation results ...... 140
4.14 Chapter Summary ...... 143
Chapter 5 Design of Virtual FACTS Devices with PHEVs Park 144
5.1 Unified Power Flow Controller (UPFC) ...... 145
5.1.1 Introduction ...... 145
5.1.2 UPFC in power system ...... 146 Contents xv
5.2 Basic Structure of Virtual UPFC (VUPFC) ...... 150
5.2.1 VUPFC model ...... 151
5.2.2 Controller design ...... 154
5.2.3 Shunt converter control ...... 156
5.2.4 Series converter control ...... 159
5.2.5 Test system design and simulation results ...... 161
5.3 Unified Power Quality Controller ...... 166
5.3.1 Construction of VUPQC ...... 166
5.3.2 Controller design ...... 169
5.3.3 Simulation results ...... 178
5.4 Chapter Summary ...... 181
Chapter 6 Power Quality Improvement of Distribution Network and Microgrid using V2G 183
6.1 Introduction ...... 184
6.2 German Pilot Project (GPP) ...... 186
6.3 Modified Test System ...... 188
6.4 Specification of the Test System ...... 191
6.5 PHEV as Storage and Network Support ...... 192
6.6 Case1: Distribution Network without DG, PHEV or MVDC . . . . . 194
6.7 Case2: Integrating DG on Distribution Network ...... 195
6.8 Case3: Distribution Network with MVDC Coupler, DG and PHEV . 197 Contents xvi
6.9 Comparisons of Over all Voltage Profiles Comparison of the Distri-
bution Network ...... 198
6.10 PHEV Parks’ Connections with Network ...... 198
6.11 Scenario 1: PHEV as Virtual Active Filter for Solar Generation . . . 199
6.11.1 Controller design ...... 200
6.12 Scenario 2: PHEV as Virtual DVR for Network ...... 206
6.12.1 VDVR design and control ...... 207
6.12.2 Simulation results of VDVR ...... 208
6.13 Scenario 3: PHEV for the Improvement in Wind Power Quality . . . 210
6.13.1 Wind generator model and PHEV connection ...... 211
6.13.2 Simulation results ...... 213
6.14 Distribution Network’s Operation as Microgrid ...... 215
6.14.1 Islanding controller ...... 216
6.14.2 Simulation results ...... 217
6.15 Chapter Summary ...... 219
Chapter 7 Conclusions 220
7.1 Directions for Future Research ...... 223
Chapter 8 Appendices 225
8.1 Appendix-I ...... 225
8.2 Appendix-II ...... 226
8.3 Appendix-III ...... 228 List of Tables
2.1 Charging Power Levels ...... 21
2.2 Annual greenhouse emission reduction from PHEVs in the year 2050 . 34
3.1 Eigenvalues with PHEV load in SMIB system ...... 63
3.2 Eigenvalues with constant load in SMIB system ...... 63
3.3 Eigenvalues with PHEV load ...... 77
3.4 Eigenvalues with constant load ...... 78
3.5 PV system’s parameters ...... 78
3.6 Distribution system data ...... 82
3.7 Parameters used for induction generator ...... 85
4.1 Bidirectional charger mode of operations ...... 107
6.1 Modified GPP distribution system data ...... 192
6.2 Parameters of loads at each bus ...... 194
xvii List of Figures
1.1 General model of PHEV ...... 3
1.2 Market share of PHEVs from 2010 to 2050 ...... 4
1.3 Most popular PHEV models ...... 5
1.4 Charging/discharging cycle of PHEV in smart grid ...... 8
2.1 Potential V2G coverage in Australia ...... 26
2.2 The components and power flow of a V2G system ...... 29
2.3 General unidirectional and bidirectional power flow topology . . . . . 29
2.4 Annual greenhouse emission reduction from PHEVs ...... 33
3.1 Ford Escape 2010 with specification ...... 54
3.2 Battery equivalent network with parasitic branch ...... 55
3.3 Battery equivalent network ...... 55
3.4 System model SMIB with PHEV ...... 58
3.5 Angle oscillation with PHEV load ...... 64
3.6 Angle oscillation without PHEV load ...... 64
3.7 Voltage oscillation with PHEV load ...... 65
3.8 Voltage without PHEV load ...... 65
3.9 Bode diagram of gain and phase of system with PHEV load . . . . . 66
3.10 PV equivalent circuit ...... 68
3.11 General diagram of PV system with PHEV load ...... 69
xviii List of Figures xix
3.12 PV converter controller system ...... 70
3.13 P-V characteristic curve of PV cell with maximum power operating
point (x-axis PV array voltage and y-axis power) ...... 71
3.14 Flowchart of P and O method ...... 71
3.15 PV cell’s performance with PHEVs load under constant radiation and
temperature ( y-axis PV array voltage (V) and x-axis time) . . . . . 74
3.16 PV cell’s performance with PHEVs load under constant radiation and
temperature (y-axis PV array current (A) and x-axis time) ...... 75
3.17 PV cell’s performance with constant load under constant radiation
and temperature (y-axis PV array voltage (V) and x-axis time) . . . 75
3.18 PV cell’s performance with constant load under constant radiation
and temperature (y-axis PV array current (A) and x-axis time) . . . 75
3.19 PV cell’s performance with PHEVs load under variable radiation and
temperature (y-axis PV array voltage (V) and x-axis time) ...... 76
3.20 PV cell’s performance with PHEVs load under variable radiation and
temperature (y-axis PV array current (A) and x-axis time) ...... 76
3.21 PV cell’s performance with constant load under constant radiation
and temperature (y-axis PV array voltage (V) and x-axis time) . . . 76
3.22 PV cell’s performance with constant load under constant radiation
and temperature (y-axis PV array current (A) and x-axis time) . . . 77
3.23 Single line diagram of test distribution system ...... 81 List of Figures xx
3.24 PHEVs connection with power system ...... 86
3.25 PHEVs load connection with power system network ...... 87
3.26 Eigenvalues of the distribution system with PHEV load ...... 88
3.27 Wind generator voltage oscillation with constant and PHEV load . . 89
3.28 Wind generator angle oscillation with constant and PHEV load . . . 89
4.1 P-Q capability of PHEV’s battery ...... 98
4.2 PHEV’s connection to power system network ...... 98
4.3 Topology of unidirectional and bidirectional power flow ...... 101
4.4 Topology of general charger ...... 103
4.5 P-Q plane showing charger operation ...... 106
4.6 Vector diagram of charger operation ...... 107
4.7 PHEV’s connection as virtual active filter in network ...... 110
4.8 pq generation in controller ...... 111
4.9 Signal generation for inverter switching ...... 111
4.10 Signal generation for inverter switching with hysteresis band . . . . . 116
4.11 Single-line diagram of the CIGRE benchmark HVDC system . . . . 119
4.12 PHEVs connection as virtual active filter in HVDC network . . . . . 120
4.13 Modified CIGRE benchmark HVDC system with virtual filter on rec-
tifier side ...... 120
4.14 Harmonics current of load (one cycle) with virtual filter (case1) . . . 122
4.15 THD with virtual filter (case1) ...... 122 List of Figures xxi
4.16 Compensating current and reactive power output from virtual filter . 122
4.17 Harmonics spectrum of source current with virtual filter(y-axis odd
harmonics) ...... 123
4.18 Harmonics spectrum of load current with virtual filter (y-axis odd
harmonics) ...... 123
4.19 Modified CIGRE benchmark HVDC system with virtual filter on both
side...... 123
4.20 Harmonics current of inverter (one cycle) with virtual filter (case2) . 124
4.21 THD with virtual filter (case2) ...... 125
4.22 Harmonics spectrum of load current with virtual filter (case2)(y-axis
odd harmonics) ...... 125
4.23 PHEV connection with wind farm ...... 126
4.24 System structure of wind turbine with directly connected SCIG . . . 129
4.25 General structure of constant-speed wind turbine model ...... 130
4.26 Pitch angle control diagram ...... 137
4.27 System configuration ...... 139
4.28 System and load current at phase A, without PHEV as filter . . . . . 140
4.29 System and load current at phase A, with PHEV as filter ...... 141
4.30 Current harmonics spectrum without filter ...... 141
4.31 Current harmonics spectrum with filter ...... 141
4.32 Compensating current ...... 142 List of Figures xxii
4.33 Load current and voltage at phase A, without filter ...... 142
4.34 Load current and voltage at phase A, with filter ...... 142
5.1 Basic single line diagram of UPFC ...... 150
5.2 Simplified VUPFC circuit ...... 151
5.3 Victor diagram of VUPFC ...... 152
5.4 Flowchart for optimum size of VUPFC ...... 155
5.5 VUPFC model using PHEV ...... 156
5.6 Shunt converter connection with PHEV in network ...... 157
5.7 Modified decoupled PQ controller for shunt converter ...... 159
5.9 DC voltage controller ...... 160
5.8 Decoupled PQ controller for series converter ...... 161
5.10 6-Bus test system for VUPFC ...... 161
5.11 Shunt converter d axis current of VUPFC ...... 162
5.12 Series converter d axis current of VUPFC ...... 163
5.13 Shunt converter q axis current of VUPFC ...... 163
5.14 Series converter q axis current of VUPFC ...... 163
5.15 Output voltage from UPFCs ...... 164
5.16 Reactive power support from UPFC ...... 164
5.17 Real power support from UPFC ...... 165
5.18 Output DC voltage from VUPFC ...... 165
5.19 Basic configuration of VUPQC ...... 167 List of Figures xxiii
5.20 Combined series and shunt active filters for compensating voltage and
current using PHEVs ...... 172
5.21 pq generation in controller ...... 173
5.22 Signal generation for series inverter switching ...... 174
5.23 Converters connection with PHEVs in VUPFC ...... 175
5.24 Signal generation for shunt inverter switching ...... 176
5.25 Connection of shunt converter at time t=0.1 sec (source voltage) . . 178
5.26 Connection of shunt converter at time t=0.1 sec (source current) . . 179
5.27 Successive connections of the shunt and series converters at time
t=0.1 sec and t=0.15 sec respectively (source voltage) ...... 179
5.28 Successive connections of the shunt and series converters at times
t=0.1 sec and t=0.15 sec respectively (source current) ...... 180
5.29 Source and load current using VUPQC ...... 181
5.30 Compensating current of VUPQC ...... 181
6.1 German pilot project ...... 188
6.2 Simplified German pilot project network ...... 189
6.3 Modified network with virtual controller model using PHEV . . . . . 191
6.4 Capability of a vehicle batteries ...... 193
6.5 PHEV battery scheme ...... 193
6.6 Voltage profiles without DG, PHEV and/or MVDC coupler ...... 195
6.7 Voltage profile with wind generator at bus 6 ...... 195 List of Figures xxiv
6.8 Voltage profile with wind generator at bus 6 and solar generator at
bus9...... 196
6.9 Voltage profile with wind generator at bus 6, solar generator at bus
9 and PHEV as battery at bus 11 ...... 196
6.10 Voltage profile with wind generator at bus 6, solar generator at bus
9 and PHEV as battery at buses 11 and 6 ...... 197
6.11 Voltage profile with PHEV, DG, and MVDC coupler ...... 197
6.12 Voltage level without DG, PHEV and MVDC ...... 198
6.13 Voltage level with DG, PHEV and MVDC ...... 199
6.14 System configuration ...... 200
6.15 pq generation in the controller ...... 202
6.16 Signal generation for inverter switching ...... 202
6.17 System and load current without filter ...... 203
6.18 System and load current with filter ...... 204
6.19 Compensating current ...... 204
6.20 Load current and voltage without filter ...... 205
6.21 Load current and voltage with filter ...... 206
6.22 Single-line diagram of the VDVR ...... 207
6.23 Hysteresis controller for virtual DVR ...... 208
6.24 Supply voltage ...... 209
6.25 VDVR voltage ...... 209 List of Figures xxv
6.26 Load voltage ...... 209
6.27 PHEVs connection with wind farm ...... 211
6.28 PHEVs’ load connection with power system network ...... 212
6.29 Wind speed signal during test ...... 213
6.30 Real power output of wind farm ...... 214
6.31 Reactive power output of wind farm ...... 214
6.32 Output voltage of wind generator after fault ...... 215
6.33 Modified GPP network as microgrid ...... 216
6.34 Microgrid control scheme for Islanding opeation ...... 218
6.35 Voltage at bus 2 during islanding operation without PHEV as battery
base ...... 218
6.36 Voltage at bus 2 during islanding operation with PHEV as battery base219
8.1 Dynamic PHEV battery design in PSCAD ...... 228 List of Symbols
Symbols
A state transition matrix of the system
B input matrix of the system
C output matrix of the system
D output matrix of the system
x(t) state vector of the system
f(t) input vector of the system
y(t) the output vector of the system
F vector function of the system
Cθ battery thermal capacity
R0 thermal resistance between the battery and its environment
PS source thermal power
Qa ambient temperature of the battery surrounding
Kc empirical coefficient for a given battery
I⋆ reference current of the battery
δ power angle of the generator
ω rotor speed in synchronous reference frames
H inertia constant of the generator
xxvi List of Symbols xxvii
Pm mechanical input power to the generator
D damping constant of the generator
′ Eq quadrature-axis transient voltage
KA gain of the exciter amplifier
Vref reference terminal output voltage
V0 terminal output voltage
′ Tdo direct-axis open-circuit transient time constant of the generator
Xd direct-axis synchronous reactance
′ Xd direct axis transient reactance
Vt terminal voltage of the generator
Idg direct axis current of the generator
Iqg quadrature axis currents of the generator
Tdo direct-axis open-circuit transient constant
TR terminal voltage regulator time constant
Cdc capacitance of charger capacitor
idc output current of regulator
I input current of battery.
Id d-axis component of stator current of the synchronous machine
′ Tqo q-axis open-circuit transient constant
′′ Tdo d-axis open-circuit sub-transient constant List of Symbols xxviii
Xls armature leakage reactance
J moment of inertia of the rotor
ra stator winding resistance of the synchronous machine
TR terminal voltage sensor time constant
Vd d-axis terminal voltage of the synchronous machine
th Vi voltage at the i node of network
Vo output voltage of the terminal voltage sensor
Ka automatic voltage regulator gain
Vq q-axis terminal voltage of the synchronous machine
Vref reference voltage of the voltage regulator
Vt terminal voltage of the synchronous machine
Vs auxiliary input signal to the exciter
Efd equivalent emf in the exciter coil
Vtr measured voltage state variable after sensor lag block
Xq unsaturated q-axis synchronous reactance of the synchronous machine
′ Xq q-axis transient reactance of the synchronous machine
′′ Xd sub-transient reactance along d-axis
′′ Xq sub-transient reactance along q-axis List of Symbols xxix
′′ Tqo q-axis open-circuit sub-transient time constant of the synchronous machine
Id q-axis components of stator current of the synchronous machine
δ rotor angle of the synchronous machine
′ Tdo d-axis transient open circuit time constant of the synchronous machine
ω angular velocity of the rotor
R rotor radius of the wind turbine
ωm rotor shaft speed of the turbine
Vw wind speed
Awt swept area of the rotor
cp power coefficient
ho air density
θ pitch angle
λ tip-speed ratio
Ng gear ratio
Tae aerodynamic torque
Hm inertia constant of the turbine
HG inertia constant of the wind generator
Ks torsion stiffness
Dm torsion damping of the turbine
DG torsion damping of the wind generator
f normal grid frequency List of Symbols xxx
γ torsion angle
Tm Mechanical torque
X′ transient reactance of the wind generator
Xm magnetising reactance of the wind generator
Rs stator resistance of the wind generator
Rr rotor resistance of the wind generator
X rotor open-circuit reactance of the wind generator
Xr rotor reactance of the wind generator
Xs stator reactance of the wind generator
ids d-axis component stator current
iqs q-axis component stator current
ωG rotor speed of the wind generator
′ To transient open-circuit rotor time constant
vtm voltage sensor output
′ Edr d-axis transient rotor voltages
s slip of the induction machine
′ Eqr q-axis transient rotor voltages
Vds d-axis stator voltage of the induction machine
Vqs q-axis stator voltage of the induction machine List of Symbols xxxi
Abbreviations and Acronyms
AVR Automatic Voltage Regulator
BESS Battery Energy Storage System
CIL Constant Impedance Load
CPL Constant Power Load
DAEs Differential Algebraic Equations
DD Distributed Generator
DEMS Decentralized Energy Management Systems
DFIGs Doubly Fed Induction Generators
DOC Depth of Charge
DVR Dynamic Voltage Restorers
EPRI Electrical Power Research Institute
ESR Equivalent Series Resistance
EV Electric Vehicle
FACTS Flexible AC Transmission System
FSWT Fixed-Speed Wind Turbine
G2V Grid to Vehicle
GHG Greenhouse-Gas
GPP German pilot Project
HVDC High Voltage Direct Current List of Symbols xxxii
ICE Internal Combustion Engine
IEC International Electrotechnical Commission
IEC International Electrotechnical Commission
IEEE Institute of Electrical and Electronics Engineers
IG Induction Generator
LV Low Voltage
MPP Maximum Power Point
MPPT Maximum Power Point tracker
MV Medium-Voltage
MVDC Medium Voltage Direct Current
ORNL Oak Ridge National Laboratory
P&O Perturb and Observe
PCC Point of Common Coupling
PHEV Plug in Hybrid Electric Vehicle
PLO Phase Locked Oscillator
PV Photovoltaic
PV-AF Photovoltaic Active Filter
PWM Pulse Width Modulation
RES Renewable Energy Source
SAE Society of Automotive Engineers List of Symbols xxxiii
SCC Standards Coordination Committees
SMIB Single Machine Infinite Bus
SOC State of Charge
STATCOM Static Synchronous Compensator
SVC Static VAr (volt-ampere reactive) Compensator
THD Total Harmonic Distortion
UCTE The Union for the Coordination of Transmission of Electricity
UPFC Unified Power Flow Controller
UPQC Unified Power Quality Conditioner
V2G Vehicle to Grid
VDVR Virtual Dynamic Voltage Restorers
VSWT Variable-Speed Wind Turbine
VUPFC Virtual Unified Power Quality Conditioner
WECs Wind Energy Converters
WT Wind Turbine
WTGS Wind Turbine Generator System Chapter 1
Introduction
In the current energy world, a few most important issues need to be addressed for the attainment of a green, clean, sustainable energy future are, use of fuel for transportation, maintaining the quality of power generation and finding smart ways of energy utilization.
Global warming and climate change is one of the most complicated scientific, economic and political issue of time. Energy researchers have focused on the fact that our burning fossil fuels and traditional means of power generation have led to the rapidly increasing destruction of our environment and climate. As to reduce our carbon footprints and become more environmentally friendly, traditional trans- portation systems require attention, plug in hybrid electric vehicles (PHEVs) and electric vehicles (EVs) have entered the energy market as potential alternatives.
With a looming energy crisis, skyrocketing gas prices, nuclear leaks in Japan and the endless debate over global warming, it is comforting to know that there is a variety of realistic and clean alternative energy sources available as a result of years of research, innovation and application. As conventional vehicles cannot be immediately replaced by EVs in a transportation system, and a number of car users
1 Section 1.1 Background 2 will initially take time to adapt to a new setup, the most promising transporta- tion solution is the emerging concept of PHEVs. The increasing cost of fossil fuel makes PHEVs a smart choice for customers who will expect to be supplied with high-quality power by the relevant utilities. The rising number of PHEV fleets pen- etrating the market presents an enormous opportunity to use vehicle-to-grid (V2G) technology to improve the power quality of a utility grid. Environmental protection and sustainable development concerns have resulted in there being a critical need for cleaner energy technologies. Reductions in the use of fossil fuels, improved energy efficiency and increases in the availability of environmentally friendly energy sources has led to the use of intermittent renewable energy sources (RESs) as potential so- lutions. To reduce transmission losses and delays in the upgrading of transmission systems, these RESs are connected close to loads in distribution networks, which give rise to a new set of power quality problems due to the intermittency of these sources and dynamics of the interfacing equipment. Therefore, to solve these prob- lems and allow the possibility of adding PHEVs to a grid, an effective scheme is essential. This dissertation focuses on the impact of penetrating PHEVs into a grid and a smart way of utilizing them to enhance the quality and reliability of a power system.
1.1 Background
The Institute of Electrical and Electronics Engineers (IEEE) defines a PHEV as a hybrid vehicle, which has at least: Section 1.1 Background 3
(1) a battery storage of 4 kWh or more for powering the motion of the vehicle;
(2) a means of recharging its battery from an external source of electricity; and
(3) an ability to drive at least ten miles in an all-electric mode and consume no gasoline [1]. A general model of a PHEV is shown in Figure 1.1 [2].
Figure 1.1. General model of PHEV
PHEV’s share of the vehicle market over the years from 2010 to 2050, as predicted by the Electric Power Research Institute (EPRI) [3], is shown in Figure 1.2 in which, evidently it is expected that they will capture the majority of the market by 2050. Section 1.1 Background 4
Conventional Vehicles Hybrid Electric Vehicles Plug ͲIn Hybrid Electric Vehicles 100% 90% 80% sales
70% 60% vehicle 50% new
40% of 30%
Share 20% 10% 0% 2010 2015 2020 2025 2030 2035 2040 2045 2050 Year Figure 1.2. Market share of PHEVs from 2010 to 2050
Recently, several vehicle companies have been producing PHEV cars, with one of the most popular being the Toyota Prius which uses relatively small batteries.
This model is designed for low speeds of about 25 mph and a low starting power which switches to gasoline for distances of more than a few miles. PHEVs with large batteries and opportunities to be charged during off-peak hours, while allowing drivers to use electric power exclusively for a city drive of about 30 to 50 miles and then switch to gasoline for longer trips, are now being introduced. PHEVs offer customers the opportunity to purchase fuel at gasoline-equivalent prices of less than
$1.76 per gallon [4]. To travel at freeway speeds, a PHEV requires about 8 kWh of electrical capacity to be delivered to its drivetrain. Although, if it were charged Section 1.1 Background 5 from a 110-volt home circuit, only about 2 kW would flow from (or to) the grid.
For a modest additional cost of upgrading to residential 220 volt wiring, it could be
fitted with 8 kW grid connections. Various PHEV and EV models are now available in the vehicle market, and some of the most popular are shown in Figure 1.3 [5].
Audi A1 E-tron BMW ActiveE BYD F3DM Chevrolet Volt
Citroen Revolte Fisker Karma Ford Escape Ford Focus
Honda Fit Hyundai Blue-Will Kia Ray Mercedes S500 Vision
Nissan Leaf Renault Fluence Z.E. Suzuki Swift Tesla Motors Roadster
Tesla Motors Model S Toyota Prius Toyota 2nd Gen RAV4 Volvo V60 Figure 1.5 Major models of PHEV and PEV Figure 1.3. Most popular PHEV models
1.1.1 Smart grid and distribution network
Depending on the functions, the technologies and benefits, smart grids are defined in a number of ways [6–8], most of which describe some of its common characteristics, Section 1.1 Background 6 such as its application of digital processing and communications to a power grid, data and information management, integration of a reliable control system, bidirectional power flow and improved infrastructure for business processes. Although a great deal of the upgrading work conducted on the electrical grid modernization, particularly substation and distribution network automation, is now included in the universal concept of the smart grid, supplementary competencies are also being developed.
According to Ausgrid Australia,“A smart grid is a new, more intelligent way of supplying electricity. It combines innovations in digital communications, sensing and metering with the electricity network to create a two-way, more interactive grid. Smart sensors and devices installed in the electricity distribution network will help achieve fewer and shorter outages.” [9]. There is increasing discussion regarding integrating the smart grid with new loads, such as PHEVs, which can be distributed sources of energy for it [10]. From the operational point of view, a distribution network and microgrid are central area for the development of a smart grid [11]. The idea of enhancing the smart grid by integrating clean, distributed and renewable generation and PHEVs into it, is currently the most interesting area of research.
1.1.2 Overview of integration of renewable energy into grid
According to the Department of Energy, USA [12], renewable distributed generation is smart because it:
• reduces greenhouse gas emissions; Section 1.1 Background 7
• improves efficiency;
• helps defer system upgrades;
• reduces peak loads;
• alleviates congestion;
• improves reliability; and
• enhances energy security.
Wind and solar power are the most promising and mature technologies of the non-hydro-renewable energies and are desirable additions to a utility grid. They are renewable, economical, and reliable over the course of a year, clean and domestically applicable. However, their biggest remaining challenges are that solar energy is not available 24 hours a day, and the wind does not always blow at the hours of the day when power loads are highest, which means that, currently, shaping wind energy to meet customer demands requires reliance on pumped or compressed air- storage technologies or polluting fossil fuels, all of which are expensive. As finding a complementary resource to provide storage for wind energy remains desirable, this dissertation attempts to determine an economical way of improving the quality and reliability of renewable energy generation using PHEVs, focusing on the impact of charging them from the grid. However, other alternatives are also available such as super conducting magnetic energy storage (SMES), capacitor energy storage CES),
flywheel energy storage but that are also very expensive. Section 1.1 Background 8
1.1.3 Putting two and two together
As combining the benefits of PHEVs, renewable energy and a smart distribution grid would be advantageous, imagine the following scenario.
• Charging/discharging cycle : On a normal working day, a PHEV owner
wakes up in the morning, having had his PHEV fully charged on low-cost off-
peak power overnight, drives to his workplace and plugs his vehicle in so that
it recharges only when the energy cost is low and is fully charged for returning
home in the evening. As electric power is used for both trips, a saving of half
the normal fossil fuel cost is achieved.
8 AM Full charged PHEV
10 AM 10 PM Selling Recharge energy to PHEV utility PHEVs in Grid
6 PM 12 PM Selling Recharging energy to PHEV utility
Figure 1.4. Charging/discharging cycle of PHEV in smart grid Section 1.1 Background 9
• Decision Making: On an exceptionally hot day, the utility operator who
knows the forecast sends a message to the PHEV owner that the price of
electricity that day will be higher and could exceed the price of fossil fuel.
In that case, the PHEV owner decides whether to charge his vehicle or use
gasoline.
• Selling Energy: After arriving at work, the PHEV is plugged back into the
grid which draws down the PHEV’s stored energy. As the utility takes the
energy at a peak period, the PHEV owner enjoys a higher rate for selling his
battery energy.
• Charging Reasonable Price: At 12 PM, as power prices drop due to solar
generation, the grid starts recharging the PHEV and, although the energy rate
is slightly expensive, it is still cheaper than fossil fuel.
• Selling Energy at Evening Peak: In the evening, when there is a peak
period of load demand in the utility grid, the grid operator determines that,
if the system can meet demands until 8 PM by relying on PHEV power with-
out needing to start a reserve generating plant, the PHEV owner obtains the
maximum rate for selling his vehicle’s energy.
• Off Peak Charging: Overnight, the PHEV is fully recharged at low-cost
off-peak power using wind energy. The full charging/discharging cycle can be
represented by the Figure 1.4 Section 1.2 Motivation for the Current Research 10
1.1.4 Requirements to make the dream a reality
To make the above scenario a reality, power systems need to be updated while maintaining standards of quality, safety and security. In summary, the major re- quirements are:
• a stable and reliable distribution power network that can operate with addi-
tional loads, such as PHEVs, the impacts of which must be determined;
• a grid control system, which includes the real-time incremental cost of electri-
cal power as well as the status of all generating units and reserve requirements;
• an automatic bidirectional charging system, including communication and
information systems;
• a PHEV charging infrastructure and charging station;
• an accounting system;
• a suitable parking location and structure;
• solutions to practical problems, such as harmonics and the dynamic impact
of interconnection of PHEVs and grid; and
• designs of several alternative methods for consuming battery power rather
than directly drawing from and drying a battery, which can reduce its life.
1.2 Motivation for the Current Research
From the above discussion and various literature reviews, we can summarize the issues relating to the charging of PHEVs and utilizing them for the betterment of grid and renewable energy-based distribution networks as follows. Section 1.2 Motivation for the Current Research 11
• Grid instability analyses need to determine the causes of instability and gain
a deeper insight into the mechanisms of the grid instability phenomenon due
to the changing nature of modern power systems, the increased use of dynamic
loads and the integration of large-scale PHEVs.
• Before large-scale PHEVs and FACTS devices can be integrated into existing
power systems, the impacts of their penetration require a thorough evaluation.
• Currently available transmission and generation facilities are highly utilized,
with many power interchanges taking place through tie lines and geographical
regions. It is expected that, as this trend will continue in the future, there will
need to be more stringent operational requirements for maintaining reliable
services and adequate system dynamic performances.
• PHEVs can be a huge distributed energy source for a power system.
• As smart grids are needed at this time, developing PHEVs for them can make
a valuable contribution.
• As using PHEVs for the direct support of a power system may reduce their
battery lives and even their performances on the road, using them in the
alternative V2G mode is essential.
• In power systems, filters (active and passive filters) are used to correct the
power factor, current harmonics compensation and for overall power quality
improvement. Besides, the shunt active filter can also compensate for load
current unbalances, and allows the power source to see an unbalanced reactive Section 1.3 Contributions of this Research 12
non-linear load, as a symmetrical resistive load. A filter is one of the devices
used to improve the power quality of a power system but comes at a high price,
which, if reduced, could enable a utility’s users to enjoy higher-quality power.
• FACTS devices are sometimes essential for a power system, especially one
with renewable energy-based generation but, as their cost is an important
issue, their use is limited.
• A quality distribution network is a fundamental requirement for a future
smart grid with a microgrid mode of operation.
• A properly designed distribution network and microgrid will be expensive as
they need devices, which improve power quality.
• PHEVs will be the first choice of the next generation and, as they will be
connected to grids for long periods, design of some expensive but important
devices for power systems using PHEVs could be an economical solution for
utility engineers and PHEV users.
1.3 Contributions of this Research
In this thesis, a novel, dynamic impact analysis of the penetration of PHEVs into the grid is presented. This research work is aimed at providing deeper insights into the mechanisms of voltage instability caused by dynamic PHEV loads and attempts to improve present power systems using V2G technology. A power system filter and various FACTS devices are designed using PHEVs and compared with those of existing models. A benchmark distribution network performance is then analyzed Section 1.3 Contributions of this Research 13 using the proposed devices. The major contributions of this thesis in this direction are as follows:
Analyses
• The impacts of PHEV loads on the dynamic behavior of a single-machine
infinite-bus (SMIB) system under both small and large disturbances are ex-
amined.
• The dynamics of a photovoltaic (PV) cell with PHEVs load, which offers a
complete system for charging PHEVs with a PV cell, is investigated. System
dynamics are analyzed at the maximum power point while the perturb and
observe (P&O) method is used to ensure proper tracking of the maximum
power point from the PV cell. A small-signal stability analysis demonstrates
the charging impact of a dynamic PHEV load using a PV cell.
• The charging effect of a dynamic PHEV in a renewable energy-based electricity
distribution system is examined. To determine the system’s adequacy, it is
necessary to perform a micro-level analysis to assess the PHEV load’s impact
on the grid. The scope of such analysis covers the performances of wind and
solar generation with dynamic PHEV loads in a distribution network.
Virtual Filter Design
V2G technology is used to design virtual active filter for power systems. The poten- tial of a low-cost filter that utilizes the reactive power and filtering capabilities of Section 1.3 Contributions of this Research 14
PHEVs parked in charging stations is investigated for the auxiliary use of a PHEV battery, which has less impact on its lifetime. Simulations are performed for:
• the CIGRE benchmark HVDC network, which demonstrates that the pro-
posed virtual active filter improves power quality while meeting IEEE Std
519-1992;
• a wind farm to improve power quality, dynamic power factor correction and
harmonics current compensation; and
• a PV system for designing a shunt Active Filter (PV-AF) system to improve
the power quality of PV generation within a benchmark distribution network
with renewable energy generation and a MVDC network.
Virtual FACTs Design
Designing FACTS devices could be another use of PHEVs, which has less impact on battery life. In distribution networks, the concept of FACTS are applied with medium power, high-speed electronic switches known as custom power devices. In this dissertation, custom power devices are also designed as virtual FACTS devices using V2G technology. The following virtual FACTS are designed for various net- work:
• A Unified Power Flow Controller (UPFC) is a FACTS device, which can
fulfill multiple power flow control objectives, such as the need for reactive
shunt compensation, phase shifting and series compensation in transmission
level. Section 1.4 Thesis Outline 15
• A Unified Power Quality Conditioner (UPQC) can be used for multiple power
quality control, such as the need for reactive power compensation, voltage
flicker and harmonics current compensation in distribution level.
• Dynamic Voltage Restorers (DVRs) are used in distribution systems to protect
sensitive loads from voltage sags in distribution level.
Smart Network Design
Integrating distributed generation (DG), especially renewable energy sources, into and maintaining the power quality of a distribution network and islanding operation of microgrid, are the major challenges for developing a smart grid. The potential of PHEVs in the V2G mode of operation to provide a low-cost means of designing a centralized power quality conditioner using a PHEV charging station is explained in this work.
1.4 Thesis Outline
Based on the above objectives, an outline of this thesis is as follows:
Chapter 1 provides the background to this dissertation, including the motiva- tion behind it, and discusses its contributions.
Chapter 2 presents an overview of PHEVs in a power system and introduces the various V2G modes of operation, the general practice of V2G technology. The eco- nomical and technical advantages and disadvantages of V2G and G2V technologies, with a literature review, are discussed. Section 1.4 Thesis Outline 16
Chapter 3 introduces dynamic model of a PHEV and briefly reviews the lit- erature on the impact of the penetration of PHEVs into a grid. A single-machine infinite-bus system, solar system, wind system and distribution network are taken into consideration to identify the impact of PHEVs’ charging. Conventional lin- earization and modal analysis techniques commonly used in small-signal stability analyses are presented to determine their impacts under both disturbances and the normal operation of a power system.
Chapter 4 presents a control design algorithm for improving the power quality of a power system’s active filter using a dynamic PHEV park and a concept of bidirectional charger design. A power system model, HVDC model, active filter, test cases and control tasks are included. The virtual filter’s performances are evaluated through simulations and discussions, and conclusions provided.
Chapter 5 presents FACTS devices, such as UPFC, and UPQC designs, using
V2G technology. It concentrates on the possible ancillary support mechanisms pro- vided by PHEVs in the V2G mode of operation. Detailed case studies are used to illustrate the performances of some of the key FACTS devices which, are actually designed as virtual devices with PHEVs.
Chapter 6 introduces a distribution network which contains most of the possible uses of PHEVs in the V2G mode of operation to improve the performance of the grid, and presents a scenario how PHEVs can help to operate a microgrid without the need for additional storage devices even in the islanding mode. Section 1.4 Thesis Outline 17
Chapter 7 provides the thesis summary, conclusions and recommendations for future research.
Chapter 8 contains the appendices. Chapter 2
V2G Technology in Future Smart Grid
In a smart power network, PHEVs can act as either loads or distributed sources of energy. The two terms most commonly used to describe the interconnection of a power network and electric vehicle are ‘Grid-to-Vehicle (G2V)’ and ‘Vehicle-to-Grid
(V2G)’. When electric vehicles are connected into the grid to recharge their batter- ies or supply energy to it, they act as loads known as the G2V or V2G modes of operation respectively. This chapter reviews the impact of implementing the G2V mode, and the benefits and drawbacks of, and strategies for, the V2G interfacing of individual vehicles with a PHEV park. The performance of a power system net- work can be improved using V2G technology, which offers reactive power support, power regulation, load balancing, and harmonics filtering, which in turn, improve its quality, efficiency, reliability and stability. To implement V2G technology, a power network might require significant changes in its structure, components and controls, the issues for which include battery life, the need for concentrated communication between vehicles and the grid, the effects on distribution accessories, infrastructure changes, and social, political, cultural and technical concerns. As storage is essen- tial for a power system, distributed electric vehicles can be an economical storage
18 Section V2G Technology in Future Smart Grid 19 solution if it has a good plan for buying and selling its energy. Bidirectional power
flow technologies of V2G systems need to be addressed and the economic benefits of V2G technologies depend on vehicle aggregation and G2V/V2G strategies. In the future, it is expected that their benefits will receive greater attention from grid operators and vehicle owners. Section 2.1 Introduction 20
2.1 Introduction
Due to environmental and climate issues, along with the rising cost of petroleum, energy security and limited reserves of fossil fuels [13–15], PHEV technology has be- come of increasing interest. However, it is in an early stage of development and faces a few problems before it can be adopted worldwide, such as technical limitations, sociocultural obstacles and the fact that PHEVs currently cost more than conven- tional vehicles [16]. According to the EPRI, penetration of PHEVs into the USA’s vehicle market will be 35% by the year 2020 [17]. To attain a stable and versatile interfacing between a grid and PHEVs, standards and codes for system requirements are developed by various organizations, such as the automotive sector, the IEEE, the
Society of Automotive Engineers (SAE) and the EPRI. In this dissertation, PHEVs are chosen for analysis as they have a few advantages over hybrid electric vehicle
(HEV) and internal combustion engine (ICE) vehicles as they can act in the dis- charge mode as V2G devices and in the charging mode as G2V devices [18]. This chapter reviews V2G/G2V technologies on grids and customer requirements, cost analysis, challenges and policies for V2G interfaces of both individual PHEVs and vehicle fleets. To assess the impacts and utilization of PHEVs in utility distribution or transmission networks, their controls and usage prototypes need to be evaluated.
The SAE has defined three levels of charger for PHEVs [19], as summarized in Ta- ble 2.1. A PHEV behaves as a load when it needs to recharge its battery in the
G2V mode and as a generator when a utility grid takes power from its battery in Section 2.2 Impact of G2V on Grid 21 the V2G mode of operation. Its recharging and discharging characteristics depend on a few factors, such as its geographical location, the number of PHEVs in that particular area, its charging levels (charging current and voltage), battery state and capacity, and the connection type used (unidirectional or bidirectional) [20], [21]
Table 2.1. Charging Power Levels Power Level Description Power Level Level 1 Opportunity charger 1.4 kW (12A) (any available outlet) 1.9 kW(20A) Level 2 Primary dedicated charger 4 kW (17A) 19.2 kW(80A) Level 3 Commercial fast charger Up to 100 kW (12A)
2.2 Impact of G2V on Grid
First-generation mass-market PHEVs, such as the Chevrolet Volt and Nissan Leaf
[22], [23], connect to the grid for only battery charging, which is the most basic configuration. G2V includes conventional and fast battery charging systems, and the latter can stress a grid distribution network because its power is high, as a typical PHEV requires more than double an average household’s load [25]. Charging practices in different locations also have an effect on the amount of power taken from an electric grid by a fleet of PHEVs; for example, charging at work in congested urban centers can lead to undesirable peak load [24], which could require significant investments in expensive peak generation. Injected harmonics and a low power factor can be serious problems if the charger does not employ a state-of-the-art conversion for charging PHEVs at night, which has minimal impact on the power grid given suitable choices for intelligent controls [17], [25–29]. The increasing exploitation of Section 2.2 Impact of G2V on Grid 22
PHEVs is still a topical area of research. One of the foremost recent studies of smart- grid development with PHEVs is [30] that recognized the complexity of studying the impact of PHEVs on a smart grid, with the results depending on many factors (power level, timing, duration of PHEV connection to the grid) and possibly affecting several variables (capacity needs, emissions generated). As mentioned above, as a charging
PHEV may present a load to an electrical grid twice the order of magnitude of that of a typical home, connecting it may create power quality problems, such as momentary voltage drops. An interesting point about the simulations in [31], which assumed no control over the charging of vehicles, is that they showed voltage drops between 5% and 10.3% depending on the time of day and season. The simulation results showed how the voltage supplied to a house changed without and with PHEV charging where, for the latter, the drop in it increased from 1.7% to 4.3% while, for the former, much more random behavior was exhibited with average voltage drops of around 4% (although this eventually reached a value close to 1.7% once the PHEV was charged). These results point to a need to improve the quality of electrical energy delivery by utilizing smart technology to coordinate the charging of PHEVs.
The study in [31] was based on simulations using residential power consumption profiles. The shorter period of AC power consumption became a switched load, with levels similar to what would be expected to be observed in PHEVs with Level
2 charging profiles, that is, the notion that the grid had not experienced high- magnitude loads with a random switching profile was not true. This switching Section 2.3 V2G Technology 23 high-power AC consumption profile was masked by its aggregated effect on the grid.
One other important aspect when studying the effect of PHEVs on an electric grid is the grid’s stability for which damping components will need to be introduced into future designs for controlling PHEV charging. In view of these observations, as the impact, that PHEV charging will have on an electrical grid still needs to be studied, in this dissertation, a small-signal analysis is conducted to identify that on the stability of a utility grid in the next chapter.
2.3 V2G Technology
V2G describes a system in which PHEVs communicate with the power grid to sell demand response services by delivering electricity into the grid or throttling their charging rate. PHEVs can serve as stored and distributed energy resources as well as reserves for unexpected outages when they have proper on-board power electronics, smart connections to the grid and interactive charger hardware control [26], [32–35].
A bidirectional charging system is essential to support energy injection into the grid [36–39], as a unidirectional charger, although simple and easy to use in terms of control, can be used only for a G2V system.
A smart charging system and proper management can shift loads and avoid peaks while a proper controller can minimize the impact of PHEVs on the utility grid [28],
[29], [40]. Smart metering, communication and control systems play important roles in the direct coordination of the V2G and G2V modes of operation. The real- time, nonlinear pricing of a utility bill is one of the important factors for obtaining Section 2.3 V2G Technology 24 higher returns from grid-connected PHEVs [41]. In the V2G mode, interconnection between the grid and vehicle is essential and an individual vehicle or even a fleet of PHEVs can take part as spinning reserves for the grid. A group of cars in a park is more convenient to manage as a load for the grid [42] and more helpful as it can work as a distributed energy resource when necessary [43]. The potential benefits and economic issues of V2G technology are of great concern for researchers nowadays [10, 40–42, 44–58]. Another current issue is the use of RESs in a power network as, due to their sporadic natures, they need storage devices for which a
PHEV’s battery can be a solution as it offers the opportunity to store wind and solar energy at times of excess generation and provide possible backup when necessary
[18], [44–46]. The implementation of V2G technology has been explored in a number of ways through different research, such as for reactive power support [40], active power regulation, load balancing by valley filling [47], [48], [59], and peak load shaving [49], [50]. These systems can enable such ancillary services as frequency control and spinning reserves [10, 18, 42, 51–53], improve grid efficiency, stability, reliability [54] and generation dispatch [55], and reduce utility operating costs and, potentially, even generate revenue [41]. In addition, PHEVs owners benefit when electricity is cheaper than fuel for equivalent distances. Researchers have estimated that potential net returns from V2G methods range from $90 to $4,000 per year per vehicle based on the power capacity of electrical connections, market value, PHEV penetration and the energy capacity of the PHEV battery [18,31,32,56–58]. Besides Section 2.4 A simple Structure of V2G System 25 the intrinsic benefits of PHEVs, emissions have been reduced [29,60,61], and it has been reported that V2G strategies have the potential to displace the equivalent of
6.5 million barrels of oil per day in the USA [49]. Peterson et al. estimated the annual net social welfare benefits from the grid to be $300-$400 [62]. The design of a power system filter is one more option for using V2G technology, which in this dissertation, is described so that its implementation makes the grid smarter.
The expected increase in the number of electric vehicles produced could have a significant impact on the potential for utility-related energy storage as these vehi- cles can provide some of its benefits. Specifically, it may be cost-effective to charge electric vehicles when energy prices are low and then dispatch the power from them to support the grid, especially during grid emergencies. Using electric vehicles as distributed storage is an important complement to the expected increase in intermit- tently RESs, such as solar and wind power outputs, which are sometimes produced when the energy demand and price are low and can change rapidly [63]. If every suburb in Australia installed just one vehicle to a grid recharge point, Australia’s
V2G coverage would look like that in Figure 2.1 [64].
2.4 A simple Structure of V2G System
C. Pang et al. summarized the requirements for a simple V2G structure with en- ergy resources and an electrical utility as being an independent system operator and aggregator, a charging infrastructure and locations, a bidirectional electrical energy flow and communication between each PHEV and the aggregator, on-board Section 2.4 A simple Structure of V2G System 26
Figure 2.1. Potential V2G coverage in Australia Section 2.4 A simple Structure of V2G System 27 and off-board intelligent metering and control, and the PHEV’s battery charger and management [65]. In short, a power connection with a grid, suitable metering and control with effective communication can build a V2G system [66]. Figure 2.2 [67] shows a simple V2G system structure and Figure 2.3 the power flows within the charger. In general, although communications must be bidirectional to report a battery’s status and receive control commands [68, 69], achieving intelligent meter- ing and control that are aware of a battery’s capacity and state-of-charge (SOC) is challenging [18], [70–72]. Both on-board and off-board smart meters have been pro- posed to support V2G methods [40], [59], [73], smart metering can make PHEVs into controllable loads to help combine them with renewable energy [74], GPS locators and on-board meters are useful [54], [73] while sensors and smart meters on charg- ing stations can monitor and exchange information with the relevant control center through a field area network [66]. Also, control and communication are essential for services such as dynamic adjustments that track intermittent resources and alter charging rates to track power prices, frequency or power regulation, and spinning reserves [18], [75–78], for which a variety of protocols have been discussed, including
Bluetooth, Home- Plug, Z-Wave and ZigBee, [79–83]. In US, the IEEE and SAE provide the necessary communications requirements and specifications [84–86] while the National Electric Infrastructure Working Council (IWC) has defined a commu- nications standard to enable PHEVs to communicate with chargers [87], [88]. PHEV chargers without state-of-the-art power electronics can produce deleterious harmonic Section 2.5 PHEV as Source of Stored Energy 28 effects on a distribution system [89]. The IEEE-519 [90], IEEE- 1547 [91], SAE-
2894 [92] and International Electrotechnical Commission’s IEC-1000-3-6 [90], [92] standards limit the allowable harmonic and DC current injections into the grid with which PHEV chargers are usually designed to fulfill. Sophisticated active power converter technology has been developed to reduce harmonic currents and provide a high power factor [37], [93–96] while shock hazard risk reduction for PHEV charg- ing is addressed in the standard for personnel protection systems for PHEV supply circuits [97].
2.5 PHEV as Source of Stored Energy
A single PHEV’s battery storage capacity is small relative to that of the grid. How- ever, the better coordination and reliability of a smart grid can be achieved by aggregating PHEVs as storage devices [35], [42], [98]. An aggregator can be a com- munication or controller device, or an algorithm that plays an effective role between
PHEV owners, the electricity market and distribution and transmission system oper- ators [99–101]. Both aggregated vehicles and the grid need to be properly controlled to maintain the stability of the grid [102]. Figure 2.2 shows an aggregator in a
V2G system. One of its major roles is to manage PHEVs to operate in the V2G mode whenever the grid needs power [103]. Each PHEV can be contracted for this service in a cost-effective way by an aggregator that understands its battery’s SOC condition [54], [104, 105]. In an aggregated smart grid environment, vehicles can Section 2.5 PHEV as Source of Stored Energy 29
Figure 2.2. The components and power flow of a V2G system
Bidirectional Bidirectional Filter AC/DC DC/DC Converter Converter
Grid
Figure 2.3. General unidirectional and bidirectional power flow topology Section 2.6 Benefits of V2G System 30 engage and disengage while performing ancillary services of the grid [75] and main- taining the maximum and minimum contract limits. Considering each vehicle as an individual decision maker and the aggregator as the coordinator, C. Wu et al. pro- posed a method of smart pricing and optimal frequency regulation [106]. Another optimal frequency regulation controller for a V2G aggregator was designed by Han et al. [107] while the western Danish power system was used for the long-term aggregation of PHEVs in [108]. In the industrial networks MOBIE [109] and Bet- ter Place [110], the aggregation concept was successfully implemented, and it was found that control and communication with individual vehicles was much difficult than with the aggregator [78].
2.6 Benefits of V2G System
PHEVs can support V2G mode of operation because on average, in the USA they travel on the road for only 4-5% of the day while sitting in home garages or parks for the rest of the time [35], [73], [76]. Several services, such as voltage and frequency regulation [35], [42], [51], [52], spinning reserves, reactive power support, peak shav- ing, valley filling (charging when demand is low), load following and energy bal- ance [40], [49], [59] could be provided by PHEVs. These services are sometimes essential for power system while using V2G system overall costs could be reduced and, thereby, prices to customers, and selling energy to the grid could improve load factors and reduce emissions [17], and possibly replacing large-scale energy storages. Section 2.6 Benefits of V2G System 31
2.6.1 Renewable energy supporting
The power quality of intermittent source of energy wind and solar can be improved using PHEVs as storage and filter devices [35], [45], [46], [108], [111–114]. The combination of PHEVs and renewable energy sources can make the grid more stable and reliable. The unpredictable nature of wind speed make the wind energy sources strongly intermittent and leading to imbalances [45], [115]. Solar radiations are available during the day while the peak energy demand occurs in the evening which refers the excess solar energy generation at the time when exciting grid does not need it [77].
A number of studies have been done to combine PHEVs with renewable energy sources for different purposes such as using as battery energy storage system (BESS) and reactive power support system. To overcome the fluctuation of wind power
Kepton and Tomic [35] investigated the possibility of using V2G technology while
Guille and Gross [42] proposed a structure using model predictive control (MPC) to analyze the positive effect of PHEVs on wind generator. To improve the power quality of a renewable energy based power network, Y. Ota et al. [116] design a control scheme of PHEVs as distributed spinning reserve. J. Wand et al. [117] have provided a combination of demand response and wind power integration while
Goransson et al. [118] elicited different strategies for integrating PHEVs into a wind- thermal power system.
A higher level penetration of renewable energy sources make the grid unstable, Section 2.6 Benefits of V2G System 32
PHEVs can improve the situation by charging and discharging their battery during the period of excess generation and the period of peak load demand respectively.
It can help the generation and load scheduling by consuming and supply energy whenever necessary [77]. Thus, V2G increases the flexibility of the grid to better utilize intermittent renewable sources.
2.6.2 Environmental benefits
PHEVs have emissions benefits over conventional vehicles, even when considering power generation emissions. CO2 emissions would fall significantly if PHEVs re- placed conventional ICE vehicles [110]. In V2G mode of operation, PHEVs could offer more environmental benefits and reduce greenhouse-gas (GHG) emissions [28],
[41]. CO2 emissions are estimated to drop from about 6.2 tons to 4 tons per year from a single vehicle [45], [68] while GHG emissions linked to driving depend on the type of fuel used for electricity generation. When the electricity is produced from fossil fuels, the environmental benefits of PHEVs are reduced, for renewable energy sources the GHG emissions almost 0 g/km while for coal-based plants it increases up to 155 g/km [119], even then their emissions may be 7-21% lower than those of
HEVs [29], [120] and 25% fewer GHG emissions than ICE vehicles [121]. The esti- mated reductions for PHEVs range from 15% to 65% in another USA-based study that examined low-carbon electricity sources [17], [29]. Long-term GHG reductions depend on reducing a grid’s carbon intensity [122], [123] and using PHEVs more than 33% emission can be reduced in future smart grid [124]. However automotive Section 2.6 Benefits of V2G System 33 and oil companies allege that EVs would have a net negative effect on the environ- ment because of lead discharges from battery manufacturing facilities and battery disposal [16], [125].
EPRI predicted the GHG impact of PHEVs over the years from 2010 to 2050 [3], as shown in Figure 2.4, in which three scenarios represent levels of both CO2 and total GHG emissions intensity and another three scenarios represent penetration of
PHEVs. Nine different outcomes are possible from these two sets of scenarios, which determine the potential long term impacts, as shown in the Table 2.2.
800
600
400 Reduction 2 2 Low
CO 200
(million metric tons) metric (million Medium 0 High Low Medium High
PHEV Penetration
Figure 2.4. Annual greenhouse emission reduction from PHEVs
From the analysis, it is found that each of the nine scenario combinations reduced annual GHG emissions significantly while reaching a maximum reduction of 612
High
474 517 612 Section 2.6 Benefits of V2G System 34
Table 2.2. Annual greenhouse emission reduction from PHEVs in the year 2050 PHEV Penetration CO Intensity 2 Low Medium High High 163 394 474 Medium 177 468 517 Low 193 478 612
million metric tons in 2050 (High PHEV fleet penetration, Low CO2 intensity case) and reductions from 2010 to 2050 can range from 3.4 to 10.3 billion metric tons.
2.6.3 Auxiliary services
To maintaining stability, reliability, supply and load balancing and overall power quality, power system sometime needs auxiliary services from external and internal network devices. PHEVs with a bidirectional charger can provide higher quality ancillary services, such services are voltage and frequency regulation, load leveling and peak demand management. A few of them are described here in the light of literature. An aggregator can be the main part of the system by creating a larger and desirable load for the utility [42], [126].
Voltage and frequency regulation
Voltage and frequency regulation in power system is always essential for the better quality power supply to the end user, V2G technology can provide this service and it could be one of the best service from PHEVs due to their high market value and minimal stress on a vehicle power storage system [73], [127]. An expensive process of cycling large generator in the network [106] is used to regulate the frequency in Section 2.6 Benefits of V2G System 35 present grid system to balance supply and demand for active power [128] and the reactive power demand is balanced by voltage regulation [128]. The charging and discharging of PHEVs can be an alternative way of frequency regulation [35].
A proper logic of charging and discharging of PHEVs can be implanted in the battery charger with a voltage control to compensate reactive power, which will select the current phase angle to operate in inductive or capacitive mode of charging
[128]. With an appropriate voltage control a PHEV can able to decide when it will charge or discharge it’s battery. As for example, when the grid voltage becomes too low, vehicle charging can stop and, when it becomes too high, charging can start [77]. Although penetration of large number of PHEV for charging batteries from the grid could be a reason of line over loading and voltage instability at a low voltage network [129], it can regulate the reactive power within the local network by V2G operation [31].
The Union for the Coordination of Transmission of Electricity (UCTE) defined three types of control for the frequency stability in the distribution network: primary, secondary and tertiary frequency control [130].
There are two regulation in power system: regulation up and down and separate prices are given for regulation down and regulation up capacity, depending on bids submitted during an auction. If a vehicle providing regulation submits a bid below the market clearing price, it is contracted for its available capacity. Over the course of the contracted hour, the vehicle will charge or discharge some percentage of its Section 2.6 Benefits of V2G System 36 contracted capacity. When the vehicle charges for regulation up, the owner will be charged for the energy consumed, and when it discharges for regulation down, the owner will be reimbursed for the energy provided. For secondary and tertiary frequency control, activation is also based on bids. When demand for regulation up arises, the lowest bid is activated first. Because delivering regulation down means charging at a lower price, this can be profitable for PEVs [56]. In [131], primary control is expected to have the highest value for V2G.
V2G research group at the University of Delaware compared the potential profit of V2G with existing grid regulation system and found that a PHEV with 10-15 kW power regulation capacity can earn $3,777–$4,000 per year [35], [73] and Brooks’s calculation on California City’s PHEVs shows the amount up to $5,038 for V2G application [132].
Load Shifting
By discharging during daily peaks and charging during low demand V2G can level the energy load. A local and global smart-charging control strategies could reduce the peak load [133]. Based on variation method an electricity pricing algorithm has been proposed for load leveling and identified an electricity price curve by M.
Takagi et al. [48] that could realize an ideal bottom charge while PHEV owners could minimize their electricity bills. L. Sana showed that even 4 million PHEVs charging load could be accommodated with the existing grid of Californian [134] and Section 2.7 Challenges to V2G Concept 37 for New York City it is observed that, up to 10% of peak capacity could be safely contributed by PHEVs at penetration levels of around 50%, which represented an economic benefit of $110 million per year [135]. Smart charger reduce peak load and shift energy demand [47], [136] while a little financial incentive for increased PHEV penetration when V2G is used for peak load reduction [62], [126].
2.7 Challenges to V2G Concept
V2G technology in a power distribution system may impact on its performance through overloading transformers and feeders, and in some cases this would reduce efficiency, produce voltage deviations and increase harmonics [137], [138]. The US
Department of Energy reported [139] specific challenges and opportunities in terms of communication needs. Security issues are another challenge at public charging facilities [140]. Battery degradation, investment costs, energy losses, resistance of the automotive and oil sectors are also impediments and barriers to V2G systems.
Rapid charging and discharging of PHEV’s battery for V2G concept may reduce the life of its battery. The rate of energy withdraw and cycling frequency determine the amount of battery degradation. Equivalent series resistance (ESR) and state of charge (SOC) are two major parameters proper controlling of which is a good way of slowing degradation [141–143].
According to Andersson [56], the investment cost of a battery is $300/kWh and a lifetime of 3,000 cycles at 80% depth of charge (DOC), the degradation cost is Section 2.8 Scope of Research 38
$130/MWh. For a 16 kWh battery Peterson et al. [144] calculated the maximum net annual degradation cost of battery for V2G services, which is only $10-$120.
Implementation of V2G technology in the present distribution network is likely to have a huge impact on equipment [145], [146]. Depending on the number and capacity of PHEV a distribution network could overload distribution transformers, increase voltage deviations, harmonic distortions and peak demand [147–152].
According to K. J . Dyke et al. [153], PHEVs penetration need a significant investment in electrical networks within the United Kingdom, while Fernandez et al.
[154] presented the impacts of investments in distribution networks and incremental energy losses for different levels of PHEV penetration.
2.8 Scope of Research
From the above literature, there are several issues, which have not yet been taken into consideration by researchers. In this dissertation, some are discussed, with the main focus being on the following.
• Consideration of PHEV battery dynamics for load calculation and a charg-
ing impact analysis on generation, which have not yet been discussed in the
literature.
• Introduction of a novel ancillary service of PHEVs through designing a filter
for a power system.
• Designs of virtual FACTS devices using PHEVs, which a few researchers have
addressed. Section 2.9 Chapter Summary 39
• A complete power quality solution for a benchmark distribution network using
V2G technology.
2.9 Chapter Summary
In this chapter, the impact of G2V and V2G technologies on power system and the benefits and challenges with the requirements and strategies for the interconnection between PHEVs and power system were reviewed. With the help of a bidirectional charger PHEVs can act like energy storage devices and serve the network whenever necessary. Unidirectional charger was the logical first step of the PHEV while the addition of on board bidirectional charger makes PHEV a smart part for the future smart grid with the opportunity of charging from any outlet of the grid and supports the network by injecting energy back to the grid. The economic benefits, CO2 emis- sions, cost and the impact on the distribution system depend upon the cooperation between PHEV owners, aggregators and efficient strategy for grid operators.
Efficiency, stability, reliability and generation dispatch of a grid can be improved by using V2G operations. This mode of operation can offer active power regulation, reactive power support, power sources, current harmonic filtering, peak shaving and load balancing by valley filling for the grid. To improve the reliability of intermittent renewable energy sources PHEVs can provide possible backup as energy storage and as load at the time of excess generation. Several auxiliary supports can be provided by PHEVs to power system, such as voltage control, spinning reserves, reduce grid operating cost and generate revenue. Based on the power market value, the number Section 2.9 Chapter Summary 40 of PHEVs and their battery energy capacities V2G mode of operation have the potential of net return between $90 and $4000 per year per vehicle.
The V2G operation includes the cost of battery degradation, the need for smart communication between the vehicles and the grid, effects on the distribution sys- tem, the requirement for infrastructure changes, and political, social, technical and cultural issues.
A number of proposed V2G technique have been discussed in this chapter, and it is shown that with a few disadvantage V2G technology is more economical and fea- sible from both owner and grid operator point of view. Political and environmental benefits can be ensured by the development of PHEVs. For the better interfacing of the PHEV and grid, the PHEV battery must have an extended life cycle with pre-determined standard of V2G and G2V connections. Chapter 3
Impact of PHEV Load on Power System
The effects of large-scale PHEV penetration need to be investigated before integrat- ing into the existing grid. This chapter analyzes the impact of PHEV load on the dynamic behavior of a SMIB system under both small and large disturbances, from which it can be summarized that PHEV loads can reduce the damping of the system under certain operating conditions.
The dynamics of a photovoltaic (PV) cell with PHEV load is investigated. This work offers a complete system for charging PHEVs with a PV cell, the dynamics of which are analyzed at the maximum power point (MPP) while the perturb and observe (P&O) method is used to ensure the tracking of the MPP from the PV cell.
The small signal stability analysis presented demonstrates that it is important to consider the dynamics of a PHEV load for charging with a PV cell.
The charging effect of a dynamic PHEV is presented in a renewable energy based electricity distribution system. For planning and design of a distribution system,
PHEVs are one of the most important factors as they will be a spinning reserve of energy for the power system [155,156] and a major load for a distribution network.
A micro-level analysis to determine the impact of a PHEV’s load on distribution
41 Section Impact of PHEV Load on Power System 42 system is essential to find out the system’s adequacy. This analysis covers the performances and stability analysis of renewable energy based distribution system, with dynamic PHEV load. Section 3.1 Introduction 43
3.1 Introduction
As the behavior of a power system becomes more complex due to continuous changes in load demand and diverse energy sources, which influence the characteristics and rates of its dynamic responses, as a result maintaining and analyzing the stability of a power system has been a challenge for power systems engineers [157].
The history of PHEVs covers a little more than a century, but most of the considerable commercial developments have taken place after 2002. Until 2010 most
PHEVs were conversions of production HEV, and the most prominent PHEVs were after market conversions of 2004 or later. As of January 2013, there are highway- capable PHEVs available in several international markets [158].
The interest in PHEV is increasing because of advances in battery and hybrid- electric power technologies, coupled with the need to take into account financial considerations, energy security requirements, environmental concerns and rising petroleum costs [159–161].
It is expected that, by the year 2030, PHEV penetration will be 25%, which represents a large additional load on power systems [162]. According to EPRI,
PHEVs would be recharged during overnight off-peak hours when there is a 60% reduction in total electricity generation. However, if 50% of all road vehicles are re- placed by PHEVs by the Year 2050, total electricity generation will need to increase by 8% [163]. Although this increased demand may produce large and unexpected peaks in power consumption, through demand management and the coordination of Section 3.1 Introduction 44 multiple PHEVs in distribution grids, charging PHEVs during off-peak hours might be possible [134], [164], [165].
Load management of power systems with increases in the penetration level of
PHEVs may not be so simple as many power system networks do not have an automation capability and even sufficient spare capacity. Distribution systems might require some changes due to new load levels, patterns and characteristics since many were designed decades ago based on considering the load levels and types at that time
[166]. Therefore, it is important to identify the effects for a particular load on the stability and control of a power system. In this study, the impact of charging PHEVs using a complete dynamic load model is analyzed. A system operator is concerned with power loss and transformer and feeder overloads while, for the customer, as well as the operator, power quality is very important. To assure the perfect operation and reliability of electrical appliances, large voltage deviations must be avoided. Overall efficiency is also an issue as the overnight recharging of PHEVs for owners, who will expect their vehicle’s batteries to be fully charged in the morning in preparation for daily use, will increase the loads of base-load power plants [31].
Although a few studies involving load-level and cost-benefit analyzes of PHEV penetration in power systems have been conducted [167–169], to date, the effects of battery charging in terms of a PHEV’s dynamic load characteristics [96], [170], [171], have not been studied. Normally, a PHEV is modeled as a constant power load
(CPL) or constant impedance load (CIL) [172]. Depending upon the dynamics Section 3.1 Introduction 45 of the vehicle’s charging system and battery, an interesting issue leads us to use a dynamic PHEV model to analyze the stability of existing power systems with
PHEVs as dynamic loads rather than CPL or CIL.
One powerful tool, which has been applied to assess a power system’s security, stability limits and regions of attraction for the post-fault equilibrium state is the energy function concept [173]. The post disturbance period is analyzed to determine whether the energy function value diminishes over time. Constructing these energy functions is much easier using a simple model of a generator with a CIL than large- order model complexities, such as excitation control, wind turbines, dynamic loads,
FACTS devices and networks with transfer conductances. However, the operating behavior of an interconnected power system can be analyzed using the theory of linear system analysis in which the dynamic behavior of the system must be implied to be linear. A better understanding of the nature of system dynamics helps to plan the control strategies necessary for the secure operation of the system.
A linearization technique is used throughout this dissertation to gain insights into problems, which could be considered in future designs of controllers. This chapter provides a general coverage of conventional linearization techniques and the modal analysis used in this research to determine the impacts of PHEV loads on different power system configurations. Renewable energy and a distribution network are used to realize the situation of a power system with PHEVs. Section 3.2 Stability Analysis by Linearization 46
3.2 Stability Analysis by Linearization
3.2.1 Linearization method
In general, to understand the behavior of a nonlinear power system in the neigh- borhood of an equilibrium point, the system is linearized, which usually, works reasonably well and has various advantages [174].
A set of nonlinear differential algebraic equations (DAEs) is used to express the power system’s dynamic behavior and algebraic equations are used to express the network’s power balance and generator’s stator current equations. A standard power
flow gives the initial operating state of the algebraic variables and by substituting them into the set of DAEs, the initial values of the dynamic variables are obtained from the solutions to the DAEs and then the set of DAEs is linearized in the vicinity of the equilibrium point.
A power system modelling approach involves forming the overall system equa- tions in the form of DAEs, as:
x˙ = F (x(t), f(t)) (3.1)
y = G (x(t), f(t)) (3.2) where x(t), f(t) and F are the state-space vector, input vector and vector function, respectively. The inputs are normally reference values, such as the speeds and voltages at individual units, and can be the voltages, reactance and power flows set in FACTS devices. y is the vector of outputs, and G is the vectors of the Section 3.2 Stability Analysis by Linearization 47 nonlinear functions relating to the systems output variables, where output can be power output, bus voltage, line power or current, etc.
Setting equation (3.1) equal to the zero vector:
x˙ = F (x(t), f(t)) = 0 (3.3)
The system is said to be at rest or at an equilibrium point since all variables are constant. Letting x0 be the state vector and f0 the input vector corresponding to the system at rest is:
F(x0, f0) = 0 (3.4)
Assuming actual system dynamics in the immediate proximity of the system’s nom- inal trajectories can be approximated as:
x = x0 + △x(t) (3.5)
f = f0 + △f(t) (3.6)
The prefix △ in equations (3.5) and (3.6) denotes a small deviation and as the new state (and every state) must satisfy equation (3.3):
x˙ = F (x(t) + △x(t), f(t) + △f(t)) (3.7) Section 3.2 Stability Analysis by Linearization 48
By time-differentiating both sides of equation (3.5):
x˙ =x ˙ 0 + △x˙(t) (3.8)
From equations (3.7) and (3.8):
x˙ 0 + △x˙(t) = F (x(t) + △x(t), f(t) + △f(t)) (3.9)
For small deviations, the non-linear function, F(x, f), in equation (3.9) can be ex- pressed in terms of a Taylor expansion, for which a general scalar function, F(x(t)), as a function of one variable, x, in a close interval around x0 is defined in [175] as:
F ′ F ′′ (x0) (x0) 2 F(x(t)) = F(x0) + (x(t) − x0) + (x(t) − x0) + ··· 1! 2! (3.10) F n(x ) + 0 (x(t) − x )n n! 0
If we omit the second and higher orders in equation (3.10):
′ F(x) = F(x0) + F (x0)(x(t) − x0) (3.11) Section 3.2 Stability Analysis by Linearization 49
So, that for ith order system, i = 1, 2, ··· , n we can write:
x˙i0 + △x˙i(t) = Fi(x(t) + △x(t), f(t) + △f(t))
∂Fi = fi(x0, f0)) + △x1(t) + ··· (3.12) ∂x1(t)
∂Fi ∂Fi ∂Fi + △xn(t) + △f1(t) + ··· + △fn(t) ∂xn(t) ∂f1(t) ∂fn(t)
Since xi0 = Fi(x0, f0) = 0:
△x˙ i(t) = Fi(x + △x, f + △f)
F F ∂ i(t) △ ··· ∂ i(t) △ = x1(t) + + xn(t) (3.13) ∂x1(t) x=x0 ∂xn(t) x=x0 f=f0 f=f0 F F ∂ i ∂ i(t) + △f1(t) + ··· + △fn(t) ∂f1 x=x0 ∂fn(t) x=x0 f=f0 f=f0
Similarly, for output signal yj(t):
G G ∂ j(t) ∂ j(t) △yj(t) = △x1(t) + ··· + △xn(t) ∂x1(t) x=x0 ∂xn(t) x=x0 f=f0 f=f0 (3.14) G G ∂ j(t) ∂ j + △f1(t) + ··· + △fn(t) ∂u1(t) x=x0 ∂fn(t) x=x0 f=f0 f=f0
Finally, in matrix form:
△x˙ = A△x(t) + B△f(t) (3.15)
△y = C△x(t) + D△f(t) (3.16) Section 3.2 Stability Analysis by Linearization 50 where
F F ∂ 1(t) ... ∂ 1(t) ∂x1(t) ∂xn(t) . . A = . ... . (3.17) F F ∂ n(t) ... ∂ n(t) ∂xn(t) ∂xn(t) x=x0 f=f0
F F ∂ 1(t) ... ∂ 1(t) ∂f1(t) ∂fn(t) . . B = . ... . (3.18) F F ∂ n(t) ... ∂ n(t) ∂fn(t) ∂fn(t) x=x0 f=f0
G G ∂ 1(t) ... ∂ 1(t) ∂x1(t) ∂xn(t) . . C = . ... . (3.19) G G ∂ n(t) ... ∂ n(t) ∂xn(t) ∂xn(t) x=x0 f=f0
G G ∂ 1(t) ... ∂ 1(t) ∂f1(t) ∂fn(t) . . D = . ... . (3.20) G G ∂ n(t) ... ∂ n(t) ∂fn(t) ∂fn(t) x=x0 f=f0 where △x(t) is the state vector of dimension n, △f(t) the input vector of dimension r, △y(t) the output vector of dimension m, A the state matrix of size n × n, B the input matrix of size n × r, C the output matrix of size m × n and D the feedforward Section 3.2 Stability Analysis by Linearization 51 matrix of size m × r.
3.2.2 Modal analysis of power systems
In this section, some basics of modal analysis, which are necessary for understanding controller design methods, are introduced. By linearizing the nonlinear power system model, about an operating point, the total linearized system model is represented by the equations (3.15) and (3.16).
By taking the Laplace transform of the equations (3.15) and (3.16), we obtain:
s△x(s) − △x(0) = A△x(s) + B△(s) (3.21)
△y(s) = C△x(s) + D△(s) (3.22)
A formal solution of the state equations results in:
△y(s) = C(sI − A)−1 [△x(0) + B△f(s)] + D△(s) (3.23) where I represents the identity matrix. The equation
det(sI − A) = 0 (3.24) is called the characteristic equation of matrix A and the values of s, which satisfy it are the eigenvalues of matrix A. The natural modes of the system’s response are related to the eigenvalues and an analysis of the eigen properties of A provides valuable information regarding the stability characteristics of the system [176]. Section 3.2 Stability Analysis by Linearization 52
Because power systems are physical systems, A is an n by n matrix and has n solutions of the eigenvalues as:
λ = λ1, λ2 . . . , λn (3.25)
For any eigenvalue, λi, the n-column vector, ϕi, which satisfies (3.26), is called the right eigenvector of A associated with the eigenvalue λi [176] as:
Aϕi = λiϕi (3.26)
Similarly, the n-row vector, ψi which satisfies:
ψiA = λiψi (3.27)
is called the left eigenvector associated with the eigenvalue λi.
Physically, the right eigenvector describes how each mode of oscillation is dis- tributed among the systems states and is called the mode shape. The left eigenvec- tor, with the input coefficient matrix and the disturbance determines the amplitude of the mode in the time-domain solution for a particular case [176].
To express the eigenproperties of A succinctly, the modal matrices are also in- troduced:
Φ = [ϕ1, ϕ2, . . . , ϕn] (3.28)
Ψ = [ψ1, ψ2, . . . , ψn] (3.29) Section 3.2 Stability Analysis by Linearization 53
If we define a transformed vector, z, as x = Φz (since ΦΨ = I, we have z = Ψx), for (u = 0):
z˙ = Φ−1AΦz = Λz (3.30)
λit λiψix(0) This means that zi(t) = e z(0) = e and finally:
∑n λit x(t) = ϕiψix(0)e (3.31) i=1
• The ith element of z(t) is called the ith mode of the system corresponding to
the eigenvalue λi.
th • The i right eigenvector, ϕi, is the mode shape corresponding to the eigenvalue
λi.
th • The j element of the left eigenvector, ψi, ψij, gives the contribution of the
jth state in the ith mode.
• For a complex eigenvalue, λi = ai + jbi and its eigenvector, ϕi = Ui + jVi, we
have:
AUi = aiUi − biVi (3.32)
AVi = aiVi + biUi (3.33) Section 3.3 Load Modeling 54
3.3 Load Modeling
On average, more than 50% of cars in the US are driven about 40 km/day [134].
To evaluate the impact of PHEVs we consider a driving range of 65 km/day, which means that the capacity of a PHEV battery will be 12 kWh because 0.186 kWh of battery energy is required to drive one km [31], [165].
Practical data available for the advanced research vehicle manufactured by Ford as shown in Figure 3.1 [177] is used in this work and its vehicle specification are same as those of the calculated model.
Vehicle Specification Model Escape 2010 Ford Output 155hp @6000 rpm, 2.5 L Battery Lithium-Ion Electric Drive and Charge System No. of cell 84 Cell voltage 3.6 V System Voltage 302 V Charging Voltage 120 V Charging Current 30 A Charger power 3.6 kW Pack Energy 12 kWh Figure 1: 2010 Ford Escape Figure 3.1. Ford Escape 2010 with specification
A dynamic model of a lead acid battery [178] is selected to develop a suitable model of PHEV load, the elements of which are not constant as they depend on electrolytic temperature as well as the battery’s state-of-charge (SOC).
The battery equivalent network is presented in Figure 3.2, in which θ represents the electrolyte temperature and Im is a integral part of the total current, I, which is used to store charge in the battery. Another part of the total current entering the Section 3.3 Load Modeling 55
R0 P
im Ip
Z (θ,SOC) Z m (θ,SOC) p
+ +
E m (θ,SOC) E p (θ,SOC) - - N
Figure 3.2. Battery equivalent network with parasitic branch battery flows through the parasitic branch. The parasitic reaction is a continuous process, that draws current but does not participate in the main reaction. The volt- age at this branch is nearly equal to the voltage at the pin while the power dissipated in the real parts of impedances Zm and Zp is converted into heat. Impedance of the main reaction branch increases with an increasing charge and as a result, the termi- nal voltage of the parasitic branch and the current, Ip, rise. At a full state of battery, the impedance of the main reaction branch approaches infinity [169], [179–182].
C1 Cn R Im P 0 R1 Rn
I1 Ip
V Em
N
Figure 3.3. Battery equivalent network Section 3.3 Load Modeling 56
This battery model can be represented as an RLC network as shown in Figure 3.3 and the number of R-L-C blocks can be limited as the specific speed of evolution of the electrical quantities evolve rapidly for PHEVs [178], [179]. The parameters used for the battery are given in Appendix 8.1.
For the third order battery dynamic model is developed considering the charging current, extracted charge and electrolyte temperature.
Charging current : a current entering the battery would be the integral of the cur- rent itself. This current has two main characteristics, that are:
1. The charging current behavior is far from being linear.
2. The charging efficiency cannot be consider equal to 1.
Extracted charge : A per unit measure of the level of the discharge of a battery correlate the charge that actually extracts from the battery, starting from a battery completely full with the charge that can be extracted under given, standard con- ditions. The charge that can be drawn from a battery with a constant discharge current at a constant electrolyte temperature is higher with higher electrolyte tem- peratures and lower discharge current. It depends also on the voltage reached at the end of the considered discharge to measure the capacity.
Electrolyte temperature : Since the batteries are extensive components, each elec- trolyte point has a temperature of its own. Starting from information on the temper- ature of the air surrounding the battery and some computation of the heat generated, the electrolyte temperature can be computed. Section 3.4 PHEV’s Impact on SMIB System 57
The dynamic equations for the model are [178], [179]:
( [ ] ) ˙ 1 1 − − VdcR0A1Qe Ke(273 + θ)Qe − I = Vdc(1 R0) Em + ⋆ + ⋆ I (3.34) T1 ZR KcCI KcCI
[ ] ˙ − − − VdcR0A1Qe Ke(273 + θ)Qe Qe = Vdc(1 R0) Em + ⋆ + ⋆ (3.35) KcCI KcCI
[ ] − ˙ 1 θ Qa θ = − Ps − (3.36) Cθ Rθ where:
Cθ is the battery thermal capacity;
R0 is the thermal resistance between the battery and its environment;
PS is the source thermal power, namely the heat generated internally in the battery;
Qa is the ambient temperature, namely the temperature of the environment(normally air)surrounding the battery;
Kc is the empirical coefficient for a given battery;
I⋆ is the reference current; and
Em, Ke, A1 are constant for a particular battery. Section 3.4 PHEV’s Impact on SMIB System 58
Xe Ie Vt IL
R0
+ Vdc Rp R1 C1 It - 0 Em V∞ E d
Figure 3.4. System model SMIB with PHEV
3.4 PHEV’s Impact on SMIB System
The dynamic PHEV model is then integrated with a single machine infinite bus system (SMIB) to investigate the impact of PHEV load on a simple power system as shown in Figure 3.4, where power supply to the load PL = 15 MW, QL = 1.5
MVAR from the infinite busbar and local generator PG = 3 MW, QG = 2.25 MVAR.
The parameters used for the SMIB are given in Appendix 8.1
3.4.1 Mathematical model of SMIB system
Utilizing the battery model in an electrical network, the total system is divided into three parts: the generator model; battery charger; and battery load. The syn- chronous generator can be modeled using the following set of differential equations Section 3.4 PHEV’s Impact on SMIB System 59
[176], [183–185]:
˙ δ = ω0ω − ω0 (3.37) D ω ( ) ω˙ = − ω + 0 P − E′ I (3.38) 2H 2H m q qg 1 ˙ ′ − − − ′ Eq = ′ [KA(Vref V0) (Xd Xd)Idg] (3.39) Tdo ˙ 1 V0 = (Vt − V0) (3.40) Tr where:
δ is the power angle of the generator;
ω is the rotor speed in synchronous reference frames;
H is the inertia constant of the generator;
Pm is the mechanical input power to the generator;
D is the damping constant of the generator;
′ Eq is the quadrature-axis transient voltage;
KA is the gain of the exciter amplifier;
Vref is the reference terminal output voltage;
Vinf is the infinite bus voltage;
V0 is the terminal output voltage;
′ Tdo is the direct-axis open-circuit transient time constant of the generator;
Xd is the direct-axis synchronous reactance;
′ Xd is the direct axis transient reactance; Section 3.4 PHEV’s Impact on SMIB System 60
XT is the total reactance at load side;
Vt is the terminal voltage of the generator;
Idg is the direct axis current of the generator;
Iqg is the quadrature axis currents of the generator;
Tdo is the direct-axis open-circuit transient constant; and
TR is the terminal voltage regulator time constant.
The terminal voltage of the generator can be expressed as:
√ ′ − ′ 2 ′ 2 Vt = (Eq XdIdg) + (XdIqg)
The charger is considered as a capacitor parallel with the load and equation is expressed as: √ ′ ′ − ′ 2 ′ 2 ′ E (Eq XdIdg) + (XdIqg) (Xd + Xe) ˙ − 1 q Vinf − Vdc = ′ + ′ (3.41) C Xd Xe XdXe and the current equations are:
( ) E′ − q Vinf Vdc Idg = ′ + ′ + ′ cos δ (3.42) Xd Xd + XT + Xe Xd + XT
( ) Vinf Vdc Iqg = ′ + ′ sin δ (3.43) Xd + XT + Xe Xd + XT Section 3.4 PHEV’s Impact on SMIB System 61
3.4.2 Linearization of SMIB with PHEV load
Using the Taylor series expansion method and truncating the higher order terms, the linearized form of equations (3.34)–(3.41) can be written as :
˙ ∆δ = ω0∆ω (3.44) D ω ( ) ∆ω ˙ = − ∆ω + 0 −E′ ∆I − I ∆E′ (3.45) 2H 2H q0 qg qg0 q 1 ˙ ′ − − − ′ ∆Eq = ′ [ KA∆V0 (Xd Xd)∆Idg] (3.46) Tdo ˙ 1 ∆V0 = (∆Vt − ∆V0) (3.47) Tr 1 X + X′ ˙ ′ − e d ∆Vdc = ′ ∆Eq ′ ∆Vt (3.48) CX X Xe d [ ( d ) ( ) ˙ 1 − R0A1Qe R0A1Vdc Ke(273 + θ) ∆I = 1 R0 + ∗ ∆Vdc + ∗ + ∗ ∆Qe T1ZR KcCI KcCI KcCI ( ) ] − Keθe ∆IZR + ∗ ∆θ (3.49) KcCI
( ) [ ] ˙ − − R0A1Qe R0A1Vdc Ke(273 + θ) ∆Qe = 1 R0 + ∗ ∆Vdc + ∗ + ∗ ∆Qe ( ) KcCI KcCI KcCI Keθe + ∗ ∆θ (3.50) K(cCI ) 1 1 ∆θ˙ = 1 − ∆θ (3.51) C0 R0 Section 3.4 PHEV’s Impact on SMIB System 62
The suffix 0 denotes the values at the operating point given by:
√ ′ − ′ 2 ′ 2 Vt0 = (Eq0 XdIdg0) + (XdIqg0)
( ) E′ V V ′ − q0 inf dc0 Idg0 = ′ + ′ + ′ cos δ0 Xd Xd + XT + Xe Xd + XT
( ) Vinf Vdc0 Iqg0 = ′ + ′ sin δ0 Xd + XT + Xe Xd + XT
The linearizations of the current equations are:
[ ] E′ − q0 ′ − Vinf sin δ0 − Vdc0 sin δ0 Vdc0 cos δ0 ∆Idg = ′ ∆Eq ′ ′ ∆δ + ′ ∆Vdc Xd Xd + XT + Xe Xd + XT Xd + XT
[ ] Vinf cos δ0 Vdc0 cos δ0 Vdc0 sin δ0 ∆Iqg = ′ + ′ ∆δ + ′ ∆Vdc Xd + XT + Xe Xd + XT Xd + XT
Finally, the following equation is used to represent the state space equation of the linearized model of the system.
∆x ˙ = A∆x (3.52) where A is the state matrix and x is the states of the system.
The elements of A that are necessary for the analysis given in Appendix 8.2.
3.4.3 Small signal stability analysis
Through load flow analysis, the operating point of the system is determined and then its operating point, state matrix is calculated using the power system parameters. Section 3.4 PHEV’s Impact on SMIB System 63
To predict the system’s response, we conduct a small-signal stability analysis and, after linearization of the system, the eigenvalues shown in Table 3.1.
Table 3.1. Eigenvalues with PHEV load in SMIB system 0.000 −5.400 −0.280 + 15.920i −0.280 − 15.920i −12.761 −0.512 + 7.220i −0.512 − 7.22i −3.333
From the eigenvalues in Tables 3.1, it is seen that the system has marginal stable
′ mode 1. The participation factors are ∆δ = 0.0072, ∆ω = 0.0002, ∆Eq = 0.0002,
∆V0 = 0.0001, ∆Vdc = 0.040, ∆I1 = 0.003, ∆Qe = 0.070, ∆θ = 0.0116 which show that the states, Vdc, ∆Qe and ∆θ have the highest participation in system marginal stability.
Another set of eigenvalues has been found using constant load for the same single machine infinite busbar system shown at Table 3.2.
Table 3.2. Eigenvalues with constant load in SMIB system −0.22.700 −11.202 −0.267 + 7.983i −0.267 − 7.983i −12.761
From Tables 3.1 and 3.2, it is clear that the system has low-frequency oscillations.
To compare the effects of the constant and dynamic PHEV loads, a simulation is Section 3.4 PHEV’s Impact on SMIB System 64 performed.
A single phase-to-ground fault is applied on the connecting line of the dynamic load and cleared after 0.05s. Figures 3.5 and 3.6 show the angles and Figures 3.7 and 3.8 show the voltage responses (in pu) of the local generator with and without
PHEV loads, respectively, from which it is clear that penetrations of PHEV loads make the system more oscillatory and reduce its damping.
Angle oscillation with PHEV load 0.09
0.08
0.07
0.06 Angle Angle (rad) 0.05
0.04 0 5 10 15 20 Time (s)
Figure 3.5. Angle oscillation with PHEV load
Angle oscillation without PHEV load 0.09
0.08
0.07
0.06 Angle(rad)
0.05
0.04 0 5 10 15 20 Time (s)
Figure 3.6. Angle oscillation without PHEV load Section 3.4 PHEV’s Impact on SMIB System 65
Voltage oscillation with PHEV load 0.8925
0.892
0.8915
Voltage 0.891
0.8905
0.89 0 5 10 15 20 Time (s)
Figure 3.7. Voltage oscillation with PHEV load
Voltage oscillation without PHEV load 0.893
0.8925
0.892
Voltage 0.8915
0.891
0.8905 0 5 10 15 20 Time (s)
Figure 3.8. Voltage without PHEV load
Again, voltage variations can influence the damping of electromechanical oscil- lations [186] and the parameters of the PHEV load on the frequency response. To investigate the load-system interaction, the system is considered as a transfer func- tion, with the load and voltage deviations the input and output respectively. Figure Section 3.4 PHEV’s Impact on SMIB System 66
3.9 presents the bode diagram of the system. It is found that damping deterioration occurs due to the load and system parameters, which reinforce oscillation.
Figure 3.9. Bode diagram of gain and phase of system with PHEV load Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 67
3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System
PV energy system is one of the cleanest power-generating technologies available today and has less impact on the environment compare to other sources. PV converts the sun’s rays into electricity and produces no air pollution, waste or noise. Increased use of PV energy to generate electricity from the sun’s rays decreases our dependence on fossil fuels and imported sources of energy. As a result, solar energy can be an effective driver of economic development. In recent year interest of using solar energy is rapidly rising and worldwide PV market installations reached a record high of 7.3 gigawatt (GW) in 2009, representing a growth of 20% over the previous year [187].
Increasing penetration of PV generation with load levels, patterns and character- istics [166], [165] need to be determined for the stability and better control of power system and it is important to identify the effects of a specific load, such as PHEV and EV. In this thesis, the impact of charging PHEVs with PV cells is analyzed using a complete dynamic load model.
To date, the effects of charging a PHEV battery using a PV cell while taking into account its dynamic load characteristics have not been studied. This thesis considers the dynamics of a vehicle’s charging system and dynamic PHEV model and analyze the stability of PV cell systems with PHEVs as dynamic loads. Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 68
3.5.1 PV generator
A current source anti-parallel to a diode is the simplest representation of an electrical equivalent circuit for a solar cell as shown in Figure 3.10. The Kirchhoff’s law gives
i
L
+
IL v C ION _
Figure 3.10. PV equivalent circuit
[ ] di I − I {exp α(v + L ) − 1} − i = 0 (3.53) L s dt
where α=q/nsKT , q = 1.6022 × 10−19C is the charge of the electron, K =
1.3807 × 10−23 J/K the Boltzman’s constant, T = 298K is the temperature and ns the number of series cells in the array. However, the solar PV cell’s inputs are the solar radiance [W/m2], temperature [0C] and PV voltage [V] while the only output is the PV current supplied by the cell [A]. Therefore the output current can be characterized by I = f(V ). Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 69
To utilize the maximum amount of energy from a PV cell, it is important to track the MPP, which varies with changing atmospheric conditions. Generally, the maximum power output occurs around the knee point of the P-V curve as shown in
Figure 3.13. In this system, a DC-DC converter is used with an intelligent algorithm between the PV array and load as a MMP tracker (MPPT), which ensures the operation of the PV at its MPP. The perturb and observe method (P & O) [188], [189] and Incremental Conductance method [189] are two proven approaches for tracking the MPP. In this work, the P & O method is chosen to obtain MPP as shown in
Figure 3.14, and implemented in PSCAD due to its simplicity and low computational demand [189].
PV array Boost Converter DC Link PHEV Load C1 Lpv ,Rpv i R0 R2 ipv l idc I Im R1
Ip
Cpv Vpv Vm Cdc Vdc E m
Figure 3.11. General diagram of PV system with PHEV load
As the PV power varies with climatic conditions, there is no explicit reference power for tuning. The PV voltage needs to be adjusted according to the solar radiation to extract the maximum PV current. With a regulation of the generator Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 70
voltage, Vpv, and inductor current, Il, and by varying the transistor’s cyclic ratio, this adjustment is possible. The regulator measures the PV voltage and current using an intelligent algorithm between the PV array and load as a MMPT, which
∗ ensures the operation of the PV cell at its MPP, therefore the adequate voltage, Vpv , which the boost converter imposes on the system is found. The reference voltage is determined by calculations of the two adequate controllers and two compensators as shown in Figure 3.12.
I1 Lpv i + - dc I
+ PV CPV m Array V I Vdc _ pv pv Vm
Comparator Vpv +- PI + - * Vpv MPPT
Figure 3.12. PV converter controller system
The voltage and current in the capacitor, Cpv, and inductance, Lpv, respectively give the optimal command of the current and voltage. The voltage control loop with
∗ the PV current compensation gives the current reference Il , whereas the current
∗ control loop with the PV voltage compensation gives the voltage reference Vpv . The controller parameters are chosen to maintain a constant PV voltage and minimize the current ripple. Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 71
Figure 3.13. P-V characteristic curve of PV cell with maximum power operating point (x-axis PV array voltage and y-axis power)
Figure 3.14. Flowchart of P and O method Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 72
3.5.2 Mathematical model of solar generator
From Figure 3.11, a mathematical model describing the boost converter connected
PV generator can be written as [190], [191]:
V V m dc = m (3.54) idc i1
and the dynamics of the PV system are:
1 Rpv i˙1 = (Vm − Vpv) − i1 (3.55) Lpv Lpv
˙ 1 Vpv = (i1 − ipv) (3.56) Cpv
while the charger dynamic equation is:
˙ 1 Vdc = (idc − I) (3.57) Cdc where
Cdc= capacitance of charger capacitor; idc=output current of regulator; and
I = input current of battery. Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 73
3.5.3 Linearization of the system model
The linearized forms of the equations (3.55)-(3.57) at MPP can be written as :
m 1 Rpv ∆i˙1 = ∆Vdc − ∆Vpv − ∆i1 (3.58) Lpv Lpv Lpv ˙ 1 ∆Vpv = ∆i1 (3.59) Cpv ˙ m 1 ∆Vdc = ∆i1 − ∆I (3.60) Cdc Cdc (3.61)
Equation (3.52) is used to represent the state space equation of the linearized model of the system, where A is the state matrix and
∆X = [i1,Vpv,Vdc,I,Qe, Θ].
3.5.4 Simulation results for PV system with PHEV load
To simulate the performance of the stand-alone PV system with a dynamic PHEV load, a PV array with 15 strings characterized by a rated current of 2 A is used.
Each string is subdivided into 15 modules characterized by a rated voltage of 8 V and connected in series. Thus, the total output voltage of the PV array is 120 V, and its output current 30 A. The value of the DC-link capacitor is 250µF, the line resistance 0.1Ω and inductance 2 µH. At this stage, the system is simulated under standard atmospheric conditions for which the values of solar irradiation and tem- perature are considered to be 1 kW m−2 and 298K respectively. Then a PHEV load Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 74 is added to the PV cell and its output voltage and current are shown in Figures 3.15 and 3.16 respectively, in which, evidently, there are some fluctuations due to the nonlinear characteristics of the PV system and the dynamic behavior of PHEV. To compare the effect of constant and dynamic PHEV loads on the PV cell, simulation is performed under constant radiation and temperature, the results from which ver- ify those obtained from the eigenvalue analysis, as shown in Figures 3.17 and 3.18.
Due to changes in atmospheric conditions, the output voltage, current, and power of the PV unit change significantly, for example, if a single module of a series string is partially shaded, its output current will be reduced and dictates the operating point of the whole string. Therefore variable radiations and temperatures are used to compare the PV cell’s performance under PHEV and constant loads with step variations in radiation from 1000 to 1300 at time, t=2.5 sec and back to the pre- vious condition at t=3.5 sec. This shows that the PHEV load affects the PV cell’s performance in terms of both voltage and current magnitudes as well as frequency oscillations, as shown in Figures 3.19, 3.20, 3.22 and 3.21.
140
120
100
80
60 Voltage (V) Voltage 40
20
0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time (Sec)
Figure 3.15. PV cell’s performance with PHEVs load under constant radiation and temperature ( y-axis PV array voltage (V) and x-axis time) Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 75
30.6
30.4
30.2
30 Current (A) Current 29.8
29.6 2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4 Time (Sec)
Figure 3.16. PV cell’s performance with PHEVs load under constant radiation and temperature (y-axis PV array current (A) and x-axis time)
140
120
100
80
60 Voltage (V) Voltage 40
20
0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Time (Sec)
Figure 3.17. PV cell’s performance with constant load under constant radiation and temperature (y-axis PV array voltage (V) and x-axis time)
30.6
30.4
30.2
30 Current (A) Current 29.8
29.6 2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4 Time (Sec)
Figure 3.18. PV cell’s performance with constant load under constant radiation and temperature (y-axis PV array current (A) and x-axis time) Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 76
140
130
120 Voltage (V) Voltage 110
100 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4 Time (Sec)
Figure 3.19. PV cell’s performance with PHEVs load under variable radiation and temperature (y-axis PV array voltage (V) and x-axis time)
30.1
30.05
30 Current (A) Current 29.95
29.9 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4 Time (Sec)
Figure 3.20. PV cell’s performance with PHEVs load under variable radiation and temperature (y-axis PV array current (A) and x-axis time)
140
130
120 Voltage (v) Voltage 110
100 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4 Time (Sec)
Figure 3.21. PV cell’s performance with constant load under constant radiation and temperature (y-axis PV array voltage (V) and x-axis time) Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 77
30.1
30.05
30 Current (A) Current 29.95
29.9 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4 Time (Sec)
Figure 3.22. PV cell’s performance with constant load under constant radiation and temperature (y-axis PV array current (A) and x-axis time)
3.5.5 Small-signal stability analysis
From the MPPT, the MPP can be determined as the operating point of the system, using which the system’s A matrix can be calculated using its parameters.
Table 3.3. Eigenvalues with PHEV load −2.0162 + 24.0642i −2.0162 − 24.0642i −1.1764 −0.4741 + 10.0083i −0.4741 − 10.0083i −2.5300
To predict the system’s response, we conduct a small-signal stability analysis, gives the eigenvalues shown in Table 3.3, which shows that the system has some low frequency oscillations in modes 1, 2, 4 and 5. The participation factors for mode
4, are ∆i1 = 0.492, ∆Vpv = 0.202, ∆Vdc = 0.561, ∆I1 = 0.205, ∆Qe = 0.702,
∆θ = 0.0236 which show that the states, ∆i1, Vdc, and ∆Qe have the highest participation in system oscillations. Section 3.5 Impact on Stand Alone Dynamic Photovoltaic (PV) System 78
Using a constant load for the same PV system, another set of eigenvalues shown in Table 3.4 is found. The parameters used for these analyses are given in Table 3.5.
Table 3.4. Eigenvalues with constant load −0.3026 + 6.8567i −0.3026 − 6.8567i −0.1862
From the modal analysis, both complex and real eigenvalues are found for the
PHEV load. Complex eigenvalues indicates that the system is oscillatory. The more accurate nonlinear simulation results validate those obtained from the small signal analysis. For a constant load of 12 kW equivalent to the PHEV load, the system still has low-frequency oscillations but they are lower in frequency as are the magnitude of its voltage and current.
Table 3.5. PV system’s parameters Cell in series: 15, parallel: 1
Module in series: 15
Short circuit generator: 30A
Open circuit voltage: 120V
Filter and grid impedance: R = 0.282 Ω, L = 0.003 H
−3 Boost converter: RPV = 0.1 Ω, LPV = 2µH,CPV = 0.5e F
DC link capacitance: Cdc = 250µF Section 3.6 Impact of Dynamic PHEV Load on Distribution System 79
3.6 Impact of Dynamic PHEV Load on Distribution System
PHEVs offer the opportunity to store wind and solar energies at times of the excess generation of power from renewable sources. In a distribution network, they have the potential to modify our consideration of not only how we drive but also the generation and use of electricity. Wind and solar are the cleanest power generating technologies and, as they have with little impact on the environment, can consid- erably reduce greenhouse gas emissions and decrease our dependence on imported sources of energy and fossil fuels. As a result, solar and wind energies can be effective drivers of economic development.
The existing literatures is focused mainly on the small signal stability of a large transmission system with large generators regulated to support transmission network
[192], [193] whereas distribution systems utilize renewable energy sources using small rated distributed generators. The sporadic nature of renewable energy sources makes it difficult to regulate generators to maintain the stability of a distribution system, which may pose an enormous threat when charging PHEVs in such a system and may need additional concern for stability of the network.
Therefore it is important to identify the effect of PHEVs, for the stability and control of a distribution system. In this section, the impacts of charging PHEVs with renewable energy sources are analyzed. Section 3.6 Impact of Dynamic PHEV Load on Distribution System 80
3.6.1 Distribution system design
The effect of integrating dynamic PHEV loads and renewable energy generation is illustrated using the distribution system as shown in Figure 3.23 based on the data presented in Table 3.6. It is a modified version of the distribution system presented in [194], [195] and, in it, three radial feeders are connected. This system is convenient for studying the dynamic interactions of various generating units located on different feeders and 5.92 MW real and 3.83 MVAR reactive loads are used as its total load.
The concept of the π model is used to design the distribution network which is similar to that of a transmission system. A synchronous generator is connected at
Bus 3 which supplies 1 MW and has a reactive power limit of 0.75 MVAR. At Bus 1, the system is connected to the grid substation, at Bus 2 a 1 MW wind generator and at Bus 12, 500 kW solar PV generator assumed as a constant active power source operating at unity power factor. To determine the impact, PHEV load is connected at Bus 4 as shown in Figure 3.25, while a shunt capacitor is used to compensate the reactive power for the wind generator [196].
3.6.2 Wind generator model
The components of a constant speed wind generator are wind turbine, drive train, and generator. The rotor with a speed of ωm and radius R, converts energy to the rotor shaft. The amount of power from the wind depends on some factors, such as the wind speed, Vw, the air density, ρ, and the swept area, Awt. The available power on the rotor, is determined using the power coefficient cp(λ, θ), which depends on Section 3.6 Impact of Dynamic PHEV Load on Distribution System 81
16 15 3 13
14 Synchronous Generator
10
9 12 1 8
Maingrid Solar Generator
5 11
DG
7 4 2 6 Wind Generator Wind
Figure 3.23. Single line diagram of test distribution system Section 3.6 Impact of Dynamic PHEV Load on Distribution System 82
Table 3.6. Distribution system data The network and bus data are given in Table X. Per unit value are based on 10 MVA, 6.6 KV Bus to Section Section End Bus End Bus Bus Resistance Reactance Load Load (pu) (pu) (MW) (MVAR) 2-4 0.075 0.100 0.500 0.400 4-5 0.080 0.110 0.750 0.375 4-6 0.090 0.180 0.500 0.200 6-7 0.040 0.040 0.375 0.300 5-11 0.040 0.040 0.000 0.000 1-8 0.110 0.110 1.000 0.675 8-9 0.080 0.110 1.250 0.750 9-11 0.110 0.110 0.150 0.025 9-12 0.080 0.110 0.000 0.000 8-10 0.110 0.110 0.25 0.225 10-14 0.040 0.040 0.000 0.000 3-13 0.110 0.110 0.250 0.225 13-14 0.090 0.120 0.250 0.275 13-15 0.080 0.110 0.250 0.275 15-16 0.040 0.040 0.525 0.250 the blade pitch angle, θ, and the ratio between the wind speed and the speed of the blade tip, called tip-speed ratio, λ = ωmR . The aerodynamic torque applied to the Vw rotor turbine can be express as [197]:
ρ 3 Tae = Awtcp(λ, θ)Vw (3.62) 2ωm
where cp is calculated by the following equation [198]:
[ ] π(λ − 3) c = (0.44 − 0.0167θ) sin − 0.00184(λ − 3)θ p 15 − 0.3θ
The dynamic characteristics of a wind turbine generator system is reproduced Section 3.6 Impact of Dynamic PHEV Load on Distribution System 83 using a two-mass drive train model in this section. The high-speed shaft is connected with the low speed shaft through a gear box, at the same time low speed shaft is attached with the drive train through wind turbine converter aerodynamic torque
Tae. The first mass term can be expressed with the blades, hub and low-speed shaft, and the second mass term with the high-speed shaft, with inertia constants Hm and
HG. A gear ratio, Ng, interconnected the shafts combined with torsion stiffness, Ks, and torsion damping, Dm and DG, which gives the torsion angle γ. With the grid frequency f. The dynamics of the shaft are represented as [197]:
1 ω˙ m = [Tae − Ksγ − Dmωm] (3.63) 2Hm
1 ω˙ G = [Ksγ − Te − DGωG] (3.64) 2HG
1 γ˙ = 2πf(ωm − ωG) (3.65) Ng
Gear box gives the mechanical power to the generator using stiff shaft. The rela- tionship between the mechanical torque and the torsional angle is given by:
Tm = Ksγ (3.66) Section 3.6 Impact of Dynamic PHEV Load on Distribution System 84
The following algebraic-differential equations are used to describe the transient model of a single cage induction generator [198], [199]:
1 s˙ = [Tm − Te] (3.67) 2HG [ ] ˙ ′ − 1 ′ − − ′ − ′ Eqr = ′ Eqr (X X )ids siωsEdr (3.68) To ˙ ′ − 1 ′ − ′ ′ Edr = ′ [Edr + (X X )iqs] + siωsEqr (3.69) To − ′ ′ Vds = Rsids X iqs + Edr (3.70)
′ ′ Vqs = Rsiqs + X ids + Eqr (3.71) √ 2 2 vt = Vds + Vqs (3.72)
′ where X = Xs + XmXr/(Xm + Xr), is the transient reactance, X = Xs + Xm,
′ the rotor open-circuit reactance, To = (Lr + Lm)/Rr, the transient open-circuit
′ time constant, vt the terminal voltage of the IG, s the slip, Edr the direct-axis
′ transient voltages, Eqr the quadrature-axis transient voltages, Vds the d-axis stator voltage, Vqs the q-axis stator voltage, Tm the mechanical torque, Te = Edrids +
Eqriqs, the electrical torque, Xs the stator reactance, Xr the rotor reactance, Xm the magnetizing reactance, Rs the stator resistance and Rr the rotor resistance. HG is the inertia constant of the IG, ωG the rotor speed of the IG, ωs the synchronous speed, and ids and iqs the d and q axis components of the stator current, respectively.
By axis transformation, the dynamic element of induction generator can be written in a synchronous rotating frame using the following relation: Section 3.6 Impact of Dynamic PHEV Load on Distribution System 85
√ ′ ′2 ′2 Eg = Eqr + Edr ( ) − ′ −1 Edr δg = tan ′ Eqr
The modified third-order induction generator model can be written as follows:
( ) 1 ′ − s˙ = EgIqs Tm (3.73) 2HG [ ( ) ] ˙ ′ − 1 ′ − ′ Eg = ′ Eg + X X Ids (3.74) T0 − ′ ˙ X X δg = ωsωr + ′ ′ Iqs (3.75) TdoEg
The parameters used for the induction generator are given in Table 3.7
Table 3.7. Parameters used for induction generator
Rs = 0.012pu, Xs = 0.074pu, Xm = 2.76pu, Rr = 0.008pu, Xr = 0.1761pu, Hm = 2.5s, HG = 0.22s 1
3.6.3 PHEV’s interface with network
The connection with the network is assumed to be realized by means of an ideal converter and a transformer with reactance, xT , as depicted in Figure 3.24. The DC voltage is regulated by means of the converter’s modulating amplitude m, as [200]:
√ ( ) 2 2 xT 2 Vs m = ps + qs + (3.76) VskVdc xT
As the DC power of the battery (Pdc=VdcI ) is considered the real power in the network ( ps=Pdc), the link with AC network is: Section 3.6 Impact of Dynamic PHEV Load on Distribution System 86
VtVs ps = − cos(θs − θt) = VdcI (3.77) xT √ 2 Vs VskmVdc xT I 2 qs = − 1 − ( ) (3.78) xT xT kmVs
√ where Vt=kmVdc, and the rectifier gain k= (3/8).
C1 R0 R2 I Im R1
ps+jqs Ip
Vdc Rp Vp E m
1:m θ Vs - s Vt - θt
Figure 3.24. PHEVs connection with power system
The power balance equations for the load bus are :
∑n − − PLs (Vs) + VsVt[Gst cos(θs θt) + Bst sin(θs θt)] = 0 (3.79) t=1 ∑n − − − QLs (Vs) + VsVt[Gst sin(θs θt) Bst cos(θs θt)] = 0 (3.80) t=1
where n is the total number of buses in the system and Yst=Gst+jBst the element of the sth row and tth column of the bus admittance matrix Y. Section 3.6 Impact of Dynamic PHEV Load on Distribution System 87
16 15 3 13
14 Synchronous Generator
10
9 12 1 8
Maingrid Solar Generator
5 11
DG
7 4 2 6 Wind Generator Wind
Figure 3.25. PHEVs load connection with power system network Section 3.6 Impact of Dynamic PHEV Load on Distribution System 88
3.6.4 Stability analysis of distribution system
The following equation is used to represent the state space equation of the linearized model of the system.
∆X˙ = A∆X
′ ′ where ∆X = [δ, ω, Eq,V0, ωm, ωG, γ, s, Eg, δg,I,Qe, Θ] are the states of the system. The eigenvalues of A provide information of small signal stability.
Through a load flow analysis, the system’s operating point can be determined and then its state matrix of the system can be calculated using power system parameters.
Equations (3.37)-(3.40), (3.63)-(3.65), (3.73)-(3.75) and (3.34)-(3.36) are linearized at the equilibrium point and the eigenvalues of the system under study are shown in Figure 3.26.
25
20
15
10
5
0
−5
−10
−15
−20
−25 −18 −16 −14 −12 −10 −8 −6 −4 −2 0
Figure 3.26. Eigenvalues of the distribution system with PHEV load
From the small signal analysis, both complex and real eigenvalues are found for the PHEV load. The system can be identified as oscillatory as it has complex Section 3.6 Impact of Dynamic PHEV Load on Distribution System 89 eigenvalues. To compare the effects of the constant and dynamic PHEV loads, a simulation in which a single-line to ground fault is applied on the connecting line of the dynamic load and cleared after 0.05 sec is performed. As the nonlinear simulation results also show oscillations, they validate those results obtained from the small signal analysis. For comparison a constant load of 12 kW is used to represent an equal dynamic PHEV load and the PHEV load shows higher voltage and angle oscillations than the constant load, as shown in Figure 3.27 and 3.28.
0.9
0.8 With Constant load 0.7 With PHEV load 0.6
Voltage (PU) Voltage 0.5 0.4
0.3 0 2 4 6 8 10 12 14 16 18 20 Time (s)
Figure 3.27. Wind generator voltage oscillation with constant and PHEV load
−0.4 PHEV Load −0.6 Constant Load
−0.8
−1 Angle Angle (rad) −1.2
−1.4 0 2 4 6 8 10 12 14 16 18 20 Time (s)
Figure 3.28. Wind generator angle oscillation with constant and PHEV load Section 3.7 Chapter Summary 90
3.7 Chapter Summary
In this chapter, the impact of dynamic PHEV loads on a SMIB system is investigated through small-signal and time-domain analyses. From the modal analysis, both complex and real eigenvalues are found. The complex eigenvalues clearly indicate that the system is oscillatory while the more accurate nonlinear simulation results validate those obtained from the small-signal analysis. Goals achieved via this study are investigations into the performances of a PV cell with a dynamic PHEV load under both constant and variable radiations and temperatures, and renewable energy in a distribution system with dynamic PHEV loads. Also, for wind generation, voltage and angle variations due to these loads are examined. The results obtained from both modal analysis at a MPP and nonlinear simulation results show that the system becomes more oscillatory with a dynamic PHEV load. It is concluded that a great deal of research into the charging of a dynamic PHEV load through a PV cell and the integration of renewable sources in a distribution system is needed. Several issues, such as the use of more accurate dynamics of the MPPT, interconnections with the grid and the consequences of battery aging for charging with a PV cell, as well as consideration of the solar power dynamics and controller design for charging with renewable sources in a distribution network, could be interesting topics for future work. Chapter 4
Design of Virtual Active Filter for Power System using PHEVs
In this chapter, V2G technology is used to design a virtual active filter, which provides the potential for a low-cost solution of active filter for power systems that utilize the reactive power and filtering capabilities of PHEVs parked in charging stations. Simulations are performed for the CIGRE benchmark HVDC network to demonstrate that the proposed virtual active filter improves the power quality while maintaining IEEE Std 519-1992.
Using V2G technology, a shunt active filter is designed for an induction generator- based wind farm to improve its power quality. Simulation results show that PHEVs have the potential to work as active filters with wind generators to improve power quality, dynamic power factor correction and harmonics current compensation.
91 Section 4.1 Introduction 92
4.1 Introduction
In recent years, there have been significant advances in battery and hybrid-electric power technologies, which coupled with energy security requirements, financial and environmental concerns and the rising costs of petroleum, make plug-in hybrid elec- tric vehicles (PHEVs) a strong alternative to conventional vehicles [161], [201]. It is envisioned that most vehicles manufactured in the future will have a plug-in op- tion for recharging their batteries and, by the year 2030, PHEV penetration will be almost 25% [172]. PHEVs can be charged from various locations, such as household electrical connections, charging stations and car parks during the day. Various re- searchers have attempted to determine the optimal placement for a PHEV’s charging station and parking [202–205]. Also, a PHEV provides the opportunity to design its bidirectional charger as an active filter for power systems.
In V2G mode of operation, PHEVs can serve grids as regulators, spinning re- serves, storage for renewable energy sources and reactive power compensators to provide power quality improvement.
Ideas for using a PHEV charging station as a spinning reserve [206], for load leveling [42], and as an external storage for renewable energy [18], have been studied while using a PHEV to improve wind power quality has been reported in [113].
Another study has demonstrated that a PHEV battery can serve as a STATCOM
[43] while effective pricing models of electricity buying and selling from a PHEV using variable price curves have been reported in [41], [207]. Section 4.2 Power Quality 93
As the harmonics in a power system cause power loss and, sometimes, oper- ational failures of electronic equipments, their reduction is essential for installed equipments [208], [209]. This is achieved by capacitor-based active filters, which are a combination of electronics converters, capacitors and switching controllers [210].
The most expensive components for the constitution of FACTS and filter devices are the capacitors [211]. In this chapter, the capacitors are replaced by a PHEV’s parking station with the PHEV’s bidirectional charger used as the converter. The p-q theory [208] is used to design a controller and investigate the possibility of em- ploying PHEVs as virtual active filter in a HVDC network and wind generator to achieve low-cost filters.
Two case studies for HVDC network have been introduced to verify the virtual active filter performance :
• Case 1: Virtual active filter for the rectifier side of the HVDC link; and
• Case 2: Virtual active filter for both rectifier and inverter side of the HVDC
link.
The virtual filter performance is also verified for a wind farm.
4.2 Power Quality
Power quality is a term that describes a set of parameters of electrical power and a load’s ability to function properly with that power. Poor electrical power quality may cause overloading of the network, and neutral wire, unsafe resonance phenomena or even damage to the load, which usually lead to high costs in countries dynamically Section 4.3 Power System Harmonics 94 developing new technologies. It is estimated that issues related to power quality cost industries hundreds of billions of dollars annually of which the finance provided to prevent them is only a small percentage. Therefore, research into methods for analyzing and improving power quality is being extensively conducted throughout the world.
4.3 Power System Harmonics
Harmonic distortions indicate the presence of frequencies at integer multiples of a power system’s fundamental frequency and occur whenever non-sinusoidal currents and/or voltages are generated in the system. Generally, it is safe to assume that the sine wave voltage generated in a central power station is purely sinusoidal and, in most areas, that found in a typical transmission system has much less than a 1% distortion. However, this may reach 5-8% closer to the load and, at some loads, the current wave forms barely resemble sine waves. Solutions to problems caused by harmonic distortion include installing active or passive filters at the load or bus, or taking advantage of transformer connections that enable cancelation of zero-sequence components.
The basic conditions that give rise to harmonic-related problems in power sys- tems are, in brief [212], [213]:
(a) Nonlinear loads;
(b) Phase imbalance;
(c) High input voltage or current; and Section 4.4 Filters in Power Systems 95
(d) Resonance
Harmonic-related problems caused by the widespread use of large-capacity nonlin- ear loads, such as rectifiers, inverters, and cyclo-converters in industries [214] and on an individual basis, and lower-capacity ones in modern office automation equip- ment [215], have led to the relatively recent creation of a new area in the power electronics field, power quality, which is now a major topic in terms of electrical power generation and distribution, and its users. As the use of such types of loads is continually increasing [216], power quality concerns already play a important task in electrical energy scenarios, imposing harmonic power consumption constraints on the implementation of new power electronic technologies. This chapter presents the issues related to methods for improving power quality, in particular, using an active
filter to reduce the harmonics in a power system using V2G technology. An EMTDC transient simulator PSCAD is used to identify the harmonics level in various case studies, which is a fast, accurate and easy to use power system simulator for design and verification of all type of power systems and power electronics control.
4.4 Filters in Power Systems
The use of an active filter in a power system was introduced to solve the prob- lems of passive filters, which tend to overload and require a considerable amount of space and expense. An active filter is a device that compensates harmonic currents in a utility AC system with the same magnitudes as, but opposite phases to, the Section 4.5 V2G Technology for Filter Design 96 harmonic currents generated by a given non-linear load. For several years, the avail- able ratings of power semiconductor switches have permitted their use in industrial applications (low and medium power active filtering). As a result, pulse-width- modulated (PWM) amplifiers with continually increasing power capabilities have become available during recent years. However, the cost of the filter is still of great concern for power system engineers. For this reason, new ideas for the application of active filtering to power systems have recently been proposed in both universities and industry. This has been motivated by the existence of problems related to the design and use of passive and active filters, together with the increasingly stringent requirements of power utilities to limit interference caused by harmonic currents.
Some of these ideas are to use:
1. magnetic-flux compensation [217], [218];
2. a current-source PWM inverter as an active source [219];
3. a capacitor-commutated inverter with PLL-generated reference signal [220];
4. a PWM CR-VSI and economic feasibility [221];
5. a combination of a series active and a shunt passive filters [222]
6. a combination an active and passive filter in a series [223]; and
7. a combination of an active filter in a series with shunt-tuned filters [224]
4.5 V2G Technology for Filter Design
V2G describes a system in which PEVs communicate with a power grid to sell demand services by delivering electricity into the grid or throttling their charging Section 4.6 PHEV Battery Modeling for Filter Design 97 rate. In this chapter, an ancillary service of V2G technology is introduced, to do which a well-designed PHEV park is essential for connecting a network and supporting it when necessary.
4.6 PHEV Battery Modeling for Filter Design
PHEVs are able to compensate reactive power to a utility grid, as reported in [43]. In this work,they are considered to be bidirectional converters connected to dynamic batteries [225] designed with rated currents of 70 A (level 2 charger) and power capabilities within the range of ±20 kW active and ±20 kVA reactive. For HVDC link virtual filters are considered for a ±20 MW power transaction with the grid, that is, a park with around 1000 vehicles, a reasonable assumption for city car parks and for wind farm the size of a PHEV park will be determined depending upon the number and capacity of wind generators in that particular farm. Here, the “+” sign means that the PHEV is in the V2G mode of operation and the “-” sign the G2V mode. The P-Q capability of a realistic PHEV battery varies between ± 138 kW and ± 126 kVA respectively [226], as shown in Figure 4.1 [43], [227] . Section 4.6 PHEV Battery Modeling for Filter Design 98
Figure 4.1. P-Q capability of PHEV’s battery
A dynamic model of a rechargeable battery [178], [228] is used to implement
V2G technology using PHEV where battery operation depends on electrolyte tem- perature, θ, and SOC, and im , an integral part of the total current, idc.
This battery model can be represented as an RLC network connected to the grid, as shown in Figure 4.2 and the number of RLC blocks can be limited because, for PHEVs, the specific speeds of evolution of their electrical quantities change rapidly [178].
C1
R2 R0 im P Idc R1 L2 R12 iBa I1 Ip va
L2 R12 iBb E m
vb
L2 R12 iBc N vc
Figure 4.2. PHEV’s connection to power system network Section 4.6 PHEV Battery Modeling for Filter Design 99
The third-order battery dynamic model is designed considering the current, elec- trolytic temperature and SOC, and its dynamic equations are [178], [179], [200],
[228]:
q˙e = idc/Ts (4.1)
i˙m = (idc − im)/Tm (4.2) − ˙ 1 θ Qa θ = − [Ps − ] (4.3) Cθ Rθ
−Beqe Vdc = Em − Vp(qe, im) + Vee − R0idc (4.4)
where Ve, represents the hysteresis phenomenon for the battery during both its charge and discharge cycles. The voltage, Vdc, increases when the battery is charg- ing and decreases when it is discharging, while the polarization voltage, Vp, depends on the sign of im as:
Rpim+Kpqe SOC if im > 0 (discharge) Vp(qe, im) = Rpim + Kpqe if i < 0 (charge) qe+0.1 SOC m
The equations for Em, R0, R1 and, R2 are:
Em = Em0 − Ke(273 + θ)(1 − SOC) (4.5) Section 4.6 PHEV Battery Modeling for Filter Design 100
R0 = R00[1 + A0(1 − SOC)] (4.6)
R1 = −R10ln(DOC) (4.7)
exp[A21(1 − SOC)] R2 = R20 ∗ (4.8) 1 + exp(A22Im/I )
The SOC and depth of charge (DOC) can be expressed as:
Qn − Qe SOC = = 1 − qe (4.9) Qn
DOC = 1 − Qe/C(Iavg, Θ) (4.10) where:
Cθ and Ps are the battery’s thermal capacity and power respectively;
R0 and Rp the thermal and polarization resistance;
Qa the ambient temperature;
I⋆ the reference current (I⋆ is a current, flows in the battery for typical use); xr the Thevenin equivalent reactance;
βe the exponential capacity coefficient;
Qe the extracted capacity in Ah;
Qn the rated battery capacity in Ah; and
K c,Em,Ke,Kp , A0, A21 and A22 are constant for a particular battery. These battery parameters are available in Appendix 8.1. Section 4.6 PHEV Battery Modeling for Filter Design 101
The behavior of the parasitic branch is strongly nonlinear and its current is given as: ( ) Vp θ Ip = VpGp exp + Ap(1 − ) (4.11) Vpo θf
The heat produced by the parasitic reaction can be calculated by means of the Joule
Law, as:
2 Ps = RpIp (4.12)
where, θf , is the electrolyte freezing temperature and Vpo, Gp and, Ap are constant.
The dynamic PHEV battery modeling in PSCAD is given in Appendix 8.3.
4.6.1 Bidirectional charger
In general, there are two different converters in a bidirectional charger. The first is connected to the grid and converts AC to DC while the second regulates the charging and discharging currents with its basic operation being to convert DC to
DC, as shown in Figure 4.3 [229].
Bidirectional Bidirectional Filter AC/DC DC/DC Converter Converter
Grid
Figure 4.3. Topology of unidirectional and bidirectional power flow
Although most studies suggest using a bidirectional charger for a PHEV, there are a number of challenges to implementing it [73], such as its frequent charge and discharge cycling regulation degrading the battery, its additional cost and its Section 4.6 PHEV Battery Modeling for Filter Design 102 interfacing and smart metering issues. Maintaining power quality is also a great concern of the V2G mode of operation because, when a PHEV is in the charging mode, it draws a sinusoidal current with a particular phase angle and reactive power and, when discharging, has to provide a similar sinusoidal current form [95], [230–
233].
To find a suitable and reliable bidirectional charger, research is continuing. Al- though the present PHEV market does not have bidirectional PHEV chargers [234], it is expected that, following their introduction, customers and utility engineers will be a part of a smart grid. A successful bidirectional charger will require extensive safety measures [235] and a guarantee that, when a vehicle is driven, its SOC will be predictable and high [62]. This chapter includes a simple model of a bidirectional charger system used for the V2G technology analysis undertaken in the rest of this thesis. V2G technology depends basically on the performance of the battery charger in order to maintain the battery’s charging condition and health in terms of its SOC.
It needs to have the capability to work as a universal converter and accept differ- ent voltage and power levels, as well as the additional characteristic of preventing overcharging [236]. Consequently, it should protect the battery from over-current, over-voltage, under-voltage and a too high temperature [237]. A PHEV battery can be charged by either an off-board or on-board charger. The later should be opti- mized to accept different charging levels as well as match different vehicle battery ʌ
Section 4.6 PHEV Battery Modeling for Filter Design 103
requirements and have the capability to use any outlet available [238]. The avail-
ability of such charging places will increase the acceptance of PHEV technology. On
the other hand, an off-board charger can use a method of fast charging whereby it
can charge a battery in 10 minutes to increase its SOC by 50% with a rating of 240
kW [106]. Nissan has claimed that their PHEV model ‘Nissan Leaf’, which has been
available since 2010, can be charged up to an 80% SOC of its 24 kWh within 30
minutes at a quick-charge station [23]. with several assumptions that will make the computation much easier. During the analysis, the positive current direction will be assumed to be from grid to the inverter as shown in Fig. 1. Therefore, positive power sign (P = active power and Q = reactive power) corresponds to the power flow from grid
Fig. 1. Representation of grid and charger. Figure 4.4. Topology of general charger
Although, due to the weight and space restrictions of a vehicle, an on-board
charger cannot be designed to be as fast as an off-board one, using a traction inverter
[239], it can charge up to an 80% SOC for a battery rated at 30 kWh within one
hour [240], which causes the majority of losses to be those of copper [241]. In the
following, a theoretical analysis of a conductive charger utilized in a bidirectional
power transfer is presented. The PHEV charger analyzed in this study is composed
of a full-bridge inverter/rectifier and a DC-DC converter. The analysis begins by
investigating the interaction between the grid and the inverter. To understand all the Section 4.6 PHEV Battery Modeling for Filter Design 104 dynamics, the basic ideal case is introduced with several assumptions, which make the computation much easier. During the analysis, the positive current direction is assumed to be from the grid to the inverter, as shown in Figure 4.4 [225]. Therefore, the positive power sign (P = active power and Q = reactive power) corresponds to the power flow from the grid to the inverter.
The system parameters are: vc(t) instantaneous charger voltage [V]; vs(t) instantaneous grid voltage [V]; ic(t) instantaneous charger current [A];
Lc coupling inductor [H];
δ phase difference between vc(t) and vs(t); and
θ phase difference between ic(t) and vs(t).
The grid voltage is assumed to be purely sinusoidal and the high-frequency com- ponents of the inverter output voltage vc(t) are neglected for analysis purposes, as shown by: √ √ vs(t) = 2Vssin(wt)vc(t) = 2Vcsin(wt − δ) (4.13)
To ensure a power transfer from the charger to the utility, a coupling inductor is used and the two voltage sources are decoupled. From Figure 4.4 and by applying the necessary mathematical transformations, the line current can be written as:
√ ic(t) = 2Icsin(wt − θ) (4.14) Section 4.6 PHEV Battery Modeling for Filter Design 105
Since the default direction for active and reactive power transfer is from grid to charger, ic(t) and vc(t) are lagging the grid voltage. Also, note that the reactance is equal to
Xc = 2πfLc (4.15)
Table 4.1 and the P-Q plane in Figure 4.5 [225] show all the different operational modes in which the system can work. To conserve the amount of energy drawn from the battery and keep the battery as undisturbed as possible, from Figure 4.4, it can be written that
Vs = Vc + jXcIc (4.16)
Using equation (4.16), the system variables are shown in the phasor diagrams in
Figure 4.6, to illustrate the differences between the operating modes. Section 4.6 PHEV Battery Modeling for Filter Design 106
Q
II I Discharging Charging Inductive operation Inductive operation
P
III IV Discharging Charging Capacitive operation Capacitive operation
Figure 4.5. P-Q plane showing charger operation Section 4.6 PHEV Battery Modeling for Filter Design 107
P Q Operation Mode of the Charger
a) Charging b) Discharging
c) Inductive operation d) Capacitive operation .
.
c) Charging and inductive operation d) Charging and capacitive operation
FigureFig. 3. Vector 4.6. Vector diagram diagram for different of charger operation operation modes.
Table 4.1. Bidirectional charger mode of operations # P Q Operating Mode 1 Zero Positive Inductive 2 Zero Negative Capacitive 3 Positive Zero Charging 4 Negative Zero Discharging 5 Positive Positive Charging and inductive 6 Positive Negative Charging and capacitive 7 Negative Positive Discharging and inductive 8 Negative Negative Discharging and capacitive
ș ș ș
į
ș Section 4.7 Controller Design 108
4.7 Controller Design
4.7.1 General control principle
The target of the controller design is to control the switching of the PHEV’s con- verter in such a way that the PHEV park can be used as a filter and compensate the reactive power.
This virtual active filter controller is divided into three functional control blocks:
1. instantaneous-power calculation block;
2. power compensating block; and
3. DC voltage regulator and current control block.
The first block calculates the instantaneous power of the nonlinear load while the second controls the behavior of the virtual active filter and determines the parts of the real and imaginary powers of the non-linear load that need to be compensated.
The third represents the DC voltage regulator for calculating an extra amount of real power, PL, to maintain the voltage at around a fixed reference value. Then, PL is added to the compensating real power and passed to the current reference calculation block with the compensating imaginary power. Then, this block determines the instantaneous compensating current references from the compensating powers and voltages.
4.7.2 Instantaneous power calculation
In this work, the p-q theory, which consists of an algebraic transformation (Clarke transformation) of the three-phase voltages and currents in the a-b-c coordinates Section 4.7 Controller Design 109 to αβ, is used to calculate the instantaneous power without considering the neutral wire [208]. The equations for the currents in the αβ coordinates can be expressed as [208], [242]:
i √ a i 1 − 1 − 1 α 2 2 2 = i (4.17) 3 √ √ b 3 − 3 iβ 0 2 2 ic
The voltage in αβ coordinates are:
v √ a v 1 − 1 − 1 α 2 2 2 = v (4.18) 3 √ √ b 3 − 3 vβ 0 2 2 vc
and the equation for p, q are :
p v v i α β α = (4.19) q −vβ vα iβ
The complete system with the controller is shown in Figure 4.7. In the first part of the controller, the instantaneous values of the real and imaginary powers are calculated. To generate the reference values of p and q, these powers pass through a selection block where the power needing to be compensated is calculated, as shown Section 4.7 Controller Design 110
AC L1 ias R11 ia
AC va To L1 R non linear ibs 11 ib Y/Y load AC vb L1 ics R11 ic
Transformer vc ABC To a ß DC voltage Regulator i PQ dc Generation L2 R12 iBa Reference Current R calculation L2 12 iBb Vdc Current L2 R 12 iBc control
Figure 4.7. PHEV’s connection as virtual active filter in network in Figure 4.8. Then, the error current signal used to switch the inverter is shown in
Figure 4.9.
A nonlinear load draws fundamental (or average) and harmonic (or oscillating) current components from the power system, which a shunt active filter can com- pensate [208]. The real and imaginary powers can be defined as the combination of the average and oscillating components while the undesirable oscillating real and reactive powers are produced by harmonic components in the load current. Section 4.7 Controller Design 111
Va Va Va ABC Ia Vb To p S a, ß Vß Vc Vß p_ref
Ia q_ref Vß
Ia Ia Iß q ABC S To Ib Vß a, ß Iß Iß Selection of power to be compensated Ic
Figure 4.8. pq generation in controller
Va
IBa p_ref S Iaref - Iaerr + Ia Vß 2 a ß V Ibref Iberr ß S To +- Ib 2 ABC Va IBb Ic + err q_ref Icref - S
IBc
Va
Figure 4.9. Signal generation for inverter switching
The compensating currents in the αβ reference are calculated for these oscillating powers and then the Clarke inverse transformation is used to calculate the amount Section 4.7 Controller Design 112 of current to be injected by the virtual active filter. To generate the current for the controller, equation (4.19) can be written as:
i v v p α 1 α β = 2 2 (4.20) vα + vβ iβ vβ −vα q
4.7.3 Power compensation
It is convenient to separate p and q into their average and oscillating parts as:
p = p + pe (4.21)
q = q + qe (4.22) where: p and q are the average parts of real and imaginary power respectively; and pe and qe the oscillating parts of real and imaginary power respectively.
From equation (4.20), it is possible to write
i v v p v v 0 α 1 α β 1 α β = 2 2 + 2 2 (4.23) vα + vβ vα + vβ iβ vβ −vα 0 vβ −vα q Section 4.7 Controller Design 113
i i αp αq , + (4.24) iβp iβq
where the instantaneous active and reactive currents in the αβ axis are denoted by iαp, and iβp and iαq and iβq respectively. The reactive power is defined as a com- ponent of the instantaneous power, as shown in equation (4.25). The instantaneous powers in the αβ axis are pα and pβ, which are combinations of the instantaneous active and reactive powers in the α and β axes, that is, pαp and pαq and pβp and pβq, respectively. In the p-q theory, the imaginary power is the sum of the products of the instantaneous three-phase voltage and current. In this work, q represents the imaginary power. Therefore, the instantaneous power, p, in the αβ axis can be expressed as:
p = vαiαp + vβiβp + vαiαq + vβiβq 2 2 − vα vβ vαvβ vαvβ = 2 2 p + 2 2 p + 2 2 q + 2 2 q vα + vβ vα + vβ vα + vβ vα + vβ
= pαp + pβp + pαq + pβq
= pα + pβ (4.25)
As the sum of the third and fourth terms on the right-hand side of equation
(4.25) is always zero, p is called the instantaneous reactive power. The instantaneous imaginary power, q, is a quantity that gives the magnitudes of the powers pαq and pβq. If the αβ variables of p and q defined in equation (4.19) are replaced by their Section 4.7 Controller Design 114 equivalent expressions referred to on the abc axis using equation (4.17) and similarly for the current, the following relation can be found.
p = vαiα + vβiβ 1 = [{(v − v ) − (v − v )}i 3 a b c a a
+ {(vb − vc) − (va − vb)}ib
+ {(vc − va) − (vb − vc)}ic] 1 = [(v − v )i + (v − v )i + (v − v )i ] (4.26) 3 ab ca a bc ab b ca bc c and
q = vβiα − vαiβ 1 = √ [(va − vb)ic + (vb − vc)ia + (vc − va)ib] 3 1 = √ (vabic + vbcia + vcaib) (4.27) 3
To draw a constant instantaneous power from the source, the shunt virtual active
filter should be installed as close as possible to the nonlinear load. In this case, we have to consider that the average real, p, is supplied by the grid while the power from the voltage regulator, PL, contributes to maintaining Vdc at around its reference value. In fact, a small amount of average PL, must be drawn continuously to supply the switching and ohmic losses in the converters. As the oscillating real pe, the power
PL and the total instantaneous reactive power (q = q + qe) are compensated by the Section 4.7 Controller Design 115 virtual filter, the compensating current will be: −e icomα 1 vα vβ p + PL = (4.28) v2 + v2 α β − − icomβ vβ vα q where pe can be calculated using p = pe in equations (4.23) and (4.24). The use of a PHEV park as a source of energy has the advantage of compensating real power p, which implies an oscillating flow of energy and protecting it from experiencing large voltage variations. If the amplitude of the AC voltage is higher than that of the DC voltage, the PWM controller loses its controllability. Although, in this case the rating of the DC capacitor needs to be high, using the PHEV park as storage can eliminate this problem. Finally, the αβ inverse transformation block in
Fig. 4.9, calculates the instantaneous current reference for the dynamic-hysteresis current control of the VSC.
4.7.4 DC voltage regulator and current controller
A hysteresis current controller is used in this paper because it offers excellent dy- namic performance and is simple to implement in real time [243]. A dynamic offset,
ε, is created from the measurement of Vdc and the DC reference voltage so that the band limits of this are:
Upper hysteresis band limit = i(ref) + ∆(1 + ε); and
Lower hysteresis band limit = i(ref) − ∆(1 + ε)
where iref = iaref , ibref , icref , and ∆ is a fixed half hysteresis band. Another slower feedback loop generates the signal from the voltage regulator PL, to maintain the Section 4.7 Controller Design 116 voltage at around a fixed reference point, as shown in Figure 4.10. It brings the energy balance and is also useful for compensating the error occurring during the transient state.
e e ∆V Hysteresis Vref ∆V Current controller S + Sh1 - IB Upper Limit PI a Ref. Current (Ia_ref) Sh2 Vdc IBb Lower Limit Actual Current PL IBc (IB_a) Sh3
Ia ref Sh4 Filter Ib Controller ref Sh5 Icref Sh1 PHEV Park Sh4 Sh6
Figure 4.10. Signal generation for inverter switching with hysteresis band
The switching logic for the hysteresis current controller is formulated as:
if iBa < (iaref − ∆(1 + ε)) Sh 1 on and Sh 4 off;
if iBa < (iaref + ∆(1 + ε)) Sh 1 off and Sh 4 on;
if iBb < (ibref − ∆(1 + ε)) Sh 3 on and Sh 6 off;
if iBb < (ibref + ∆(1 + ε)) Sh 3 off and Sh 6 on;
if iBc < (icref − ∆(1 + ε)) Sh 5 on and Sh 2 off; and
if iBc < (icref + ∆(1 + ε)) Sh 5 off and Sh 2 on. Section 4.8 Virtual Filter for HVDC Test System 117
4.8 Virtual Filter for HVDC Test System
The use of an active filter in a HVDC network was first demonstrated in 1993 at Skagerrak3 HVDC Intertie and then at Baltic Cable HVDC Link in 1994 and
Chandrapur-Padghe HVDC Power Transmission in 1998 [210]. Its main aim is to reduce harmonics and compensate reactive power at the same time.
The filtering capability of a single vehicle was utilized in photovoltaic and wind power systems in [111] and [112] respectively. However, these studies did not deal with the capability of a fleet of such vehicles parked in a charging station or their usefulness in large numbers in an HVDC transmission network. This section presents a way of analyzing the filtering and reactive power transaction capabilities of PHEV
fleet in the V2G mode of operation.
A simple HVDC network structure can be described as a transmission system with a set of rectifiers and inverters for interfacing a DC line with an AC network.
A future smart grid composed of an HVDC system has been anticipated since early nineties [244]. However, as the drawback of an HVDC network has been the high cost of its power electronics based converters and inverters control, huge efforts have been made over the last few years to reduce costs and improve HVDC technology
[245–247]. As a result, several American states, some European and Asian coun- tries and Australia currently have HVDC power networks. A HVDC network has some important niche applications compared with other systems or devices, such as transmitting electrical power over long distances and in submarine connections. Section 4.8 Virtual Filter for HVDC Test System 118
An HVDC link converter behaves like a non-linear load [248] and the harmonics created from the non-linear operation of converters on both the AC and DC sides of the link can be identified as characteristic or non-characteristic. Under ideal conditions, characteristic harmonics are related to the pulse number of the converter but as, in a real system, ideal conditions are not achievable and non-characteristic harmonics are usually present in an HVDC network.
The CIGRE benchmark model [249], [250], [251], is taken as our base system to address harmonic problems, which exhibit complex operational characteristics.
The system shown in Figure 4.11, where the T-section represents the DC line. The control model at the rectifier is the constant current control and at the inverter the constant extinction angle (γ) control. The converters (rectifier and inverter) are modeled using a 6-pulse Graetz bridge block, which consists of an internal phase- locked oscillator (PLO), firing angle measurements and firing and valve blocking controls while each thyristor has a built-in RC snubber circuit. Both the inverter and converter sides have similar models of the converter transformer, which is a com- bination of a three-phase two-winding transformer, one with a grounded Wye-Wye connection and the other a grounded Wye-Delta connection. Saturation character- istics are used in the model with a tap-setting arrangement and smoothing reactors are inserted on both sides with an equivalent T-network to model the DC line.
Three-phase AC voltage sources are used to represent the supply voltages on both the converter’s sides while tuned filters and reactive power supports are provided on Section 4.9 HVDC Case Study 119 the AC sides of the rectifier and inverter, as shown in Figure 4.11 [250].
0.151 H 0.5968H 0.5968H AC 3.737 Ω 0.0365H0.7406 Ω 0.0365H AC 25 Ω 25 Ω 3.342µF 422.84kV : 230 kV 7.522µF 2160.633 345kV : 422.84 kV Ω 1196MVA 1196MVA 24.81 Ω 0.1364 H 0.0606H 29.76 Ω 6.685µF 26µF 15.04µF 13.23 Ω
74.28µF 167.2µF 116.38 Ω 261.87Ω 0.1364 H Low- Frequency Filter 15.04µF Low- Frequency Filter 6.685µF 12 Pulse 12 Pulse Rectifier Inverter 0.0061 H
High- Frequency Filter High- Frequency Filter 83.32 Ω 37.03 Ω
Figure 4.11. Single-line diagram of the CIGRE benchmark HVDC system
The aim of this research is to replace the tuned filters and reactive power supports from both the converter’s sides using the PHEV as a virtual active filter to obtain a low-cost filter and demonstrate that the HVDC terminal is a suitable place at which to connect a PHEV park, as shown in Figure 4.13 and Figure 4.19, while connecting the PHEV as a virtual active filter in the HVDC network is shown in Figure 4.12
4.9 HVDC Case Study
The 12-pulse converter of an HVDC network is a major source of harmonics in power systems. To observe the performance of the virtual active filter, the following two case studies are introduced.
1. Virtual active filter on rectifier side; and
2. Virtual active filter on both rectifier and inverter sides Section 4.9 HVDC Case Study 120
AC L1 ias R11 ia
AC va L1 R 12 pulse DC ibs 11 ib converter link AC Y/Y of HVDC vb L1 ics R11 ic
Transformer vc ABC To a ß DC voltage Regulator i PQ dc Generation L2 R12 iBa Reference Current R calculation L2 12 iBb Vdc Current L2 R 12 iBc control
Figure 4.12. PHEVs connection as virtual active filter in HVDC network
4.9.1 Case 1: Virtual filter at rectifier side
0.151 H 0.5968H 0.5968H AC 3.737 Ω 0.0365H0.7406 Ω 0.0365H AC 25 Ω 25 Ω 422.84kV : 230 kV 2160.633 345kV : 422.84 kV Ω 1196MVA 7.522µF 1196MVA 24.81 Ω 26µF 0.0606H 15.04µF 13.23 Ω
167.2µF 116.38 Ω 12 Pulse 12 Pulse Low- Frequency Filter 15.04µF V Rectifier Inverter dc 0.0061 H 1:m High- Frequency Filter 37.03 Ω PHEV
Figure 4.13. Modified CIGRE benchmark HVDC system with virtual filter on rectifier side
In this study, a virtual active filter is connected on the rectifier side with a Section 4.9 HVDC Case Study 121
fixed tuned filter on the inverter side. Three different analyzes are carried out to identify the performances of the virtual active filter in the HVDC link. A real-time harmonic current spectrum is analyzed and it is found that the different orders of the harmonic currents have maximum and minimum levels of magnitude of 0.21% and 0.01% respectively, as shown in Figure 4.14, and are periodic at a period twice that of the system for the p-q theory. The THD of the current of the virtual filter is also investigated and found to be 4.16% for the transient condition, which according to IEEE standard 519 [252], is within acceptable limits, while that of the steady state is 0.3%, as shown in Figure 4.15.
To obtain individual harmonic spectra, another analysis is carried out. As 12- pulse converters do not produce significant 5th- and 7th-order harmonics, their major contributor, the 11th-order harmonics [253], is observed for both the source and load currents using the virtual filter, as shown in Figure 4.17 and 4.18 respectively. The magnitude of the 11th-order harmonic of the source current is 0.058% and that of the load current 0.011%. To realize the compensating current and instantaneous reactive power characteristics, their wave forms are shown in Figure 4.16 where the capability of filtering by PHEV park is justified. Section 4.9 HVDC Case Study 122
0.25 11th order
0.2 7th order
5th order 0.15
0.1
0.05 Harmonics current in % In in current Harmonics 0 0.06 0.061 0.062 0.063 0.064 0.065 0.066 0.067 0.068 Time (Sec)
Figure 4.14. Harmonics current of load (one cycle) with virtual filter (case1)
5
4
3
2 THD (%) THD
1
0 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 Time (Sec)
Figure 4.15. THD with virtual filter (case1)
Reactive power and Compensating current 0.4
0.2
0 PU
−0.2 Reactive power Compensating current −0.4 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 Time (Sec)
Figure 4.16. Compensating current and reactive power output from virtual filter Section 4.9 HVDC Case Study 123
Figure 4.17. Harmonics spectrum of source current with virtual filter(y-axis odd har- monics)
Figure 4.18. Harmonics spectrum of load current with virtual filter (y-axis odd harmon- ics)
4.9.2 Case 2 : Virtual filter on both converter sides
0.151 H 0.5968H 0.5968H AC 3.737 Ω 0.0365H0.7406 Ω 0.0365H AC 25 Ω 25 Ω 422.84kV : 230 kV 2160.633 345kV : 422.84 kV Ω 1196MVA 1196MVA 24.81 Ω 0.0606H 26µF
12 Pulse 12 Pulse V Rectifier Inverter dc Vdc 1:m 1:m Vs - θs Vt - θt PHEV PHEV
Figure 4.19. Modified CIGRE benchmark HVDC system with virtual filter on both side Section 4.9 HVDC Case Study 124
Virtual active filters are used on both sides of the HVDC link to justify the performance of the PHEV-based virtual filter. As all the simulation results are taken from the inverter side, to determine the real-time harmonic current spectrum for the virtual filter there, Figure 4.20 helps to identify that its performance has maximum and minimum harmonic current levels of 0.32% and 0.013% respectively, while the
THDs of its currents in the transient and steady state conditions are 4.3% and
0.4% respectively, as shown in Figure 4.21. The harmonic load current’s individual spectrum analysis also shows that the 11th-order harmonic current’s magnitude is 0.029% for the virtual active filter as shown in Figure 4.22. As the analyzes demonstrate that the performance of the virtual filter can satisfy the requirements of IEEE Std 519-1992, using it could be an economic solution for the HVDC link.
11th order 0.3 7th order 0.25 5th order 0.2
0.15
0.1
0.05 Harmonics current in % of in In current Harmonics
0 0.06 0.061 0.062 0.063 0.064 0.065 0.066 0.067 0.068 Time (Sec)
Figure 4.20. Harmonics current of inverter (one cycle) with virtual filter (case2) Section 4.9 HVDC Case Study 125
5
4
3
THD (%) THD 2
1
0 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 Time (Sec)
Figure 4.21. THD with virtual filter (case2)
Figure 4.22. Harmonics spectrum of load current with virtual filter (case2)(y-axis odd harmonics) Section 4.10 Virtual Active Filter System for Improving Wind Power Quality 126
4.10 Virtual Active Filter System for Improving Wind Power Quality
Wind is one of the cleanest power generating technologies available today and, as it has little impact on the environment, can dramatically reduce greenhouse gas emis- sions while using PHEVs their excess generation of power during off peak, especially at night can be stored. PHEVs have the potential to revolutionize the generation and use of electricity in our everyday life.
Grid
Wind Generator
Transformer Non linear load Line of common PHEV Coupling
Figure 4.23. PHEV connection with wind farm
IEC standard 61400-21 was developed and released by the International Elec- trotechnical Commission (IEC) as part of the IEC 61400 standards for testing and assessing the power quality characteristics of grid-connected wind energy converters
(WECs) in a consistent and accurate way. To easily compare WECs of different types, it states that, in terms of measurement, the key power quality characteris- tics are the wind turbine (WT) specifications, active and reactive power output, Section 4.11 Modeling of Wind Farm 127
flicker, harmonics, control of active and reactive powers, protection, voltage dip re- sponse and reconnection time. Harmonics is one of the main reasons for significant power losses in distribution and communication systems and, sometimes, operational failures of electronic equipments. Consequently, as the power quality delivered to consumers is an object of great concern, it is mandatory to solve the harmonic problems [254]. Therefore, we propose implementing V2G technology to design an active filter to improve the quality of wind power output under such circumstances through compensating reactive power and harmonic current. A simple connection of a PHEV to a wind power system is shown in Figure 4.23.
4.11 Modeling of Wind Farm
In recent years, significant interest in using wind power for the generation of elec- tricity has been observed worldwide. However, its increasing penetration in inter- connected power systems requires particular attention to be paid to its modeling in order to study its impact and controls. Wind power is particularly interesting from the modeling viewpoint since it combines stochastic models (such as wind speed), mechanics (such as WT), electrical machines, power electronics and controls. For power system stability, modeling these systems requires careful examination of their equipment and controls to determine the characteristics that must be achieved in the time frames and bandwidths of such studies. The performance of a wind farm or a model of it is extremely dependent on the equipment used. In the literature, there are four notions of WTs based on speed, that is, constant speed, limited variable Section 4.11 Modeling of Wind Farm 128 speed, variable speed with a partial-scale frequency converter and variable speed with a full-scale frequency converter [198]. The fixed-speed squirrel-cage induction generator (SCIG) is the oldest and most used generator to date. This type of wind farm is known as the ‘Danish concept’ and is relatively cheap and electrically effi- cient. From a market perspective, the dominant WT is the doubly fed IG (DFIG).
However, this thesis focuses on the fixed-speed WT (FSWT) technology. A FSWT schematic structure with a SCIG is shown in Figure 4.24. It is one of the simplest types of WT technology that converts the kinetic energy of wind into mechanical energy using which the generator operates and produces electrical energy for direct delivery to the grid. The rotational speed of the generator, depending on the number of poles, is relatively high (in the order of 1000–1500 rpm for a 50 Hz system fre- quency). A gear box is used to transform this rotational speed to maintain turbine efficiency and reduce the mechanical stress from such a high-speed rotation.
A reactive power supply from the external grid is essential to flux the rotor circuit through the stator of the fixed-speed IG (FSIG). This reactive power consumption results in the unit demonstrating a low full-load power factor. To maintain power grid stability and reduce transmission loss, capacitor banks or reactive power com- pensation devices (SVCs or STATCOMs) are installed to compensate the reactive power consumed by the FSIG rather than take in reactive power from the grid. The main challenge involved in utilizing a FSIG in wind generation is the excessive re- active power it absorbs from the power system, which has become an issue of great Section 4.11 Modeling of Wind Farm 129
Bypass Turbine
Transformer Gear Box
Squirrel cage Starter Induction Compensator Generator
Figure 4.24. System structure of wind turbine with directly connected SCIG concern when wind generation is connected in a weak power system with a limited reserve of reactive power and during voltage sag conditions arising from switching-in or system short-circuit fault events. Wind power has evolved rapidly over the last two decades in terms of WT power ratings and, consequently, the rotor diameters of WTs have increased. In the past few years, a different type of development has taken place: instead of continuously increasing WT-rated power, WT manufacturers have focused on developing WTs that are more reliable, grid code-compliant and suitable for different installation environments- onshore and offshore. Recently, a commercial offer from the wind industry, which has the majority of WTs has been rated at around 2 to 3 MW. As wind farms become a larger part of the total gener- ation of power systems worldwide, issues related to integration, stability effects and voltage impacts become increasingly important. Adequate load flow and dynamic Section 4.11 Modeling of Wind Farm 130 simulation models (encompassing all their significant air-dynamical, mechanical and electrical factors) are necessary to evaluate their impacts on power systems.
4.11.1 Dynamic models of wind generators
This dissertation uses a model of an IG written in an appropriate d-q reference frame to facilitate an investigation into control strategies. Figure 4.25 depicts the general structure of a model of a constant-speed WT, the most important components of which are the rotor, drive train and generator, combined with a wind speed model.
Pm P and Q
Wind V P Induction Frequency w Rotor m Shaft speed generator and model model model model grid model
ωg V and f
Figure 4.25. General structure of constant-speed wind turbine model
4.11.2 Rotor model
WTs are the main components of wind farms. They are usually mounted on towers to capture the most kinetic energy and, because wind speed increases with height, taller towers enable them to capture more energy and generate more electricity. The three-bladed rotor, consisting of three blades and a hub, is the most important and most visible part of a WT through which the energy of the wind is transformed into mechanical energy that turns the main shaft of the WT.
The rotor of a WT, with radius Ri, converts energy from the wind to the rotor
shaft, rotating at the speed of ωmi . The power from the wind depends on the wind
speed, Vwi , air density, ρi, and swept area, Awti . From the available power in the Section 4.11 Modeling of Wind Farm 131
swept area, the power on the rotor is given based on the power coefficient (cpi (λi, θi)), which depends on the pitch angle of the blade, θi and the ratio between the blade
ωmi Ri tip and wind speeds denoted as the tip-speed ratio, λi = . The aerodynamic Vwi torque applied to the rotor for the ith turbine by the effective wind speed passing through the rotor is given as [197]:
ρ T = i A c (λ , θ )V 3 (4.29) aei wti pi i i wi 2ωmi
where cpi is approximated by [198]:
[ ] − − π(λi 3) − − cpi = (0.44 0.0167θi) sin 0.00184(λi 3)θi 15 − 0.3θi where i = 1, ··· , n and n is the number of WTs.
A controller equipped with a WT starts up the machine at a wind speeds of about 8 to 16 miles per hour (mph) and shuts it off at about 55 mph. Turbines do not operate at wind speeds above about 55 mph because they might be damaged.
The radius of a 2 MW WT is about 80 m, the typical air density is 1.225 kg/m3, cp is in the range of 0.52–0.55, towers range from 60 to 90 meters (200 to 300 feet) high and the blades rotate at 10-22 revolutions per minute.
Equation 4.29 shows that aerodynamic efficiency is influenced by variations in the blade’s pitch angle. Regulating the rotor blades provides an effective means of regulating or limiting the turbine’s power during high wind speeds or abnormal conditions. A pitch control turbine performs power reduction by rotating each blade Section 4.11 Modeling of Wind Farm 132 about its axis in the direction of the angle of attack. In comparison with passive stall, pitch control provides greater energy capture at the rated wind speed and above. The aerodynamic braking facility of pitch control can reduce extreme loads on a turbine and also limit its power input so as to control possible over-speed of the machine if the loading of the turbine-generator system is lost, for instance, because of a power system fault. In a pitch-controlled WT, electronic controllers check the power output of the turbine several times per second and when it becomes too high, a message is sent to the blade-pitch mechanism, which immediately turns the rotor blades slightly in an attempt to restore this output to an acceptable value. In this work, the pitch-rate limit is set to the typical value of 12 deg s−1.
4.11.3 Shaft model
A two-mass drive train model of a WT generator system (WTGS) is used in this dissertation as drive train modeling can satisfactorily reproduce the dynamic char- acteristics of a WTGS because the low-speed shaft of a WT is relatively soft [255].
Therefore, although it is essential to incorporate a shaft representation into the constant-speed wind turbine model, only a low-speed shaft is included while the gearbox and high-speed shaft are assumed to be infinitely stiff. AS the resonance frequencies associated with gearboxes and high-speed shafts usually lie outside the frequency bandwidth of interest [256], we use a two-mass representation of the drive train.
The drive train attached to the WT converts the aerodynamic torque (Taei ) on Section 4.11 Modeling of Wind Farm 133 the rotor into the torque on the low-speed shaft scaled down through the gear-box to the torque on the high-speed shaft. The first mass term stands for the blades, hub and low-speed shaft and the second the high-speed shaft, which have inertia
constants of Hmi and HGi respectively. The shafts are interconnected by a gear
ratio (Ngi ) combined with torsion stiffness (Ksi ) and torsion damping (Dmi and
DGi ), which result in a torsion angle, (γi). The normal grid frequency is f and the dynamics of the shaft are represented as in [197]:
1 − − ω˙ mi = [Taei Ksi γi Dmi ωmi ] (4.30) 2Hmi 1 − − ω˙ Gi = [Ksi γi Tei DGi ωGi ] (4.31) 2HGi − 1 γ˙i = 2πf(ωmi ωGi ) (4.32) Ngi
The generator receives the mechanical power from the gear-box through the stiff shaft and the relationship between the mechanical torque and the torsional angle is given by:
Tmi = Ksi γi (4.33)
The gear-box connects the low-speed and high-speed shafts and increases the rotational speeds from about 30 to 60 rotations per minute (rpm) to about 1000–
1800 rpm, the rotational speed required by most generators to produce electricity. Section 4.11 Modeling of Wind Farm 134
4.11.4 Induction generator model
An IG can be represented in different ways, depending on the level of detail re- quired, and is characterized mainly by the number of phenomena, such as the stator and rotor flux dynamics, magnetic saturation, skin effects and mechanical dynam- ics, included. Although a very detailed model which includes all these dynamics is possible, it may not be beneficial for stability studies because it increases the com- plexity of the model and requires time-consuming simulations. More importantly, it has been shown in stability studies that not all these dynamics have significant influence. A comprehensive discussion of comparisons of different IG models can be found in [198]. As including iron losses in a model is a complicated task, they are neglected in stability studies. Also, as the main flux saturation is only of impor- tance when the flux level is higher than the nominal level, it can be neglected for most operating conditions and the skin effect should only be taken into account for a large-slip operating condition, which is not the case for a FSWT. Another constraint of including dynamics in a model is the availability of relevant data. As, typically, saturation and skin effect data are not provided by manufacturers, in general, it is impractical to use them in WT applications and, to represent FSIG models in power system stability studies [257] the stator flux transients can be neglected in terms of voltage relationships. All these arguments lead to the conclusion that rotor dynamics are the only major factors required to be considered in an IG model for a voltage stability analysis. Therefore, representing a third-order model of an IG Section 4.11 Modeling of Wind Farm 135 offers a compatibility with the network model and more efficient simulation times.
Although the main drawbacks of the third-order model are its inability to predict peak transient current and, to some extent, its less accurate estimation of speed, at a relatively high inertia, it is sufficiently accurate. The transient model of a SCIG is described by the following DAEs [198], [258]:
1 − s˙i = [Tmi Tei ] (4.34) 2HGi [ ] ˙ ′ 1 ′ ′ ′ E = − E − (Xi − X )ids − siωsE (4.35) qri T ′ qri i i dri oi [ ] ˙ ′ 1 ′ ′ ′ E = − E + (Xi − X )iqs + siωsE (4.36) dri T ′ dri i i qri oi
V = R i − X′i + E′ (4.37) dsi si dsi i qsi dri
V = R i + X′i + E′ (4.38) qsi si dsi i qsi qri √ v = V 2 + V 2 (4.39) ti dsi qsi
′ where Xi = Xsi +Xmi Xri /(Xmi +Xri ) is the transient reactance, Xi = Xsi +Xmi the rotor open-circuit reactance, T ′ = (L + L )/R the transient open-circuit time oi ri mi ri constant, v the terminal voltage of the IG, s the slip, E′ the direct-axis transient ti i dri voltages, E′ the quadrature-axis transient voltages, V the d-axis stator voltage, qri dsi
Vqsi the q-axis stator voltage, Tmi the mechanical torque, Tei = Edri idsi + Eqri iqsi ,
the electrical torque, Xsi the stator reactance, Xri is the rotor reactance, Xmi the
magnetizing reactance, Rsi the stator resistance, Rri the rotor resistance, HGi the Section 4.11 Modeling of Wind Farm 136
inertia constant of the IG, and idsi and iqsi the d- and q-axis components of the stator current, given by:
∑n [ ] ′ − − ′ Idi = Edrj(Gij cos δji Bij sin δji) Eqrj(Gij sin δji + Bij cos δji) (4.40) j=1 ∑n [ ] ′ ′ − Iqi = Edrj(Gijsinδji + Bijcosδji) + Eqrj(Gijcosδji Bijsinδji) (4.41) j=1
The equations that describe a SCIG and DFIG are identical except that the former’s rotor is short-circuited. The converter for the VSWTs [198] used in this thesis consists of two VCSs connected back to back which enables variable-speed operation of the WTs using a decoupling control scheme to separately control the active and reactive components of the current. The modeling of IGs for power-flow and dynamic analyses is discussed in [198], [258] while a general model representing
VSWTs in simulations of power system dynamics is presented in [259].
4.11.5 Aggregated model of wind turbine
The development of aggregated models of wind farms is also an important issue because, as the sizes and numbers of turbines on them increase, representing wind farms as individual turbines increases complexity and leads to time-consuming sim- ulations, which are not beneficial for stability studies of large power systems. To aggregate WTs, models of several identical WTs (including some facing the incoming wind) are combined in a single-turbine model with a higher rating. The parameters are obtained by preserving the electrical and mechanical parameters of each unit, and by increasing the nominal power to the equivalent of the turbines involved in Section 4.11 Modeling of Wind Farm 137 the aggregation process [260]. This aggregated model reduces computational and simulation times in comparison with those of a detailed model with different rep- resentations of tens or hundreds of turbines and their interconnections. However, specific care is required in choosing what to aggregate in order to be as close to reality as possible. In addition, this type of modeling is difficult for WTs without a parallel distribution (namely as an array, which is the most common distribution for offshore, but not onshore, wind farms).
Pitch angle controller for wind generator
The pitch angle of the wind generator, θ, is controlled to avoid super-synchronous speeds, as shown in Figure 4.26 [200]
ωm + k p θ
1 Tp s - ωref
Figure 4.26. Pitch angle control diagram
The controller’s dynamic equation can be expressed as:
˙ θ = (Kf(ωm − ωref ) − θ)/Tp (4.42) where f is a function, which determines the pitch angle set point with a predefined limit of (ωm − ωref ) Section 4.12 Harmonics in Wind Power 138
4.12 Harmonics in Wind Power
The harmonic current emission from a wind farm is an important issue in power quality. Since harmonic sources are treated as current injections [261], the total harmonic current distortion can be defined as [262]: v u u∑∞ 1 t 2 THDI = Ih (4.43) In h=2
where Ih are individual harmonic components and In The rated wind farm current.
4.13 Network Interfacing of Wind Generator
In a synchronously rotating reference frame the link between a network and the stator machine active and reactive power production are:
pt = Vdsi Idi + Vqsi Iqi (4.44)
− 2 2 qt = Vqsi Idi Vdsi Iqi + bc(Vds + Vqs) (4.45)
where bc, is the conductance, which determines the amount of compensation current for both the harmonics current and reactive power support for the wind farm. To obtain better results, we have divided the compensator into two parts, a fixed compensator, which is the minimum requirement for a wind farm under normal operating conditions, bcf , and a virtual compensator, which provides reactive power under both normal and voltage sag conditions with an harmonic current compensation capacity, bcv. The second part of the compensator is the virtual active Section 4.13 Network Interfacing of Wind Generator 139
filter for the wind farm. Therefore the term bc can be expressed as bc=bcf +bcv and bcv can be written as:
Icomp bcv = (4.46) vti
where Icomp, is the compensating current from the filter system, as shown in Figure
4.27.
L R ias 1 11 ia A v a 6 pulse 35mH L1 R rectifier ibs 11 i b as a source vb of 3.5 Ω ics L1 R11 harmonics ic
vc
dq a ß ABC ABC Wind farm To To To Fixed To ABC dq a ß a ß compensator DC voltage Regulator PQ idc Generation
L2 R12 iBa Reference Current L2 R12 calculation iBb Vdc Current
L2 R12 control iBc
Figure 4.27. System configuration
The term Icomp, denotes both the Idsicomp, and Iqsicomp, generated using the
Park-Clarke transformation [263], and the compensating current in the α-β axis is calculated using the equation (4.28). In this case study a single induction generator based wind farm is considered with nonlinear load. Possibility of using PHEV as
filter for individual generator is investigated here with generator capacity 1.8 MW and using the power, voltage base, Sbase = 10kV A, Vbase = 240V , respectively. Section 4.13 Network Interfacing of Wind Generator 140
4.13.1 Simulation results
A 6-pulse rectifier is used as a source of harmonics with a grid-connected wind gen- erator and a PHEV connected at the point of common coupling as a shunt active
filter, as shown in Figure 4.27. Simulation results demonstrate that the system and load current harmonics decrease by using a PHEV as an active filter, as shown in
Figure 4.28 and 4.29 respectively. The current harmonic spectra of the source and load sides without and with the filter are shown in Figure 4.30 and 4.31 respectively.
The current compensated by the filter shown in Figure 6.19 improves the power fac- tor of the system, as shown in Figure 4.34. The angle difference between the voltage and current is reduced from that shown in Figure 4.34 without any filter to that illustrated in Figure 4.33 using the PHEV as a filter.
Figure 4.28. System and load current at phase A, without PHEV as filter Section 4.13 Network Interfacing of Wind Generator 141
Figure 4.29. System and load current at phase A, with PHEV as filter
Figure 4.30. Current harmonics spectrum without filter
Figure 4.31. Current harmonics spectrum with filter Section 4.13 Network Interfacing of Wind Generator 142
Figure 4.32. Compensating current
Figure 4.33. Load current and voltage at phase A, without filter
Figure 4.34. Load current and voltage at phase A, with filter Section 4.14 Chapter Summary 143
4.14 Chapter Summary
The goal achieved through this study is an investigation into the filtering and reac- tive power compensation performances of PHEVs used as an active filter in a HVDC network and in a wind farm. We examine the HVDC link’s converters and system’s harmonic current with a virtual active filter in the system for two different cases.
As the results obtained from simulations show that the harmonic reduction perfor- mances of a HVDC network with a PHEV used as an active filter are within the acceptable range, opportunities arise to use PHEVs and ensure a connecting point of a PHEV park at an HVDC converter terminal. The load and system currents, and load’s voltage and current with and without using PHEVs as active filters in a wind power system, are investigated. As simulations show that both power quality and the power factor are improved with a PHEV used as an active filter, a large number of low-power active filters in the same power system, close to each problematic load or generator group can be used to avoid the circulation of current harmonics and reactive currents through power lines. Chapter 5
Design of Virtual FACTS Devices with PHEVs Park
A Unified Power Flow Controller (UPFC) is an electrical device, that combines to- gether the features of two FACTS devices: the Static Synchronous Compensator
(STATCOM) and the Static Synchronous Series Compensator (SSSC), which can fulfill various power flow control objectives, such as the needs of reactive shunt com- pensation, phase shifting and series compensation. However, as they are expensive, they are not widely used. In this chapter, the potential of PHEVs in V2G mode of operation for the design of a virtual UPFC (VUPFC) using a PHEV charging station is explained.
A Unified Power Quality Conditioner (UPQC) is a type of power quality control device, which can be a reactive power compensator, a voltage flicker controller and a harmonics current compensator. However, as it is also expensive, and not widely used. In this chapter, a simple structure of a UPQC using PHEVs is proposed.
144 Section 5.1 Unified Power Flow Controller (UPFC) 145
5.1 Unified Power Flow Controller (UPFC)
5.1.1 Introduction
The increasing amount of PHEV fleet penetration into the market offers an enormous opportunity to use V2G technology to improve the power quality of the utility grid.
PHEVs can be charged from a household’s electrical connection, a charging station and even a parking lot during the day while their bidirectional chargers can be designed as UPFC converters.
The power system parameters (transmission voltage, line impedance and phase angle), which determine the transmittable power can be controlled with any com- bination in real-time using a UPFC, which can be described as a generalized power
flow controller able to maintain some of the real and reactive powers necessary for both normal and temporary system operating conditions. Compared with other
FACTs devices, such as a static compensator (STATCOM) and thyristor controlled series capacitor (TCSC), it is evident that a UPFC is unique in its ability to control both real and reactive powers [264], [265].
According to Mithulananthan Nadarajah et al. although a UPFC is one of the most versatile of the FACTS controllers developed to date, it is not widely used for power quality improvement because of its cost [266], which on average, is twice that of a STATCOM per kVA [267]. The most expensive components for the constitution of FACTS and HVDC devices are the capacitors [268]. Given this situation, this chapter investigates the possibility of employing PHEVs as UPFCs through their Section 5.1 Unified Power Flow Controller (UPFC) 146 bidirectional converters, which can run in different modes of operation according to need, as reported in [269].
5.1.2 UPFC in power system
In this section, a survey of the literature related to UPFC operation, modeling and control is presented. The UPFC proposed by L. Gyugyi in 1991 [270–272],is one of the most complex FACTS devices in present power systems. It is primarily used for the independent control of the real and reactive powers in transmission lines to achieve the flexible, reliable and economical operation and loading of a power system. Until recently, all four parameters affecting the real and reactive power
flows on a line, namely, the line impedance and voltage magnitudes at the terminals of the line and power angle, have been controlled separately using either mechanical or other FACTS devices, such as a static VAR compensator (SVC), TCSC or phase shifter. However, the UPFC allows the simultaneous or independent control of these parameters by transferring from one control scheme to another in real-time.
In addition, it can be used for voltage support, transient stability improvement and the damping of low-frequency power system oscillations. Because of its attractive features, the modeling and control of a UPFC have come under intense investigation in recent years. Several references to the development of a UPFC steady state and dynamic and linearized models can be found in the technical literature. A steady state model, referred to as an injection model, is described in [273], in which a
UPFC is modeled as a series reactance at each end of which the dependent loads Section 5.1 Unified Power Flow Controller (UPFC) 147 are injected. This model is simple and helpful for understanding the impact of a
UPFC on a power system. However, to determine the desired load flow solution, the amplitude modulation and phase-angle control signals of the series-voltage source converter have to be adjusted manually. If a UPFC is operated in the automatic control mode (that is, to maintain a pre-specified power flow between two power system buses, the sending and receiving buses, and to regulate the sending-end voltage at a specific value), the UPFC’s sending end is transformed into a PV bus and its receiving end into a PQ bus so a conventional load flow program can be performed [274]. This method is simple and easy to implement but will only work if real and reactive power flows and the sending bus voltage magnitude are controlled at the same time. It should also be mentioned that there is no need for an iterative procedure such as that used in [274] to compute the UPFC’s control parameters as they can be calculated directly after the conventional load flow solution is found.
Because of the advantages the automatic power flow control mode offers, it is used as the basic operational mode for most practical applications. A Newton-Rhapson based algorithm for large power systems with embedded FACTS devices is derived in [275] and in [276] is extended to include UPFC applications. It allows either the simultaneous or independent control of the real and reactive powers and voltage magnitudes but is complicated and difficult to implement. It considerably increases the order of the Jacobian matrix in the iterative procedure and is sensitive to initial condition settings, which if improperly selected, can cause the solution to oscillate Section 5.1 Unified Power Flow Controller (UPFC) 148 or diverge. A UPFC dynamic model known as the fundamental frequency model can be found in [277], [274], [278] and [279]. It consists of two voltage sources, one in a series and the other in a shunt, connected to a power network to represent series and shunt voltage source inverters both of which are modeled to inject only voltages of the fundamental power system frequency. As the model in [278] neglects the DC-link capacitor dynamics, the results obtained from it are inaccurate but, as those in [277], [274] and [279] include these dynamics, they can be used to study the effect of a UPFC on a real power system’s behavior. In this study, to design a
VUPFC, the dynamics of a PHEV’s battery is considered.
The basic control design of a UPFC involves control of the real and reactive power flows, and the sending bus and DC voltage magnitudes. The most frequently used control scheme is based on the vector-control approach proposed by Schauder and Metha in 1991 [280], which allows decoupled control of the real and reactive powers, thereby making it suitable for UPFC applications, which can be accom- plished by transforming the three-phase balanced system into a synchronously ro- tating orthogonal system. A new coordinate system is chosen in such a way that its d-axis component coincides with the instantaneous voltage vector and its q-axis component is orthogonal to it. In this control system, the d-axis current compo- nent contributes to the instantaneous real power and the q-axis current accounts for the reactive power, and it can be applied for both series and shunt converter control [277], [270], [278], [281]. Another approach for the automatic power flow Section 5.1 Unified Power Flow Controller (UPFC) 149 control of a series converter is to decompose the voltage drop between the sending and receiving buses into two components: one in phase with the sending bus, which has a strong influence on the reactive power flow, and the other orthogonal to it, which mainly influences the real power flow [279]. Also, the shunt converter can use decoupled P-Q controllers to control the sending bus voltage and DC voltage mag- nitudes [113]. This control scheme is simple and easy to implement and is used in this thesis. UPFC damping controller designs can be found in [279] and [281–284].
An additional control can be applied to the shunt inverter through modulating the voltage-magnitude reference signal or to the series inverter through modulating the power reference signal. Although, in [281] and [284], the slip of the desired machine is used as the input signal to the damping controller, in general, as it is difficult to obtain, this kind of control is not feasible. Therefore, controllers that depend on local measurements, such as the tie-line power flow or the UPFC terminal voltage phase angle difference, are more appropriate [279], [283].
To achieve the stated objective, the following research tasks are performed.
• A UPFC load flow or steady state model needed to initialize the simulation
is developed.
• A UPFC dynamic model with controllers that can be used for transient sta-
bility studies is developed.
• The model is interfaced with the power system.
• To demonstrate the performance of the controller under dynamic conditions, Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 150
a power system extensively used in the literature, which consists of two ar-
eas, each with two generating plants, is used. The simulation results show
that, when a large disturbance is applied, the VUPFC can enhance the power
system’s operation and performance in the same way as a standard UPFC
available in market.
5.2 Basic Structure of Virtual UPFC (VUPFC)
A UPFC can provide several basic control parameters for a power system and act in accordance with control objectives, such as reactive shunt compensation, series compensation and phase shifting. These objectives are fulfilled through a pair of transformers connected to a transmission line, one in a series to inject voltage into the line and the other in a shunt to achieve the control objectives. Besides the transformers, a UPFC contains back-to-back AC to DC voltage source converters with a common DC-link capacitor, as shown in Figure 5.1. In this study, a PHEV park replaces this capacitor and converts by the bidirectional charger of the PHEV park, as shown in Figure 5.5.
Vh - θh Vk- θk
ph+jqh pk+jqk
h k
Vdc
Figure 5.1. Basic single line diagram of UPFC Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 151
5.2.1 VUPFC model
The VUPFC is designed based on [285–287], in which the active power demand of the network is provided by a shunt converter but, as it can also provide or absorb reactive power when necessary, it can independently compensate the reactive power for the line. By injecting a voltage with a controllable magnitude and phase angle in a series with the line, the series converter performs the function of a VUPFC, as shown in Figure 5.2.
- ‘ V h Vh - Vk Vse jXhk - - + Ise
pse qse
psh
h - k Ish qsh
Figure 5.2. Simplified VUPFC circuit
¯ The circuit model is represented by one series voltage source (Vse) and one shunt
¯ current source (Ish) defined as:
− ¯ j(θh ϕ) j(α+θh) Vse = (vp + vq)e = Vsee (5.1)
¯ θh Ish = (ip + jiq)e (5.2) where:
¯ ¯ vp is the component of Vse that is in phase with the line current (Ise); Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 152
¯ ¯ vq is the component of Vse that is in quadrature with the line current (Ise);
¯ ip is the component of Ish in phase with the voltage (¯vh); and
¯ iq is the component of Ish in quadrature with the voltage (¯vh).
- ish
vq iq - Vse
- α ip vh Ɵh
Φ - - Φ vp
ih
Figure 5.3. Victor diagram of VUPFC
The equivalent circuit vector diagram of the VUPFC is shown in Figure 5.3 and the resulting power equations that describe its power injection model are:
1 ph = (VhVk sin(θh − θk) + VhVse sin α) (5.3) Xhk
pk = −ph (5.4)
1 2 qh = (vh − VhVk cos(θh − θk) + VhVse cos α) − iqvh (5.5) Xhk
1 2 qk = (vk − VhVk cos(θh − θk) − VkVse cos α) (5.6) Xhk
The reactance, Xhk, is defined as the series reactance of the loss-less transmission line connected in series to the VUPFC device and expressed as:
2 Xhk = Xseu max(SB/SS) (5.7) Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 153
where Xse, is the series transformer reactance, u, the injected voltage magnitude in per unit, SB, the system base power and SS, the nominal rating power of the series converter.
The differential equations of the VUPFC are:
1 v˙p = (vp0 − vp) (5.8) Tr 1 v˙q = (vq0 − vq) (5.9) Tr
1 ref i˙sh = [Kr(v − vh) − ish] (5.10) Tr
where Tr, is the regulator time constant and Kr, the regulator gain.
Operation of the VUPFC demands that the series and shunt branches have the proper power ratings to enable the VUPFC to carry out the pre-defined power flow objective. The algorithm for VUPFC rating starts with definition of the series transformer short circuit reactance, xse, and the system base power, SB. Then, the initial estimation is given for the series converter rating power, Ss, and the maximum magnitude of the injected series voltage, u.
Load flows are computed changing the angle α between 0o and 360o in steps
o of 10 , with the magnitude u kept at its maximum value umax. Such rotational changes in this VUPFC parameter influence the active and reactive power flows in the system with the largest impact being on the power flowing though the line on which the VUPFC is installed. Therefore, regulation of the active and reactive Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 154 power flows through the series branch of the VUPFC could be set as the initial pre- defined objective to be achieved within the VUPFC steady state operation. To get the optimum size of the VUPFC, the used algorithm is summarized in a flowchart as shown in Figure 5.4
To design a VUPFC using PHEVs, it is essential to assess the PHEVs’ interface with the power system and the dynamics of their batteries. As they need an elec- tronic converter to connect to electrical networks for battery charging, in this work, a bidirectional converter is considered as the charger and a dynamic battery model of a PHEV considering the dynamic response of the electrolytic temperature and battery’s state-of-charge (SOC) is used as in [288], [289].
The controls of the bidirectional converters for the VUPFC are designed for ±
20 MW of power transaction with the grid indicate a park with around 1334 vehicles
(15 kWh each vehicle). This type of large parking is quite reasonable to assume in a typical city or even in a business center.
Therefore our proposed VUPFC model can be represented by Figure 5.5
5.2.2 Controller design
As the UPFC is realized as a combination of series and shunt converter-based FACTS devices with a common DC voltage, Vdc, we have two different simple controller structures. The shunt converter is controlled by controlling the AC and DC voltages to obtain the firing angle and modulating amplitude while the series converter has a simple decoupling controller for the active and reactive powers. Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 155
Define Xse,SB
umax, Initila Ss
Calculate Xhk s s S S
Perform load flow Increase Decrease
Is load flow No requirement fulfiled?
Yes
Calculate pse, qse,Ss
No Is Ss No Yes minimum? If max SB >Ss
Yes
Perform load flow Claculate psh, qsh Output Ss, umax
Figure 5.4. Flowchart for optimum size of VUPFC Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 156
Vh - θh Vk- θk RT+jXT ph+jqh Ih Ik pk+jqk
mse :1 h m:1 k
psh+jqsh pse+jqse Rsh+jXsh Rse+jXse
Vdc Shunt idc R Series Converter 0 Converter Vp
Ip Rp R2
PHEV Park R1 C1
Em im
Figure 5.5. VUPFC model using PHEV
5.2.3 Shunt converter control
A bidirectional converter (as a rectifier and inverter ) and a transformer with reac- tance, xsh, are shown in Figure 5.6, where the DC voltage is regulated by means of the converter’s modulating amplitude, m, as in [200]:
√ ( ) 2 2 xsh 2 Vs m = psh + qsh + (5.11) VskVdc xsh Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 157
C1 R0 R2 idc im R1
psh+jqsh Ip
Vdc Rp Vp E m
1:m θ Vs - s Vt - θt
Figure 5.6. Shunt converter connection with PHEV in network
As the DC power of the battery (Pdc=VdcI ) is considered the real power in the network ( ps=Pdc), the link with the AC network is:
VtVs psh = sin(θt − θs) = VdcI (5.12) xsh 2 VtVs Vs qsh = cos(θt − θs) − (5.13) xsh xsh
√ where Vt=kmVdc , and the rectifier gain is k= (3/8). Therefore the relationship between θt and θs can be expressed as:
( ) xshI θt = θs + asin (5.14) kmVs
And the final equation for qsh is:
√ ( ) 2 2 Vs VskmVdc xshI qsh = − 1 − (5.15) xsh xsh kmVs Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 158
where psh is the real power, qsh the reactive power, Vs the voltage, θs the phase angle at the connecting bus. Vt and θt the voltage and phase angle respectively before the transformer.
The PHEV is connected to the ac network through the bidirectional converter and completed by the control that regulates the modulating amplitude, m and firing angle, α. Special care is taken to develop the operating limits of the converter, where
α and m both limited by the boundary conditions as:
αmin ≤ α ≤ αmax
mmin ≤ m ≤ mmax
The PHEV’s battery current, im is subjected to a constant power control as :
[ ] lim 1 Vdcidc i˙m = − im (5.16) Tm Em0
The battery current set point is limited by the SOC and then the currents ishd and ishq are regulated through a set of PI controllers as shown in Figure 5.7 [290]. Vshdr and Vshqr, determine the amplitude and firing angle passing through another set of
PI controllers, which can be expressed by:
m˙ = (Km(Vref − Vs) − m)/T (5.17)
x˙ a = Ki(Vref − Vs) (5.18)
0 = Kp(Vref − Vs) + xa − α (5.19) Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 159
√ 2 2 where Vs = Vshd + Vshq
Ps Vshd Ishd max
1 Ki Ps_ref ++ Kp + S Ishd_ref - 1 + s Trs
Ishd min Vshdr a max Ishd XL a Generation a of m and a min mmax m XL Ishq m Vshqr min Qs Ishq max
- Ki Qs_ref + Kp +1 S Ishq_ref 1 + s Trs Vshq Ishq min
Figure 5.7. Modified decoupled PQ controller for shunt converter
5.2.4 Series converter control
A decoupled P-Q controller shown in Figure 5.8 [277] is used to control the series converter of the VUPFC and the output variables (X1 and X2) of PI controllers are used to calculate its output voltages (Vsed and Vseq) respectively.
The dynamic equations for the series converter can be expressed as [200]:
( ) 2P kref x˙ 1 = KI − Ikd (5.20) Vkd ( ) ˙ 2Pkref Ikd = x1 − KIkd + Kp − Ikd (5.21) ( ) Vkd 2Qkref x˙ 2 = KI − Ikq (5.22) Vkd ( ) ˙ 2Qkref Ikq = x2 − KIkq + Kp − Ikd (5.23) Vkd Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 160
The other control parameters are:
R + R K = T se Ω (5.24) XT + Xse √ Vkd = 2V k (5.25) √ Vhd = 2V h cos(θk − θh) (5.26) √ Vhq = 2V h sin(θk − θh) (5.27) ( ) ∗ 2P kref − x1 = x1 + KI ΩIkd (5.28) ( Vkd ) ∗ 2Qkref x2 = x2 + KI + ΩIkd (5.29) Vkd X − X V = V − V − T se X (5.30) sed hd kd Ω 2 X − X V = V − T se X (5.31) seq hd Ω 2 √ 1 2 2 Vse = √ V + V (5.32) 2 sed seq √ ( ) 8 Vse mse = (5.33) 3 Vdc where Ω is the fundamental frequency base in rad/s.
The DC voltage is controlled between its maximum and minimum limits by a set of PI controllers, as shown in Figure 5.9.
max dVdc
V m K 1dc dc Kpdc 1+ Ks Idc dVdc Vdc - T dc1 +S s Tr+ 1 + 1 +S s Tr
min dVdc Vdc_ref
Figure 5.9. DC voltage controller Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 161
Ikd Imax Ikd Pk_ref Ikd_ref X*1 X1 + 1 1 - + + PI 1 +PI s Tr - + 1 + s Tr
Imin ω1 G 1 + s Tr Vkd ω 1- sT
G ω1 ω 1 + s Tr Vkq 1- sT Imax Ikq_ref X2 Qk_ref X*2 - + 1 + 1 - 1 +PI s Tr + + 1PI + s Tr Ikq Ikq Imin
Figure 5.8. Decoupled PQ controller for series converter
5.2.5 Test system design and simulation results
BUS 5 BUS 6
LOAD BUS 4
BUS1 BUS 3 BUS 2
UPFC AC AC
Figure 5.10. 6-Bus test system for VUPFC
To validate the proposed UPFC model, a test system operating at 230 kV, with two Thevinen impedance sources connected through transmission lines and a T-tap Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 162
0.4
0.2
0
−0.2 I ref shd
Current (PU) Current I −0.4 shd
−0.6 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (Sec)
Figure 5.11. Shunt converter d axis current of VUPFC terminated with a ∆-Y transformer and rated at 230 kV/25 kV, is designed as shown in Figure 5.10. The Y-Y connected shunt transformer of the VUPFC is rated at 20
MVA and 230kV/21kV, and the Y-∆ connected series transformer at 20 MVA and
92kV/21kV.
In support of using a PHEV park as a VUPFC, two types of simulation results are presented. In the first phase of this study, to illustrate the characteristics of the
VUPFC, the internal output and reference values of the currents are presented. In the second, the performance of the VUPFC is compared with that of a standard
UPFC connected between buses 2 and 3 with a fault applied between buses 4 and 5 at t=0.5 sec and abolished at t=0.75 sec. Figure 5.11 and 5.12 show the shunt and series converters’ d-axis currents respectively, and Figure 5.13 and 5.14 their q-axis currents respectively, all of which follow the reference current both before and after the fault. Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 163
1
0.8
0.6
0.4 I Current (PU) Current kd 0.2 I ref kd 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (Sec)
Figure 5.12. Series converter d axis current of VUPFC
1.5
1
0.5
Current (PU) Current I 0 shq I ref shq
−0.5 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (Sec)
Figure 5.13. Shunt converter q axis current of VUPFC
0.2
I 0 kq I ref Kq −0.2
−0.4 Current(PU) −0.6
−0.8 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time(Sec)
Figure 5.14. Series converter q axis current of VUPFC Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 164
30
20
10 Voltage (kV) Voltage Standard UPFC VUPFC 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Time (Sec)
Figure 5.15. Output voltage from UPFCs
30
20
10
MVAR standard UPFC 0 VUPFC −10 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Time (Sec)
Figure 5.16. Reactive power support from UPFC
A comparison study has been made between a standard and the virtual UPFC to verify the performance of the proposed UPFC in the transmission network shown in
Figure 5.15, where red color shows the standard and green color shows the VUPFCs voltage output during the fault and normal condition where the performance of
VUPFC is quite satisfactory with comparison of a standard UPFC
Figure 5.16 and 5.17 illustrate that, although the reactive and real power sup- ports from the VUPFC for the network are of the same level as those from the Section 5.2 Basic Structure of Virtual UPFC (VUPFC) 165
30
20 MVA 10 Standard UPFC VUPFC
0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Time (Sec)
Figure 5.17. Real power support from UPFC
2.5
2
1.5 V ref
Voltage (pu) Voltage dc 1 V dc
0.5 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (Sec)
Figure 5.18. Output DC voltage from VUPFC standard UPFC, their oscillations are much higher. However, by designing proper controllers and filters rather than the simple converter controllers used in this work, this problem could be minimized. In Figure 5.18, evidently, the DC voltage from the PHEV park attains the reference level. Section 5.3 Unified Power Quality Controller 166
5.3 Unified Power Quality Controller
A UPQC is a combination of series and shunt active power filters, which simulta- neously compensates the voltage and current [222], [291], [292]. It can make a sig- nificant contribution to harmonic compensation in a distribution system by being connected close to loads that generate harmonic currents. Although its principal function is to compensate harmonic currents and the imbalances of a non-linear load, it can also compensate harmonic voltages and the imbalances of a power sup- ply. Therefore, to supply improved power quality to harmonically sensitive loads.
The UPQC combines the principles of both shunt current compensation and series voltage compensation in a single device. Two PWM converters, coupled back to back through a common DC link, form a PHEV park and an integrated controller, which provides both voltage and current references for them, are the principal parts of the virtual UPQC (VUPQC) shown in Figure 5.19.
5.3.1 Construction of VUPQC
The VUPQC has three main parts:
1. series and shunt converters
2. VUPQC controller
3. PHEV park as DC link
The basic configuration of a VUPQC is similar to that of a VUPFC but instead of on the network side, to ensure the most effective use of a VUPFC in a microgrid or distribution network, an additional current measurement of the nonlinear load is Section 5.3 Unified Power Quality Controller 167
Vs VL Is IL
Is IL VL Vc Ic Vs Harmonic Sensitive Load
Rse+jXse Rsh+jXsh
Series Shunt Converter Converter
Vdc * * Voltage V c I c Current Control Control Vf If Is UPQC Vs Controller IL
Figure 5.19. Basic configuration of VUPQC needed if a VUPFC and shunt converter of the VUPQC need to be connected close to the load. The series and shunt converters are designed as series and shunt active
filters to work as controlled voltage and current sources respectively. An instanta-
∗ neous algorithm provides the compensating voltage reference, v C and compensating
∗ current reference, i C to be synthesized by the converters.
In a microgrid or distribution network, the source voltage can be unbalanced or distorted. However, as the VUPQC can fulfill the demand of a critical load that requires high power quality to operate, even with an unbalanced source, it is a most powerful compensator. In Figure 5.19, the supply voltage (VS) is already unbalanced Section 5.3 Unified Power Quality Controller 168 or distorted when applied to the load. The total load includes non-linear loads that inject a large amount of harmonic current into the network, which should be filtered and the current, IL, represents the loads that should be compensated. The series and shunt active filters within the VUPQC compensate the harmonic current and perform the following tasks.
• Series Active Filter:
Compensates the source voltage harmonics.
Compensates at the fundamental frequency.
Blocks the harmonic currents flowing to the sources.
Improves system stability.
• Shunt Active Filter:
Compensates the load current harmonics.
Compensates at the fundamental frequency.
Compensates the reactive power of the load.
Regulates the PHEV’s DC-link voltage.
As the VUPQC can simultaneously compensate all undesirable currents and voltages through its series and shunt active filters, it guarantees that both the compensated voltage, VL, at the load terminal and the compensated current, Is, drawn from the power system become balanced. Moreover, with the load reactive power compen- sated, the voltage and current are sinusoidal and in phase. Furthermore, the shunt active filter provides DC-link voltage regulation, absorbing or injecting energy from Section 5.3 Unified Power Quality Controller 169 or into the power distribution system to cover losses in the PWM converters, and correct the ultimate transient compensation errors that lead to unexpected transient power flows into the VUPQC. It might be interesting to design VUPQC controllers that allow compensating functionalities other than those listed above to be selected.
5.3.2 Controller design
To develop a controller, as the UPQC is realized as a combination of series and shunt converter-based FACTS devices with a common DC voltage ,Vdc, we design two converters to control it. In both cases, the p-q theory is used as the base control theory, with the shunt converter controlled by a controlling current run through a hysteresis current controller (the same as that explained in Section 4.7.4) to obtain its switching signal and the series converter controlled by a simple PWM voltage controller with an additional unwanted high-order harmonics voltage, Vf .
To design controllers for VUPQC, a three -phase system is considered without a neural wire and simplification has been achieved through the elimination of all calculations for zero-sequence components, therefore two measurements are sufficient instead of three. When the zero-sequence components are neglected the following relations are valid [293]:
va + vb + vc = 0 (5.34)
ia + ib + ic = 0 (5.35) Section 5.3 Unified Power Quality Controller 170
Under this constraints, the simplified Clark transformation without zero-sequence components is:
vab + vbc + vca = 0 (5.36)
where vab = va − vb, vbc = vb − vc, vca = vc − va. From equation 5.35, the phase voltage can be calculated as:
Va = (vab − vca)/3 (5.37)
Vb = (vbc − vab)/3 (5.38)
Vc = (vca − vbc)/3 (5.39)
Putting vca = −(vab + vbc), from equation 5.36, further simplification can be made and the phase voltage can be calculated as: v 2 1 a v 1 ab v = −1 1 (5.40) b 3 vbc vc −1 −2
Using clark transformation the voltage in the α − β coordinates is:
V √ a v 1 − 1 − 1 αvupqc 2 2 2 = V (5.41) 3 √ √ b 3 − 3 vβvupqc 0 2 2 Vc Section 5.3 Unified Power Quality Controller 171
This voltage can be directly determined from two line voltage as:
√ v 1 2 V αvupqc 2 ab = √ (5.42) 3 3 vβvupqc 0 2 Vbc
For current calculation only two measurements of line current is enough, provid- ing that ic = −(ia + ib), and the equations are: √ 3 isα 2 2 0 Isa = (5.43) √ √ 3 3 isβ 2 3 Isb
The equations for p and q are:
p v v i vupqc α β sα = (5.44) qvupqc −vβ vα isβ
Therefore compensating voltage for the controller are:
v∗ v v p cα 1 α β vupqc = 2 2 (5.45) ∗ vα + vβ v cβ vβ −vα qvupqc
Similarly the compensating current component on α − β axis are: i∗ v v p cα 1 α β vupqc = 2 2 (5.46) ∗ vα + vβ i cβ vβ −vα qvupqc
The inverse Clark transformation in equations (5.45) and 5.46 give the reference voltage and current as: Section 5.3 Unified Power Quality Controller 172
Linear load Vca
Isa ILa
Unbalance / Isb ILb Non linear distorted load supply Vsa VLa Isc ILc
Ica Icb Icc
Vfa Vfb Vfc
LS
Lf
Ifa Ifb Ifc
* * V ca I ca * VUPQC * Current Voltage V cb I cb controller * controller * controller V cc I cc
Figure 5.20. Combined series and shunt active filters for compensating voltage and current using PHEVs Section 5.3 Unified Power Quality Controller 173
The reference current: v∗ 1 −0 ca √ v∗ 2 √ cα ∗ = 1 3 (5.47) v cb 2 2 3 ∗ √ v cβ ∗ − 1 − 3 v cc 2 2
Similarly the reference: i∗ 1 −0 ca √ i∗ 2 √ cα ∗ = 1 3 (5.48) i cb 2 2 3 ∗ √ i cβ ∗ − 1 − 3 i cc 2 2
The designed controller is shown with the total system in Figure 5.20. The instantaneous values of the real and reactive powers are calculated, in the first section of the controller, as in Figure 5.21 and then the current signal is used to switch the inverter as shown in Figure 5.24.
Va Va va ABC Ia vb To pvupqc S a, ß Vß vc Vß
Ia
Vß I i a a Iß qvupqc ABC S To ib Vß a, ß Iß Iß ic
Figure 5.21. pq generation in controller Section 5.3 Unified Power Quality Controller 174
PWM voltage control should allow the series active filter to generate nonsinu- soidal voltages according to their references (v∗Ca, v∗Cb, and v∗Cc), which can vary widely in frequency and amplitude. Therefore, three minor feedback control loops using the actual values of vfa, vfb and vfc are implemented to minimize possible de-
∗ ∗ ∗ viations between the reference values (v Ca, v Cb, and v Cc) and the compensating voltages (vCa, vCb, and vCc) generated at the primary sides of the series transformers.
The gain, Kr, which multiplies the errors between the reference values and actual values of the compensating voltages is set as high as possible considering that the new
∗ ∗ − reference values given to the PWM control equal to vC + Kv(vCk vfk), k = a, b, c, should not exceed the amplitude of the triangular carrier as shown in Figure 5.22.
Va
∆V
Vsa Vfa p Se1 S Va_ref V*ca Ia_ref - - Se2 Kr + + Ia Vß a ß Se3 2 Vb_ref V*cb Vß Ib_ref S To Kr + +
- - PWM Se4 Ib ABC 2 Controller Va Vsb Vfb Vc_ref V*cc Se5 Kr + + q Ic_ref - - S Se6 Vsc Vfc
Va
Figure 5.22. Signal generation for series inverter switching
The principal goal of the series active filter of the UPQC is to compensate har- monics and imbalances in the supply voltage at its left side that is it should com- pensate all voltage components in the supply voltage, which do not correspond to its fundamental positive-sequence component. Section 5.3 Unified Power Quality Controller 175
Vs VL Is IL
Is
IL Vs Harmonic Vc Sensitive Load VL
Ls Lf
Vs Vdc idc Shunt Converter Series Converter R0 Vp
Ip Rp R2
R1 C1 E m im
Vf
∆V PWM Hysteresis Voltage Control Current Control If
Positive sequence DC Voltage Control * Voltage detector * I c V c
Ploss I PQ theory for PQ theory for s series converter shunt converter control V' control Vs IL
Figure 5.23. Converters connection with PHEVs in VUPFC Section 5.3 Unified Power Quality Controller 176
Va
∆V
Ifa p Sh1 S ploss Ia_ref - + Sh2 Ica Sh3 Vß 2 a, ß Vß Ib_ref S To + - Sh4 Icb
2 ABC Hysteresis Va Ifb
Current controller Sh5 + q Ic_ref - S Sh6
Ifc
Va
Figure 5.24. Signal generation for shunt inverter switching
The voltage convention adopted in Figure 5.20, leads to the following relationship:
v V V La sa a v = V + V (5.49) Lb sb b
vLc Vsc Vc
Therefore the power equations are:
The power of the load are:
pload = vLαiLα + vLβiLβ
qload = vLαiLβ − vLβiLα Section 5.3 Unified Power Quality Controller 177
The power of the shunt active filter are:
pshunt = vLαiCα + vLβiCβ
qshunt = vLαiCβ − vLβiCα
The power of the series active filter are:
pseries = vCαiSα + vCβiSβ
qseries = vCαiSβ − vCβiSα
The compensated power are:
p = vLαiSα + vLβiSβ
q = vLαiSβ − vLβiSα
The system shown in Figure 5.23 is considered in the following performance analysis. The rated supply line-to-line voltage is 380 V and as the average switching frequency of the PWM converters should lie between 10 kHz and 15 kHz, it is reasonable to assume that the VUPQC can compensate harmonics up to about 1 kHz. Next, some issues to guide parameter optimization in the power circuit and controller of the VUPQC are addressed.
Loads: one three-phase thyristor converter; firing angle = 150; Idc = 24 A; and commutation inductance= 10 mH. Section 5.3 Unified Power Quality Controller 178
PHEV park: As the VUPQC is designed for a distribution network with a line to line voltage of 380 V, a single PHEV working with the given load is sufficient.
5.3.3 Simulation results
Several simulation cases with different combinations of harmonic pollution and volt- age imbalances in the supply voltage are investigated. Connecting the series and shunt converters of the VUPQC does not disturb the system as they act almost instantaneously after unblocking the PWM converters at their reference value (300
V) while the VUPQC controller is already in operation and has reached the steady state. Figure 5.25 and 5.26 show the compensated voltages and currents with the shunt converter connected at t = 0.1 sec.
400
200
0 Voltage (V) Voltage −200
−400 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 Time (Sec)
Figure 5.25. Connection of shunt converter at time t=0.1 sec (source voltage) Section 5.3 Unified Power Quality Controller 179
40
20
0 Current (A) Current −20
−40 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 Time (Sec)
Figure 5.26. Connection of shunt converter at time t=0.1 sec (source current)
Figure 5.27 and 5.28 show the compensated voltages and currents with successive connections of the shunt converter at t = 0.1 sec and series converter at t = 0.15 sec. It is possible to see spikes in the currents even after the connection of the series active filter, the principal reason for which is that, in this simulation case, the dynamic characteristics of the PHEV are used. However, if a small passive filter was used, this problem could be eliminated.
400
200
0 Voltage (V) Voltage −200
−400 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 Time (Sec)
Figure 5.27. Successive connections of the shunt and series converters at time t=0.1 sec and t=0.15 sec respectively (source voltage) Section 5.3 Unified Power Quality Controller 180
40
20
0 Voltage (V) Voltage −20
−40 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 Time (Sec)
Figure 5.28. Successive connections of the shunt and series converters at times t=0.1 sec and t=0.15 sec respectively (source current)
Figures 5.27 and 5.28, refer to a case, which experiences the following events:
1. connection of the load to the network at 0.04 sec;
2. connecting shunt converter at 0.08 sec; and
3. connecting series converter at 0.15 sec.
Firstly, the load is connected and then, after 0.08 sec, the shunt converter is con- nected, which is, perhaps, the worst transient period for the VUPQC because the
PHEV battery begins to discharge and supply power to the load. Because the series active filter is still disconnected during this period, the source voltage has harmon- ics but, at 0.15 sec, the series converter starts working and immediately reduces them. Figure 5.29 shows the source and load currents after the connection of both the shunt and series converters in which, evidently, both are in phase. Figure 5.30 shows he compensated currents drawn from the PHEV Section 5.4 Chapter Summary 181
40
20
0 Current (A) Current −20
−40 0.2 0.205 0.21 0.215 0.22 0.225 0.23 0.235 0.24 Time (Sec)
Figure 5.29. Source and load current using VUPQC
15
10
5
0
Current (A) Current −5
−10
−15 0.15 0.155 0.16 0.165 0.17 0.175 0.18 Time (Sec)
Figure 5.30. Compensating current of VUPQC
5.4 Chapter Summary
This chapter investigates the performance of dynamic PHEVs as a VUPFC and examines the system currents at different points in the network. To determine the accuracy of the designed system, we compare the output of the VUPFC with that Section 5.4 Chapter Summary 182 of a standard UPFC with the same rating. The results obtained from simulations show that a PHEV can work as a VUPFC to improve power quality.
The performance of dynamic PHEVs as a VUPQC is also investigated in this chapter. To justify the performance of a series and shunt converter connected PHEV, the converters are connected to the system at different times. As the series converter can reduce the voltage harmonics of the source and the shunt converter compensate the load current harmonics, the use of a VUPQC in a distribution system could be a good solution to the power quality problem. Chapter 6
Power Quality Improvement of Distribution Network and Microgrid using V2G
Integrating distributed generation (DG), especially from renewable energy sources, while maintaining power quality in a distribution network are the major challenges for developing a smart grid. In this thesis, the potential of PHEVs in a V2G mode of operation, which provides a low-cost means of improving the power quality of a distribution network, is explained. To demonstrate the performances of PHEVs, a benchmark distribution network is used with medium-voltage DC (MVDC) coupler links. A photovoltaic active filter (PV-AF) system is designed to improve the quality of the PV generator power and a virtual dynamic voltage restorer (DVR) model is designed for the mitigation of a bus voltage under certain conditions is explained.
Several case studies, in which voltage levels are used to monitor the performances of PHEVs and justify their capability to improve power quality in a network, are presented. The simulations undertaken demonstrate that PHEVs have the poten- tial to work as power quality conditioners in a distribution network and improve the islanding operation of microgrids.
183 Section 6.1 Introduction 184
6.1 Introduction
Over the past few decades, interest in distributed power generation and microgrids has increased rapidly mainly because of the need to protect a network from large blackouts, reduce its power losses, integrate renewable and sustainable energy gen- eration in it on a small scale and improve its power quality. However, the sporadic nature of renewable energy sources present a great challenge for power engineers attempting to incorporate generation in a distribution network. For power qual- ity and stability, DG-based and microgrid networks need battery storage which is very expensive. Also, as voltage levels and distributed generator failures and dis- turbances may significantly impact on a distribution network, maintaining power quality is the most important task in a DG-based distribution network. In the early twentieth century, the phenomenon of disturbances in power system equipment and the resulting non-ideal waveforms of the supply voltage was identified as being due mainly to the transformer and rotating machinery although that view is as old as a power network itself [294]. The development of power electronics-based technologies and equipment has brought economic benefits as well as power quality challenges for power system engineers. The increasing amounts of renewable and sustainable energy sources in a power system, especially a distribution network, have created more disturbances and led to greater interest in maintaining power quality. Power quality has become the most used buzzword in the power industry since the late Section 6.1 Introduction 185
1980s but how to define and incorporate it, has met with a great deal of disagree- ment, with various terms, such as voltage quality, voltage level, and current quality, stability, reliability and power factors, proposed as a measure of power quality. The
first of these, voltage quality, is interpreted as a quality of the product delivered by the utility to the customers while current quality is concerned with deviations in the current waveform from the ideal one which relates to what the consumer takes from the utility. Of course, they are strongly related because, if either the voltage or current deviates from the ideal, it is difficult for the other to be ideal. Although a supplementary term for the latter can be the quality of consumption, it is not in common use [295]. In some standards, the term power quality has gained a degree of official status; for example, one of the Standards Coordination Committees (SCCs) of the Institute of Electric and Electronic Engineers (IEEE) in the USA, the SCC 22, has the title Power Quality [296]. Although, within the documents (standards) of the International Electrotechnical Commission (IEC), the international standards- setting organization, the term power quality did not appear officially until 2002, it was in their first draft [297]. Despite the use of different definitions, there is a need to pay considerable attention to the quality of power and seriously address the issue of current and voltage distortions, a major form of which is harmonic distortion.
As far as power quality definitions are concerned, in [294], the authors describe it as an ultimately consumer-driven issue which gives the end user precedence and defines a power quality problem as any power problem manifested in a voltage or Section 6.2 German Pilot Project (GPP) 186 current, or frequency deviations that result in the failure or misoperation of cus- tomer equipment. Moreover, a utility may define power quality as reliability and show statistics that its system is, for example, 99.98% reliable whereas manufac- turers of load equipment may define it as those characteristics of a power system that enable their equipment to work properly. However, these characteristics can be very different depending on the criteria. In this chapter, voltage levels are used to determine power quality.
6.2 German Pilot Project (GPP)
The electricity distribution integrating systems of new generation (EDISON), stor- age and coupling technologies using advanced information and communication sys- tems for dispatch, is a pilot project established by the Federal Ministry for Economy and Technology in Germany to explore the operational characteristics of a distribu- tion microgrid with a large share of renewable sources [298]. The aims of the project are defined as:
1. Developing methods for existing networks to analyze their conformity and relia- bility in terms of large-scale dispersed generation;
2. Establishing a decentralized energy management system (DEMS) to identify the dispatchability of microgrids with a large share of DG and storage; and
3. Identifying the future requirements for the operation of new technologies, such as fuel cells, PHEVs, innovative battery storage units and MVDC couplers. Section 6.2 German Pilot Project (GPP) 187
For this project, a 20 kV distribution network in a rural area is characterized as follows. One part of its network serves a village with 3200 inhabitants with a peak load 3.5 MW and connected to the main network through a 4 km overhead line. The dead-end feeder of the network supplies power to a farm of 200 m from the village border. Direct connection between this two network is not possible be- cause the threshold exceeds the residual currents in both the neutrally compensated networks. In a worst-case scenario the voltage level drops to 95% in the remote location of the network from the feeding substation and its assessment targets are unacceptable, with 1.14 hour of outage and 3.1 MWh energy not served in time.
To overcome this situation several changes are proposed in the project as shown in
Figure 6.1 [299], and the distribution network was equipped with the following com- ponents and CIGRE established this network as benchmark distribution network for further investigations.
A MVDC network for a 2 MVA power transfer between the village and neighboring network as a second supply infeed; a wind power plant of 1.2 MW on a hill; 4 low- voltage networks with 800 kW of batteries; fuel cells for household CHPs of 50 kW el; and PV units of 200 kW distributed in the whole network.
The dispersed generation and battery unit with 2.77 MW power and a MVDC coupler with 2 MVA power transfer capacity in the network, ensure a fully served village during the outage of the feeding line. Section 6.3 Modified Test System 188
Fig. 2. Test network derived from German MV distribution Figure 6.1. German pilot project
6.3 Modified Test System
Recently, distribution networks and microgrids have been studied and tested around the world. The most commonly used guide for the design, operation and integration of distributed resources with electric power systems in the islanding and DG modes is the IEEE Std P1547 [300]. However, there is no generally accepted benchmark Section 6.3 Modified Test System 189 test system for distributed networks and microgrids although some typical micro- grid configurations have been reported. CIGRE Task Force C6.04.02 designed three benchmark networks which are concerned mainly with the network side of DG in- tegration of which the medium-voltage (MV) rural distribution network benchmark derived from a German MV distribution network (German pilot project) is one of them and the system has been simplified by K. Rudion et al in [301]. For the differ- ent case studies, the simplified system is used in this thesis, as shown in Figure 6.2, and studied in the following ways.
Bus 1
Bus 2 Bus 13
Bus 3
Bus 4 Bus 14
Bus 5
Bus 12 Bus 15 Bus 6 Bus 9 Bus 11
Bus 8 Bus 10 MVDC
Bus 7 Rural industrial area
Small City
Figure 6.2. Simplified German pilot project network Section 6.3 Modified Test System 190
• Case 1: No distributed generator is added to the system
• Case 2: Distributed generators are added to the system in the following se-
quence:
a) a squirrel cage induction generator at bus 6 (1.8 MW);
b) solar generators (30kW) are added at bus 9 with a PV-AF (photovoltaic
active filter) system designed using V2G technology;
c) PHEVs at bus 11 as storage devices (400kW) and as virtual DVR for the
network;
d) PHEVs at bus 6 to support wind generation (400kW).
• Case 3: A MVDC link connected between buses 9 and 15 with virtual active
filter at bus 9.
The modified pilot project network is shown in Figure 6.3 Section 6.4 Specification of the Test System 191
Bus 1
PHEV Park Bus 2 Bus 13
Bus 3
Bus 4 Bus 14
Bus 5
Bus 12 Bus 15 Bus 6 Bus 9 Bus 11
Bus 8 Bus 10 MVDC
Bus 7 Rural industrial area
Small City
Figure 6.3. Modified network with virtual controller model using PHEV
6.4 Specification of the Test System
This network supplies a small town and the surrounding rural area. Its rated voltage level is 20 kV supplied from a 110 kV transformer station. Most connections are by cables, with some sections of overhead lines. The network in Figure 6.1 has 30 nodes. To reduce its size to the level required for study while maintaining its realistic character, the number of nodes is reduced and the resulting network proposed as a benchmark is shown in Figures 6.3 and 6.33. It is decomposed into two separate sub-networks, 1 and 2, which are supplied by 110/20 kV transformers referred to as
TR1 and TR2 respectively. The MVDC coupler is between buses 9 and 15 and the Section 6.5 PHEV as Storage and Network Support 192 purpose of sub-network 2 is to study this coupling. The parameters of the network elements used for the case studies are given in Table 6.1.
Table 6.1. Modified GPP distribution system data
Bus to Bus R(Ω/km) X(Ω/km) C(nF/km) L(km) 1-2 — — — — 2-3 0.579 0.367 158.88 2.82 3-4 0.164 0.113 6608 4.42 4-5 0.040 0.040 0.375 0.300 5-6 0.040 0.040 0.000 0.000 6-7 0.110 0.110 1.000 0.675 7-8 0.080 0.110 1.250 0.750 8-9 0.110 0.110 0.150 0.025 9-10 0.080 0.110 0.000 0.000 10-11 0.110 0.110 0.25 0.225 11-12 0.040 0.040 0.000 0.000 12-4 0.040 0.040 0.000 0.000 1-13 0.110 0.110 0.250 0.225 13-14 0.090 0.120 0.250 0.275 13-15 0.080 0.110 0.250 0.275 14-15 0.040 0.040 0.525 0.250
6.5 PHEV as Storage and Network Support
P-Q capability of a realistic PHEV battery is identified in [302] and is shown in
Figure 6.4, where the P-Q capability of a battery vary within ± 138 kW and ± 138 kVA respectively. Section 6.5 PHEV as Storage and Network Support 193
Figure 6.4. Capability of a vehicle batteries
A simple scheme for charging and discharging of PHEVs is designed based on the dynamic equation of the battery and its control and is shown in Figure 6.5
1/Vdc Idcmax SOC
Idc Pref
1 Im 1 Rp 11 + s Ts Tr SOC Idcmin Discharging Charging 11 1T + s Tr
1/(qe+1) Ro
Kp S 1/SOC S
1 - - Ve + + Bqe Ts + V1 Vdc Pdc
Em
Figure 6.5. PHEV battery scheme
According to the pilot project, the maximum requirement for the storage device is 800 kW. Taking into consideration that a normal PHEV has a capacity of 15 kW, Section 6.6 Case1: Distribution Network without DG, PHEV or MVDC 194 the required number of PHEVs in this network is 54 which can represented by a small market car park or a large office car park.
6.6 Case1: Distribution Network without DG, PHEV or MVDC
For the first case study, the network is arranged without any DG and with the load shown in Table 6.2
Table 6.2. Parameters of loads at each bus
Bus No. Pmax [p.u] Qmax [p.u] 2 0.20000 0.04100 4 0.00500 0.00208 5 0.00432 0.00108 6 0.00725 0.00182 7 0.00550 0.00138 8 0.00077 0.00048 9 0.00588 0.00147 10 0.00574 0.00356 11 0.00545 0.00162 12 0.00331 0.00083 13 0.20000 0.04700 14 0.00032 0.00020 15 0.00537 0.00257
The steady-state voltage profile of the distribution test system without DG is shown in Figure 6.6 and it can be seen that all bus voltages are within permissible limits (0.05 pu). Section 6.7 Case2: Integrating DG on Distribution Network 195
Voltage Magnitude Profile 1
0.99
0.98
0.97 Voltage [p.u.] Voltage 0.96
0.95 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Bus No.
Figure 6.6. Voltage profiles without DG, PHEV and/or MVDC coupler
6.7 Case2: Integrating DG on Distribution Network
To observe the impact of DG in a distribution network, a wind generator with a capacity of 1.8 MW is connected at bus 6 and the nodal voltages of the test system are shown in Figure 8.1. It is found that few nodes are below the permissible voltage level due to the integration of a squirrel cage induction type of wind generator.
Voltage Magnitude Profile (with wind generator at bus 6) 1
0.98
0.96 Voltage [p.u.] Voltage 0.94
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Bus No.
Figure 6.7. Voltage profile with wind generator at bus 6
In the next step of DG integration, a solar energy generator with a capacity of
30 kW is installed at bus 9 to identify the impact on the overall network voltage Section 6.7 Case2: Integrating DG on Distribution Network 196 level and it is found that voltage levels at buses 8, 9 and 10 increase while that at bus 6 is still under the acceptable level, as shown in Figure 6.8.
Voltage Magnitude Profile (with wind generator at bus 6 and solar genrator at bus 9) 1
0.98
0.96 Voltage [p.u.] Voltage 0.94
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Bus No.
Figure 6.8. Voltage profile with wind generator at bus 6 and solar generator at bus 9
At this stage, a PHEV park is connected at bus 11 and the voltage levels observed show that they increase at several buses, particularly buses 10, 11and 12, and that at at bus 6 is just above the acceptable limit, as shown in Figure 6.9.
Voltage Magnitude Profile (with wind generator, solar genrator and PHEVs as DG) 1
0.98
0.96 Voltage [p.u.] Voltage 0.94
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Bus No.
Figure 6.9. Voltage profile with wind generator at bus 6, solar generator at bus 9 and PHEV as battery at bus 11
To support the reactive power and improve the voltage level at bus 6, another
PHEV park is added to it and it is found that the voltage levels of all buses increase Section 6.8 Case3: Distribution Network with MVDC Coupler, DG and PHEV 197 and maintain acceptable levels, as shown in Figure 6.10.
Voltage Magnitude Profile 1
0.99
0.98
0.97
0.96
Voltage [p.u.] Voltage 0.95
0.94
0.93 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Bus No.
Figure 6.10. Voltage profile with wind generator at bus 6, solar generator at bus 9 and PHEV as battery at buses 11 and 6
6.8 Case3: Distribution Network with MVDC Coupler, DG and PHEV
For this study, a medium-voltage DC network is installed between buses 9 and 15.
The aim is to support the peak power demand of a city and it is found that, with this connection, the voltage of every bus increases and operates at a higher level, as shown in Figure 6.11
Voltage Magnitude Profile 1
0.99
0.98
0.97 Voltage [p.u.] Voltage 0.96
0.95 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Bus No.
Figure 6.11. Voltage profile with PHEV, DG, and MVDC coupler Section 6.9 Comparisons of Over all Voltage Profiles Comparison of the Distribution Network 198
The connection of a PHEV park and a distribution network improves the voltage level and power quality, the impacts of which are described in the following sections.
6.9 Comparisons of Over all Voltage Profiles Comparison of the Distribution Network
The overall voltage after integrating DG, PHEVs as storage and a MVDC coupler link increases and is maintained at a higher level. Comparisons of the two conditions, one without and one with DG, PHEVs and a MVDC coupler link are shown in
Figure 6.12 and 6.13 in which it is clear that, after integrating them, the overall voltage level of the network is much better.
Figure 6.12. Voltage level without DG, PHEV and MVDC
6.10 PHEV Parks’ Connections with Network
As described before PHEV parks are connected to the network in the following ways.
1. At bus 6 to support the reactive power for the wind generator.
2. At bus 9 as a virtual filter for the solar generator. Section 6.11 Scenario 1: PHEV as Virtual Active Filter for Solar Generation 199
Figure 6.13. Voltage level with DG, PHEV and MVDC
3. At bus 11 as storage and as virtual DVR for the network.
The connections and their controls of the connections are described in the fol- lowing.
6.11 Scenario 1: PHEV as Virtual Active Filter for Solar Generation
This section presents the use of PHEVs with PV sources as an implementation of
V2G technology for designing a PV shunt active filter (PV-AF) system to improve the power quality of PV generation. In order to combine PV sources with PHEVs as an active filter to regulate both active and reactive power injections to the main, a system model with a PV cell and dynamic model of PHEVs is used. The simple battery scheme shown in Figure 6.5 is proposed for the control of the charging and discharging of PHEVs using a power electronic interface. The active filter controller Section 6.11 Scenario 1: PHEV as Virtual Active Filter for Solar Generation 200 is designed based on the p-q theory. Simulation results show that PHEVs have the potential to work as active filters for PV generation for power quality improvement, dynamic power factor correction and harmonics current compensation. The PV generator model and associate controller are used as discussed in Section 2.6.1.
6.11.1 Controller design
The controller design in this work using p-q theory, it consists of an algebraic trans- formation (Clarke transformation) of the three-phase voltages and currents in the a-b-c coordinates to α − β.
Lpv ,Rpv i1 ipv idc L R ias 1 11 iaf + vaf
Cpv L1 Vpv R11 Non ibs ibf Vm linear V dc1 vbf Loads
L R ics 1 11 icf _ vcf * Vpv VPV I pv m ABC Angle To MPPT * Controller Firing Pulse PQ I l a ß Generator Regulation Magnitude
C1 PQ I R2 R0 i Generation m P dc R1 L2 R12 iBa I1 Ip Reference Current L2 R12 calculation iBb Vdc Hystersis E m Current L2 R12 iBc control
N
Figure 6.14. System configuration Section 6.11 Scenario 1: PHEV as Virtual Active Filter for Solar Generation 201
The three phase current from solar PV can be express as follows :
− − ia = [vdc1a vaf R11iaf ]/L1 (6.1)
− − ib = [vdc1b vbf R11ibf ]/L1 (6.2)
− − ic = [vdc1c vcf R11icf ]/L1 (6.3)
Vdc1 = fa1iaf + fb1ibf + fc1icf (6.4)
fa1, fb1, fc1 are the switching functions.
The equations for current in the α − β coordinates are expressed as:
1 0 i √ af i αf 2 √ = − 1 3 i (6.5) 3 2 2 bf iβf √ − 1 − 3 2 2 icf
The voltage in the α − β coordinates is:
1 0 v √ af v αf 2 √ = − 1 3 v (6.6) 3 2 2 bf vβf √ − 1 − 3 2 2 vcf
The equation for p, q is:
p v v i pv αf βf αf = (6.7) qpv −vβf vαf iβf Section 6.11 Scenario 1: PHEV as Virtual Active Filter for Solar Generation 202
To generate the reference current for the controller: i v −v p αref 1 αf βf pv = 2 2 (6.8) vαf + vβf iβref vβf vαf qpv
Based on the p-q theory, a controller is developed and the complete system is shown in Figure 6.14. For the controller, firstly, the instantaneous values of the real and reactive powers are calculated (Figure 6.15) and then the error current signal is used to switch the inverter available (Figure 6.16).
Vaf Vaf vaf ABC Iaf vbf To ppv S a, ß Vßf vcf Vßf
Iaf
Vßf
Iaf iaf Ißf qpv ABC S To Vßf ibf a, ß Ißf Ißf icf
Figure 6.15. pq generation in the controller
Vaf
∆V
Iaf ppv Sh1 S ploss Ia_ref - + Sh2 Ica Sh3 Vßf 2 a, ß Vßf Ib_ref S To + - Sh4 Icb
2 ABC Hysteresis Vaf Ibf
Current controller Sh5 + Ic_ref - qpv S Sh6
Icf
Vaf
Figure 6.16. Signal generation for inverter switching Section 6.11 Scenario 1: PHEV as Virtual Active Filter for Solar Generation 203
The simulation results show that the system and load current harmonics decrease with the use of a PHEV as an active filter (Figures 6.17 and 6.18) and the power factor also improves (Figures 6.21 and 6.20) while the compensating current is presented in Figure 6.19.
Figure 6.17. System and load current without filter Section 6.11 Scenario 1: PHEV as Virtual Active Filter for Solar Generation 204
Figure 6.18. System and load current with filter
Figure 6.19. Compensating current Section 6.11 Scenario 1: PHEV as Virtual Active Filter for Solar Generation 205
Figure 6.20. Load current and voltage without filter Section 6.12 Scenario 2: PHEV as Virtual DVR for Network 206
Figure 6.21. Load current and voltage with filter
6.12 Scenario 2: PHEV as Virtual DVR for Network
DVR are utilized in distribution systems to protect sensitive loads from voltage sags [303]. In this section, V2G technology with a bidirectional converter is used to design a virtual DVR (VDVR) based on the hysteresis voltage control theory.
The potential of a low-cost solution to the power quality problem that utilizes the mitigation capabilities of PHEVs parked in charging stations is investigated. Sim- ulations are performed to show that the proposed VDVR improves power quality while maintaining the voltage level at the user end. In this study, a comprehensive way of utilizing PHEV batteries and their bidirectional chargers in a charging station as VDVRs is demonstrated in the V2G mode of operation. The use of PHEVs as Section 6.12 Scenario 2: PHEV as Virtual DVR for Network 207
VDVRs, including the dynamic behavior of their batteries and series-compensating strategies, is developed and integrated in a real-life low-voltage power system which has not been dealt with in previous studies.
AC is - + iL vDVR + + vs vL - - Load
Line Filter
+ Hysteresis Vdc - Controller Series PHEV Converter
Figure 6.22. Single-line diagram of the VDVR
6.12.1 VDVR design and control
In this section, a low voltage VDVR for sensitive load is designed as shown in
Figure 6.22. An hysteresis current controller is used because it offers an excellent dynamic performance and is very simple to implement in real time. It consists of a comparison between the source voltage, VS, and the tolerance limits (VH ,VL) around the reference voltage ,Vref, and while the source voltage, VS, is between the upper limit and lower limit (VH and VL respectively), no switching occurs but when the out- put voltage crosses the upper limit (lower band) it decrease (increase). Figure 6.23 Section 6.12 Scenario 2: PHEV as Virtual DVR for Network 208 shows the implementation of the hysteresis voltage controller. The reference three- phase voltage signals generated are compared with the three-phase source voltages to generate the switching pulses of the IGBTs in the VSC. In hysteresis control, each phase is regulated independently and the hysteresis band, h, is the difference between VH and VL (h = VH − VL) and inversely proportional to the switching frequency of IGBTs.
Vdvr_c Vdvr_b Vdvr_a
Upper Limit S1 Vsa Ref. Voltage (Vref_a) Vsb Lower Limit S2 Supply Voltage P Vsc (Vsa) S3 S1 S4
Vref_a S4 N Vref_b S2 S5 S5 Vref_c PHEV S6 S6 P S3 N
Figure 6.23. Hysteresis controller for virtual DVR
6.12.2 Simulation results of VDVR
The system shown in Figure 6.22 is simulated for voltage sag and the results are provided in Figures 6.24, 6.25 and 6.26. The voltage sag starts at 1 sec. and ends at 1.1 sec, the simulation mode is a fixed step of 2µ sec, the load is a linear load represented as an R-L load with values of R = 31.84Ω and L = 0.139H. Figure 6.24 Section 6.12 Scenario 2: PHEV as Virtual DVR for Network 209 shows the responses of the system to 30% three phase voltage sag with a +300 phase jump in phase a. Figures 6.25 and 6.26 are the VDVR and load voltages respectively.
400
200
0 Voltage (V) Voltage −200
−400 0.9 0.95 1 1.05 1.1 1.15 1.2 Time(Sec)
Figure 6.24. Supply voltage
300
200
100
0
Voltage (V) Voltage −100
−200
−300 0.9 0.95 1 1.05 1.1 1.15 1.2 Time (Sec)
Figure 6.25. VDVR voltage
400
200
0
Voltage (V) Voltage −200
−400
0.9 0.95 1 1.05 1.1 1.15 Time (Sec)
Figure 6.26. Load voltage Section 6.13 Scenario 3: PHEV for the Improvement in Wind Power Quality 210
The proposed modeling of a VDVR based on hysteresis voltage control using
V2G technology where generation of the hysteresis band and the quality of the load and DVR voltage are studied under voltage sags and the VDVR’s capability to maintain the load voltage under them is verified using time domain simulations.
This model can be further enhanced by using a fixed-frequency hysteresis voltage controller.
6.13 Scenario 3: PHEV for the Improvement in Wind Power Quality
This section presents the use of PHEVs in a wind farm as an implementation of V2G technology for power smoothing in generation systems in which power flow variations can occur. A system model of a wind farm and a dynamic model of PHEVs are used. A feed-forward compensating control strategy is presented to improve the wind farm’s performance. A simple battery scheme with a power electronic interface is used to control the charging and discharging of the PHEVs. The simulations carried out demonstrate that the PHEVs have the potential to combine a wind farm’s output and provide a constant output power. This analysis covers the performance of wind generation with PHEVs, as well as a stability analysis of the power grid, to demonstrate that using PHEVs as a battery energy storage system can improve the overall performance of a wind farm. Section 6.13 Scenario 3: PHEV for the Improvement in Wind Power Quality 211
6.13.1 Wind generator model and PHEV connection
The same wind generator model discussed in section 2.72 is used in this case study and the connection of PHEVs with wind farm is shown in Figure 6.27
DC AC AC DC
AC Network DC Bus PCC PHEV
Figure 6.27. PHEVs connection with wind farm
PHEVs connection with wind generator at the point of common coupling (PCC) is assumed to be realized as a bidirectional converter and a transformer with re- actance xwt as shown in Figure 6.28. AC voltage is regulated by the converter modulating amplitude am, as [200]:
a˙ m = (Km(Vref − Vs − am)/Tm (6.9)
The amplitude control are limited by the boundary conditions as follows:
max min am ≤ am ≤ am
The link with the PCC is: Section 6.13 Scenario 3: PHEV for the Improvement in Wind Power Quality 212
VtwVsw psw = sin(θtw − θsw) (6.10) xwt 2 VtwVsw Vsw qsw = cos(θtw − θsw) − (6.11) xwt xwt
The reference voltage Vref and the initial value of the inverter amplitude am0 are calculated based on the power flow solution: √ 2 xtw 2 Vgw am0 = Pgw + (Qgw + ) (6.12) VswkVdc xtw
Vref = Vgw + amkm (6.13)
where, psw is the real power, qsw is the reactive power, Vsw is the voltage and
θsw is the phase angle at the connecting bus. Vtw and θtw, are the voltage and phase angle before the transformer, Vg is the generator voltage and km is the gain of voltage control loop.
psw+jqsw
Vdc
1:am θ Vsw - sw Vtw - θtw
Figure 6.28. PHEVs’ load connection with power system network Section 6.13 Scenario 3: PHEV for the Improvement in Wind Power Quality 213
6.13.2 Simulation results
A random wind signal is used to test the system’s operation, as shown in Figure 6.29.
To justify the effectiveness of the PHEV’s battery, a computer simulation using
PSCAD is carried out. The wind generator with PHEVs connected at the end of the system is shown in Figure 6.27. To compensate fluctuations of wind farm’s terminal bus voltage and oscillations of the transmission line, PHEVs’ battery are connected to the point of common coupling (PCC) through a three-phase transformer.
Figure 6.29. Wind speed signal during test
As the battery’s characteristic of absorbing or producing reactive power enables control of the wind farm’s power factor and the potential to aid in voltage regulation, the voltage of the PCC remains stable. Figures 6.30 and 6.31 show the use of a
PHEV’s battery to improve the output from the wind farm and help to compensate its reactive power demand. Section 6.13 Scenario 3: PHEV for the Improvement in Wind Power Quality 214
Figure 6.30. Real power output of wind farm
Figure 6.31. Reactive power output of wind farm
To investigate the improvement in transient stability achieved using the battery, a three-phase fault is applied at 5 sec and cleared after 0.18 sec. The simulation results show that the proposed controller is able to stabilize the PCC voltage and absorb the imbalance in the system generated by the induction generator (Figure 6.32). Section 6.14 Distribution Network’s Operation as Microgrid 215
Figure 6.32. Output voltage of wind generator after fault
6.14 Distribution Network’s Operation as Microgrid
In this section, the use of PHEVs in a microgrid as an implementation of V2G tech- nology for power smoothing in a microgrid islanding operation, which is a major challenge, is illustrated by the GPP distribution system operating in the islanding mode with dynamic PHEVs. A control strategy for improving the microgrid’s per- formance during islanding is presented and a simple battery scheme is used to control the charging and discharging of the PHEVs. The simulation results show that the
PHEVs can improve the microgrid’s islanding operation. The implementation of
V2G technology is proposed to improve the quality and stability of a microgrid which has an inherent property of islanding. The GPP distribution network is con- sidered to operate in a microgid mode where the MDVC link and transformer TR2 are disconnected from the network, as shown in Figure 6.33 . As one of the aims of the pilot project was for the system to operate in the islanding mode for only the city area, in this section, using PHEVs to smooth islanding and then provide the support required by the microgrid for its operation, is demonstrated. Section 6.14 Distribution Network’s Operation as Microgrid 216
Bus 1
DC Bus
PHEV Park Bus 2 Bus 13
Bus 3
Bus 4 Bus 14
Bus 5
Bus 12 Bus 15 Bus 6 Bus 9 Bus 11
Bus 8 Bus 10 MVDC
Bus 7 Rural industrial area
Small City
Figure 6.33. Modified GPP network as microgrid
6.14.1 Islanding controller
The objective of the microgrid islanding controller is to maintain the voltage magni- tude in the islanded microgrid and obtain a frequency deviation of zero. To achieve this, it is necessary to begin the relevant control mechanisms and the inverters must
find new voltage and frequency references to maintain good power quality. In this case, the microgrid can be considered as an inverter-dominated system because its frequency is controlled by the power electronics. Processing of signals in two coor- dinate systems is the characteristic feature for the voltage and current controller.
The stationary αβ and then transform into synchronously rotating dq coordinate Section 6.14 Distribution Network’s Operation as Microgrid 217 system combined with PI controllers allows the elimination of steady state error.
The voltage equations in dq reference are as follows:
did vd = Vsd − Rf id − Lf + ωLf iq (6.14) dt diq 0 = Vsq − Rf id − Lf + ωLf id (6.15) dt
The DC bus voltage is set at the reference voltage by controlling the power flow from the PHEV batteries’ charging station while a DC/DC converter plays the role of a voltage stabilizer and the voltage error feeds a PI controller having the duty ratio db. The energy from the PHEV batteries is used when renewable sources are insufficient and the batteries are recharged when there is an excess of energy.
The frequency and magnitude of the voltage is set by the self-commuted inverter.
Figure 6.34 shows the inverter control scheme for a grid-connected operation.
6.14.2 Simulation results
To investigate the improvement in the islanding operation of the microgrid using
PHEV batteries, the main grid line is disconnected from the microgrid at 2 sec. The simulation results show that the proposed controller is able to stabilize the DC bus voltage and absorb the system’s imbalance faster than under the previous condition, as shown in Figures 6.35 and 6.36 Section 6.14 Distribution Network’s Operation as Microgrid 218
db
Vd PI V* * d I d + + 1 1 VDC PI PI S - - 1 + s Tr - 1 + s Tr +
* V DC * * * Id V V V V* V* V* ωLf sd dq sα ab sa a b c Sa * V sb Sb * * * ωLf V sq V sβ V sc Sc Iq ab abc Sa Sb Sc * V q * + + 1 I q - PI PI1 S - 1 + s Tr 1 + s Tr V q Id Ia ab abc Ib Iq I dq ab c
ω
Vd Va Ia Ib Ic ab abc Vb
Vq dq ab Vc
Bus 2
Figure 6.34. Microgrid control scheme for Islanding opeation
Figure 6.35. Voltage at bus 2 during islanding operation without PHEV as battery base Section 6.15 Chapter Summary 219
Figure 6.36. Voltage at bus 2 during islanding operation with PHEV as battery base
6.15 Chapter Summary
In this chapter, a benchmark distribution network designed for a rural German area is used to identify the performances of PHEVs in improving power quality in terms of voltage levels in the V2G mode of operation. A PV-AF system is designed to improve the PV power quality, and the real and reactive power variations as well as the transient stability of the wind power system are investigated. Also, a virtual DVR is designed to mitigate voltage sag for the network and using the
PHEV park as battery-based reactive power support for wind energy generation is justified. The distribution network is converted to a microgrid to observe the PHEV park’s performance during islanding and, in all cases, it is found that PHEVs have the capability to improve the power quality of both a distribution network and microgrid . Chapter 7
Conclusions
This chapter summarizes the findings from this research study, presents the con- clusions drawn and discusses proposed future areas of research. This work will become more relevant in a smart and distributed generation based grid environment as developed countries look to become energy independent and more conscious of the negative environmental impacts of increasing emissions from conventional power plants.
This dissertation is just the first step towards the vision of V2G technology without affecting the battery life in a distributed generation based grid and even for the transmission line. It should be noted that in this thesis FACTS devices and
filters are modeled at a high level of abstraction that relies on simple analytical models that are essentially identical to those used in the study of V2G schemes for congestion control in smart networks. Our future work should aim to refine and extend the current model, taking into account many realistic characteristics of the electric power grid.
As the penetration levels of PHEVs in power systems increase, more research is required to know the impact and taking benefits from them. Also when sufficient data for actual PHEV in the production becomes available, the developed prediction
220 Section Conclusions 221 tools and the PHEV model can be further examined and analyzed for the prediction of large-scale penetration. As PHEVs are characterized by their variability and uncertainty, the integration of PHEVs facilities into utility grids has several impacts on their optimum power flow, transmission congestion, power quality issues, system stability, load dispatch, protection system, economic analysis and electricity market clearing prices. These impacts present major challenges to power system operators.
This thesis tackles some of these challenges.
This dissertation presents several case studies for capturing the impact of PHEVs penetration into the grid caused by the dynamics of the battery load. The case studies conducted are on: (i) voltage instability; and (ii) the effects on different generation such as wind generation, solar generation and a distribution network with renewable energy sources. The devices considered in this thesis are synchronous generators, induction generators (IGs), exciters, HVDC network, PHEV as load and sources, bidirectional charger, FACTS devices and filters. The effect of PHEVs integration into power systems have been investigated by modal analysis as well as by detailed nonlinear simulations.
The central contribution of this dissertation is the design of active filter with simple control technique using V2G technology, which enhance voltage stability and improve power quality by reducing harmonics. This is achieved by reformulating the PQ power theory of filter. Section Conclusions 222
The performances of the proposed filter are validated through simulations. Dif- ferent test power systems are selected and controllers are designed for them. Differ- ent of simulation cases are conducted, which include load and generation changes and fault conditions. The test systems considered here include: (i) a simple but representative single wind generator system; (ii) a CIGRE benchmark HVDC net- work; and (iii) a three phase solar system in a distribution network. Performances of the proposed filter are also compared with IEEE standards for power quality. The simulation results show that the proposed filters are capable of providing better responses during normal and abnormal power system operating conditions.
The next contribution is the design of FACTS devices with the implementation of V2G technology. The FACTS devices are sometimes essential for power system but the main disadvantages of these devices is the cost, mainly for its capacitor and switching devices. In this thesis a comprehensive way of designing FACTS devices are explained using PHEVs, which will act as virtual FACTS devices and power quality conditioner. To make an economic FACTS devices solution, we have designed (i) Virtual Unified Power Flow Controller (VUPFC); (ii) Virtual Dynamic
Voltage Restorer (VDVR); and (iii) for power quality Virtual Unified Power Quality
Conditioner (VUPQC). Simulations have been done and demonstrated that PHEVs have the potential to work as Virtual FACTS devices. These works could be the guideline for the future researcher in the way of implementing V2G technology in smart grid. Section 7.1 Directions for Future Research 223
From this work, the following conclusions can be drawn:
• The dynamic nature of PHEV load has been used to analyze the impact of
PHEV in a smart grid.
• Integration of PHEVs can affect voltage profile and cause voltage unbalance
problem for renewable energy based generation and distribution networks.
• Power quality of renewable energy based generation and even traditional gen-
eration can be improved by using the virtual filter with V2G technology.
• Virtual FACTS can be designed using a single vehicle or even with the fleet.
• Reactive power support is possible from PHEVs whenever necessary.
• HVDC network can use the fleet as the filter and the HVDC converter terminal
can be a PHEV charging station.
• A low-voltage distribution network can be more stable and the power quality
of that network can be improved using PHEVs in V2G mode of operation.
• Microgrid operation can be easier with PHEVs as a source of energy during
islanding.
7.1 Directions for Future Research
Although this research achieved promising results in analyzing dynamic PHEVs impact and utilization in power systems, the work does not end here. It is the starting point of a new era of using PHEV in various ways, which will have less effect on vehicle battery life. The proposed power system design method may be further improved and consolidated by the following processes: Section 7.1 Directions for Future Research 224
i) Implementing the proposed controllers in a real power system will provide
more confidence in the proposed method.
ii) The demand side management for PHEVs can be included in future research.
iii) To design the FACTS devices we have used a simple control system for switch-
ing circuit. To make the system more stable and useful design robust controller
for those proposed virtual FACTS could add a new area under power system
engineering.
iv) The V2G technology can be an economic solution for the power system en-
gineer to design the FACTS and filter device, however the actual figure of
economic benefit was not with the bound of this research. In future, the
quantification of economic benefits from the virtual devices could be added. Chapter 8
Appendices
8.1 Appendix-I
Battery, SMIB and Induction Generator’s Parameters:
The parameters used for the Battery are as follows:
Parameters referring to the battery capacity:
⋆ 0 I = 49 A, KC = 1.18, C1 = 261.9 Ah,Θf = −40 C
Parameters referring to the main branch of the electric equivalent:
−3 0 Ts = 28800 s, Em0 = 2.135 V , Ke = 0.580e V/ C, A0 = −0.30, R00 = 2.0 mΩ,
R10 = 0.4 mΩ
Parameters referring to the parasitic reaction branch of the electric equivalent:
Ep = 1.95 V , Vpo = 0.1 V , Ap = 2.0, Vpo = 0.1 V , Gpo = 2 pS
Parameters referring to the battery thermal model :
0 0 Cθ = 15 W h/ C, Rθ = 0.2 C/W
SMIB Parameters:
Synchronous generator parameters
′ ′ Xd = 2.1 pu,Xd = 0.4 pu, H = 3.5 s, Tdo = 8 s, D = 4.
Automatic voltage regulator (AVR) parameters
225 Section 8.2 Appendix-II 226
KA = 50, Tr = 0.1 s.
Transformer parameter
XT = 0.016 pu
Infinite bus voltage
Vinf = 1.0
The parameters used for the Induction Generator are:
Rs = 0.012 pu, Xs = 0.074 pu, Xm = 2.76 pu, Rr = 0.008 pu, Xr = 0.1761 pu,
Hm = 2.5 s, HG = 0.22 s
8.2 Appendix-II
Element of State Matrix: PHEV with SMIB We have calculated all the elements of
A. The important elements for stability studies are given below:
a11 = a13 = a16 = 0
−ωs0 ′ Vinf cos δ0 Vdc0 cos δ0 a21 = Eq0[ ′ + ′ ] 2H Xd + XT + Xe Xd + XT
ω a = − s0 I 23 2H qg0
−ωs0 ′ Vdc0 sin δ0 a26 = Eq0 ′ 2H Xd + XT Section 8.2 Appendix-II 227
′ − Xd Xd Vinf sin δ0 Vdc0 sin δ0 a31 = ( ′ )[ ′ + ′ ] Tdo Xd + XT + Xe Xd + XT
E′ X′ − X 1 q0 d d − a33 = ′ ′ ′ Xd Tdo Tdo
X′ − X E′ cos δ − d d m0 0 a36 = ( ′ ) ′ Tdo Xd + XT
′ ′ − ′2 Eq0Xd Xd Idg0 Vinf sin δ0 Vdc0 sin δ0 a41 = ( )[ ′ + ′ ] TrVt0 Xd + XT + Xe Xd + XT
′2 Xd Iqg0 Vinf cos δ0 Vdc0 cos δ0 + ( )[ ′ + ′ ] TrVt0 Xd + XT + Xe Xd + XT
′ − ′ ′ ′ − ′2 ′ Eq0 XdIdg0 Eq0Xd Xd Idg0 Eq0 a43 = + ′ TrVt0 TrVt0 Xd
′ ′ ′2 ′2 E X − X Idg0 V cos δ X I V sin δ − q0 d d dc0 0 − d dg0 dc0 0 a45 = [ ′ ′ ] TrVt0 Xd + XT TrVt0 Xd + XT
− ′ E′ cos(δ − δ ) −X X q0 0 m0 a63 = ′ ′ Tdom Xd + XT
′ − 1 Em0 a66 = ( ′ + ′ ) Tdom X Section 8.3 Appendix-III 228
′ ′ − X − X Eq0 cos(δ0 δm0) a71 = ′ ′ ′ TdomEm0 Xd + XT
′ ′ − X − X Eq0 sin(δ0 δm0) a73 = ′ ′ ′ TdomEm0 Xd + XT
′ X − X Iqm0 a76 = ′ ′ ′ TdomEm0 Em0
8.3 Appendix-III
Figure 8.1. Dynamic PHEV battery design in PSCAD References
[1] Position Statement: Plug-in Electric Hybrid Vehicles. IEEE-
USA Board of Directors. 15 June 2007. [online] Available:
www.ieeeusa.org/policy/positions/phev0607.pdf.
[2] Plug-in hybrid electric vehicles. [online] Avaiable: www.climatelab.org.
[3] EPRI. Environmental assessment of plug-in hybrid electric vehicles, nation-
wide greenhouse gas emissions. Technical Report Vol.1, July 2007.
[4] Sherry Boschert. Plug-in Hybrids: The Cars that will Recharge America. New
Society Publishers, Canada, 2006.
[5] Xiaohu Zhou. Design and Control of Bi-Directional Grid-Interactive Con-
verter for Plug-in Hybrid Electric Vehicle Applications. PhD thesis, Graduate
Faculty, North Carolina State University, 2011.
[6] Smart grids european technology platform. [online] Available:
www.smartgrids.eu.
[7] U.S. Department of Energy. Smart grid / department of energy. [online]
Available: http://energy.gov/oe/technology-development/smart-grid.
[8] S. Massoud Amin and B.F. Wollenberg. Toward a smart grid: power delivery
for the 21st century. IEEE Power and Energy Magazine, 3(5):34 – 41, sept.-
oct. 2005.
[9] Ausgrid. What is a Smart Grid? [online] Available:
229 References 230
www.smartgridsmartcity.com.au.
[10] E. Keane and D. Flynn. Potential for electric vehicles to provide power system
reserve. In IEEE Innovative Smart Grid Technologies (ISGT), pages 1 –7, Jan.
2012.
[11] S.P. Chowdhury S. Chowdhury and P. Crossley. Microgrids and Active Dis-
tribution Networks. The Institution of Engineering and Technology, 2009.
[12] Department of Energy, USA. Enhancing the smart grid: Integrating clean
distributed and renewable generation.
[13] S. E. Gay M. Ehsani, Y. Gao and Modern Electric A. Emadi. Modern Electric,
Hybrid Electric, and Fuel Cell Vehicles. Boca Raton, FL: CRC Press, 2005.
[14] J. Larminie and J. Lowry. Electric Vehicle Technology Explained. . New York:
John Wiley, 2003.
[15] S.G. Wirasingha and A. Emadi. Pihef: Plug-in hybrid electric factor. IEEE
Transactions on Vehicular Technology,, 60(3):1279 –1284, Mar. 2011.
[16] Benjamin K. Sovacool and Richard F. Hirsh. Beyond batteries: An examina-
tion of the benefits and barriers to plug-in hybrid electric vehicles (PHEVs)
and a vehicle-to-grid (V2G) transition. Energy Policy, 37(3):1095 – 1103, 2009.
[17] M. Duvall and E. Knipping. Environmental assessment of plug-in hybrid elec-
tric vehicles. Final Report pp. 156, Nationwide Greenhouse Gas Emissions,
EPRI and NRDC, 2007 2007.
[18] Willett Kempton and Jasna Tomi´c. Vehicle-to-grid power implementation: References 231
From stabilizing the grid to supporting large-scale renewable energy. Journal
of Power Sources, 144(1):280–294, 2005.
[19] SAE electric vehicle and plug-in hybrid electric vehicle conductive charge cou-
pler, SAE standard j1772, January 2010.
[20] J. Lassila, J. Haakana, V. Tikka, and J. Partanen. Methodology to analyze
the economic effects of electric cars as energy storages. IEEE Transactions on
Smart Grid,, 3(1):506 –516, Mar. 2012.
[21] Gan Li and Xiao-Ping Zhang. Modeling of plug-in hybrid electric vehicle
charging demand in probabilistic power flow calculations. IEEE Transactions
on Smart Grid, 3(1):492 –499, Mar. 2012.
[22] GM-Volt: Latest chevy volt battery pack and generator details and clarifica-
tions 2011.
[23] Nissan zero emission website, leaf specs, 2010. [online] Available: www.nissan-
zeroemission.com/EN/LEAF/specs.html.
[24] G. T. Heydt. The impact of electric vehicle deployment on load management
strategies. IEEE Power Engineering Review, PER 3(5):41 –42, May 1983.
[25] S. Meliopoulos, J. Meisel, G. Cokkinides, and T. Overbye. Power system level
impacts of plug-in hybrid vehicles. Technical report, PSERC Document 09-12,
2009.
[26] H. Lund and W. Kempton. Integration of renewable energy into the transport
and electricity sectors through V2G. Energy Policy, 36(9):3578 – 3587, 2008. References 232
[27] Craig H. Stephan and John Sullivan. Environmental and energy implica-
tions of plug-in hybrid-electric vehicles. Environmental Science & Technology,
42(4):1185–1190, 2008.
[28] Ramteen Sioshansi and Paul Denholm. Emissions impacts and benefits of
plug-in hybrid electric vehicles and vehicle-to-grid services. Environmental
Science & Technology, 43(4):1199–1204, 2009.
[29] Constantine Samaras and Kyle Meisterling. Life cycle assessment of green-
house gas emissions from plug-in hybrid vehicles: Implications for policy. En-
vironmental Science & Technology, 42(9):3170–3176, 2008.
[30] S. W. Hadley. Impact of plug-in hybrid vehicles on the electric grid. Technical
report, ORNL, October 2006.
[31] K. Clement-Nyns, E. Haesen, and J. Driesen. The impact of charging plug-in
hybrid electric vehicles on a residential distribution grid. IEEE Transactions
on Power Systems, 25(1):371 –380, Feb. 2010.
[32] E. Larsen, D.K. Chandrashekhara, and J. Ostergard. Electric vehicles for
improved operation of power systems with high wind power penetration. In
IEEE Energy 2030 Conference, pages 1 –6, nov. 2008.
[33] F. Locment, M. Sechilariu, and C. Forgez. Electric vehicle charging system
with pv grid-connected configuration. In IEEE Vehicle Power and Propulsion
Conference (VPPC), pages 1 –6, sept. 2010.
[34] Y. Gurkaynak and A. Khaligh. Control and power management of a grid References 233
connected residential photovoltaic system with plug-in hybrid electric vehi-
cle (phev) load. In Twenty-Fourth Annual IEEE Applied Power Electronics
Conference and Exposition, 2009, pages 2086 –2091, Feb. 2009.
[35] Willett Kempton and Jasna Tomi. Vehicle-to-grid power fundamentals: Cal-
culating capacity and net revenue. Journal of Power Sources, 144(1):268 –
279, 2005.
[36] B. Singh, B.N. Singh, A. Chandra, K. Al-Haddad, A. Pandey, and D.P.
Kothari. A review of single-phase improved power quality AC-DC convert-
ers. IEEE Transactions on Industrial Electronics, 50(5):962 – 981, Oct. 2003.
[37] B. Singh, B.N. Singh, A. Chandra, K. Al-Haddad, A. Pandey, and D.P.
Kothari. A review of three-phase improved power quality AC-DC convert-
ers. IEEE Transactions on Industrial Electronics, 51(3):641 – 660, June 2004.
[38] X. Zhou, S. Lukic, S. Bhattacharya, and A. Huang. Design and control of grid-
connected converter in bi-directional battery charger for plug-in hybrid electric
vehicle application. In IEEE Vehicle Power and Propulsion Conference, 2009,
pages 1716 –1721, Sept. 2009.
[39] Xiaohu Zhou, Gangyao Wang, S. Lukic, S. Bhattacharya, and A. Huang.
Multi-function bi-directional battery charger for plug-in hybrid electric vehicle
application. In IEEE Energy Conversion Congress and Exposition, pages 3930
–3936, Sept. 2009.
[40] S. De Breucker, P. Jacqmaer, K. De Brabandere, J. Driesen, and R. Belmans. References 234
Grid power quality improvements using grid-coupled hybrid electric vehicles.
In The 3rd IET International Conference on Power Electronics, Machines and
Drives, pages 505 –509, Mar. 2006.
[41] Ahmed Yousuf Saber and Ganesh Kumar Venayagamoorthy. Intelligent unit
commitment with vehicle-to-grid a cost-emission optimization. Journal of
Power Sources, 195(3):898 – 911, 2010.
[42] Christophe Guille and George Gross. A conceptual framework for the vehicle-
to-grid V2G implementation. Energy Policy, 37(11):4379–4390, 2009.
[43] P. Mitra, G.K. Venayagamoorthy, and K.A. Corzine. Smartpark as a virtual
STATCOM. IEEE Transaction on Smart Grid, 2(3):445 –455, Sept. 2011.
[44] V. Marano and G. Rizzoni. Energy and economic evaluation of PHEVs and
their interaction with renewable energy sources and the power grid. In IEEE
International Conference on Vehicular Electronics and Safety, (ICVES), pages
84 –89, Sept. 2008.
[45] W. Short and P. Denholm. Preliminary assessment of plug-in hybrid electric
vehicles on wind energy markets. Technical report, Tech. Report National
Renewable Energy Lab NREL/TP-620-39729, April 2006.
[46] J. M. Latorre A. Ramos, L. Olmos and I. J. Perez-Arriaga. Modeling medium
term hydroelectric system operation with large-scale penetration of intermit-
tent generation. In XIV Latin and Iberian Conf. Operations Research, 2008. References 235
[47] F. Koyanagi and Y. Uriu. A strategy of load leveling by charging and discharg-
ing time control of electric vehicles. IEEE Transactions on Power Systems,
13(3):1179 –1184, Aug. 1998.
[48] M. Takagi, Y. Iwafune, K. Yamaji, H. Yamamoto, K. Okano, R. Hiwatari, and
T. Ikeya. Electricity pricing for PHEV bottom charge in daily load curve based
on variation method. In IEEE Innovative Smart Grid Technologies (ISGT),
pages 1 –6, jan. 2012.
[49] K. P. Schneider M. Kintner-Meyer and R. G. Pratt. Impacts assessment of
plug-in hybrid vehicles on electric utilities and regional US power grids part
1: Technical analysis pacific northwest national laboratory. Technical report,
Pacific Northwest National Lab. Tech. Rep., PNNL-SA-61669, 2007.
[50] E. Sortomme and M.A. El-Sharkawi. Optimal scheduling of vehicle-to-grid
energy and ancillary services. IEEE Transactions on Smart Grid,, 3(1):351
–359, Mar. 2012.
[51] W. Kempton, J. Tomic, A. Brooks S. Letendre, and T. Lipman. Vehicle-to-
grid power: Battery, hybrid, and fuel cell vehicles as resources for distributed
electric power in california. Technical report, Tech. Rep. California Air Re-
sources Board and California Environmental Protection Agency, CEPA, Los
Angeles, CA, Res. Rep. UCD-ITS-RR-01-03, 2001.
[52] S.G. Wirasingha, N. Schofield, and A. Emadi. Plug-in hybrid electric vehicle
developments in the US: Trends, barriers, and economic feasibility. In IEEE References 236
Vehicle Power and Propulsion Conference, pages 1 –8, Sept. 2008.
[53] D. Dallinger, D. Krampe, and M. Wietschel. Vehicle-to-grid regulation reserves
based on a dynamic simulation of mobility behavior. IEEE Transactions on
Smart Grid, 2(2):302 –313, June 2011.
[54] Anurag K. Srivastava, Bharath Annabathina, and Sukumar Kamalasadan.
The challenges and policy options for integrating plug-in hybrid electric vehicle
into the electric grid. The Electricity Journal, 23(3):83 – 91, 2010.
[55] P. Denholm and W. Short. An evaluation of utility system impacts and benefits
of optimally dispatched PHEVs. Technical report, Tech. Report National
Renewable Energy Lab NREL/TP- 620-40293, October 2006.
[56] S.-L. Andersson, A.K. Elofsson, M.D. Galus, L. Gransson, S. Karlsson,
F. Johnsson, and G. Andersson. Plug-in hybrid electric vehicles as regulat-
ing power providers: Case studies of Sweden and Germany. Energy Policy,
38(6):2751 – 2762, 2010.
[57] C. Camus, J. Esteves, and T.L. Farias. Electric vehicles and electricity sector
regulatory framework: The portuguese example. In EVS24, 2009.
[58] R. Sioshansi and P. Denkolm. The value of plug-in hybrid electric vehicles as
grid resources. Energy Journa, 31(3):1–23, 2010.
[59] W. Shireen and S. Patel. Plug-in hybrid electric vehicles in the smart grid envi-
ronment. In IEEE Transmission and Distribution Conference and Exposition,
pages 1 –4, Apr. 2010. References 237
[60] M. Grahn, C. Azar, M. I. Williander, J. E. Anderson, S. A. Mueller, and
T. J. Wallington. Fuel and vehicle technology choices for passenger vehicles
in achieving stringent CO2 targets: Connections between transportation and
other energy sectors. Environmental Science & Technology, 43(9):3365–3371,
2009.
[61] S. Karlsson L. Goransson and F. Johnsson. Plug-in hybrid electric vehicles as
a mean to reduce CO2 emissions from electricity production. In in Proceding
Electric Vehicle Sympsium 24, Stavanger, Norway, May 2009.
[62] Scott B. Peterson, J.F. Whitacre, and Jay Apt. The economics of using plug-
in hybrid electric vehicle battery packs for grid storage. Journal of Power
Sources, 195(8):2377 – 2384, 2010.
[63] A. Zahedi. Electric vehicle as distributed energy storage resource for future
smart grid. In 22nd Australasian Universities Power Engineering Conference
(AUPEC), pages 1 –4, sept. 2012.
[64] Introduction to Vehicle to Grid (V2G). [online] Avaiable:
http://www.v2g.com.au/.
[65] C. Pang, P. Dutta, S. Kim, M. Kezunovic, and I. Damnjanovic. PHEVs as
dynamically configurable dispersed energy storage for V2B uses in the smart
grid. In 7th Mediterranean Conference and Exhibition on Power Generation,
Transmission, Distribution and Energy Conversion (MedPower 2010), pages
1 –6, Nov. 2010. References 238
[66] Wencong Su, H. Eichi, Wente Zeng, and Mo-Yuen Chow. A survey on the elec-
trification of transportation in a smart grid environment. IEEE Transactions
on Industrial Informatics, 8(1):1 –10, Feb. 2012.
[67] M. Yilmaz and P.T. Krein. Review of benefits and challenges of vehicle-to-grid
technology. In IEEE Energy Conversion Congress and Exposition (ECCE),
pages 3082 –3089, Sept. 2012.
[68] J. DiPeso. Cars to grid: An electrifying idea. Environmental Quality Manage-
ment, vol. 18(no. 2):89 – 94, 2008.
[69] H. Turker, S. Bacha, D. Chatroux, and A. Hably. Aging rate of low voltage
transformer for a high penetration of plug-in hybrid electric vehicles (PHEVs).
In IEEE Innovative Smart Grid Technologies (ISGT), pages 1 –8, Jan. 2012.
[70] A.Y. Saber and G.K. Venayagamoorthy. Plug-in vehicles and renewable energy
sources for cost and emission reductions. IEEE Transactions on Industrial
Electronics,, 58(4):1229 –1238, Apr. 2011.
[71] plug-in electric vehicles: A practical plan for progress, the report of an expert
panel, school of public and environmental affairs at indiana university, feb.
2011. Technical report, School of Public and Environmental Affairs at Indiana
University, Feb 2011.
[72] Electrification of the transportation system. Technical report, MIT Energy
Initiative Symp., MITei USA, April 2010.
[73] Jasna Tomicand Willett Kempton. Using fleets of electric-drive vehicles for References 239
grid support. Journal of Power Sources, 168(2):459 – 468, 2007.
[74] D. Rua, D. Issicaba, F.J. Soares, P.M.R. Almeida, R.J. Rei, and J.A.P. Lopes.
Advanced metering infrastructure functionalities for electric mobility. In In-
novative Smart Grid Technologies Conference Europe (ISGT Europe), 2010
IEEE PES, pages 1 –7, Oct. 2010.
[75] Casey Quinn, Daniel Zimmerle, and Thomas H. Bradley. The effect of commu-
nication architecture on the availability, reliability, and economics of plug-in
hybrid electric vehicle-to-grid ancillary services. Journal of Power Sources,
195(5):1500 – 1509, 2010.
[76] W. Kempton and S. Letendre. Electric vehicles as a new power source for
electric utilities. Transport. Res. D, vol. 2(3):157–175, 1997.
[77] Kristien Clement-Nyns, Edwin Haesen, and Johan Driesen. The impact of
vehicle-to-grid on the distribution grid. Electric Power Systems Research,
81(1):185 – 192, 2011.
[78] T. Markel, M. Kuss, and P. Denholm. Communication and control of electric
drive vehicles supporting renewables. In IEEE Vehicle Power and Propulsion
Conference, 2009, pages 27 –34, Sept. 2009.
[79] Wencong Su, Wente Zeng, and Mo-Yuen Chow. A digital testbed for a
PHEV/PEV enabled parking lot in a smart grid environment. In Innovative
Smart Grid Technologies (ISGT), 2012 IEEE PES, pages 1 –7, Jan. 2012.
[80] H. Chaudhry and T. Bohn. Security concerns of a plug-in vehicle. In IEEE References 240
Innovative Smart Grid Technologies (ISGT), pages 1 –6, Jan. 2012.
[81] Zigbee smart energy overview. Technical report, 2011. [Online]. Available:
http://www.zigbee.org/Standards/ZigBeeSmartEnergy/Overview.aspx.
[82] P. Kulshrestha, K. Swaminathan, Mo-Yuen Chow, and S. Lukic. Evaluation
of zigbee communication platform for controlling the charging of phevs at a
municipal parking deck. In IEEE Vehicle Power and Propulsion Conference,
pages 1211 –1214, sept. 2009.
[83] H. A. Latchman S. Katar M. K. Lee, R. E. Newman and L. Yonge. Home-
plug 1.0 powerline communication lansprotocol description and performance
results. International Journal of Communication Systems, Vol.16:447–473,
2003.
[84] Communication between plug-in vehicles and the utility grid, May 2011.
[85] Communication between plug-in vehicles and off-board dc chargers, 2011.
[86] Communication between plug-in vehicles and the utility grid for reverse power
flow, 2011.
[87] V.C. Gungor, D. Sahin, T. Kocak, S. Ergut, C. Buccella, C. Cecati, and G.P.
Hancke. Smart grid technologies: Communication technologies and standards.
IEEE Transactions on Industrial Informatics, 7(4):529 –539, Nov. 2011.
[88] J.C. Ferreira, V. Monteiro, J.L. Afonso, and A. Silva. Smart electric vehicle
charging system. In IEEE Intelligent Vehicles Symposium (IV), pages 758
–763, June 2011. References 241
[89] S. Bai and S. Lukic. Design considerations for dc charging station for plug-in
vehicles. In IEEE Vehicle Power and Propulsion Conference (VPPC), 2011,
pages 1 –6, Sept. 2011.
[90] Wilsun Xu. Comparisons and comments on harmonic standards IEC 1000-3-6
and IEEE Std. 519. In Harmonics and Quality of Power, 2000. Proceedings.
Ninth International Conference on, volume 1, pages 260 –263 vol.1, 2000.
[91] P. Komarnicki M. Geske, T. Winkler and G. Heideck. Controlled battery
charger for electric vehicles. PIERS Online, vol. 6(6):532536, 2010.
[92] Power quality requirements for plug-in vehicle chargers - part 1: Requirements,
2011.
[93] C.C. Chan and K.T. Chau. An overview of power electronics in electric vehi-
cles. IEEE Transactions on Industrial Electronics, 44(1):3 –13, Feb 1997.
[94] C. Aguilar, F. Canales, J. Arau, J. Sebastian, and J. Uceda. An integrated
battery charger/discharger with power factor correction. In 26th Annual IEEE
Power Electronics Specialists Conference, 1995, volume 2, pages 714 –719
vol.2, Jun 1995.
[95] Young-Joo Lee, A. Khaligh, and A. Emadi. Advanced integrated bidirectional
ac/dc and dc/dc converter for plug-in hybrid electric vehicles. IEEE Transac-
tions on Vehicular Technology, 58(8):3970 –3980, Oct. 2009.
[96] J.C. Gomez and M.M. Morcos. Impact of EV battery chargers on the References 242
power quality of distribution systems. IEEE Transactions on Power Deliv-
ery, 18(3):975 – 981, July 2003.
[97] D. Aggeler, F. Canales, H. Zelaya-De La Parra, A. Coccia, N. Butcher, and
O. Apeldoorn. Ultra-fast dc-charge infrastructures for ev-mobility and future
smart grids. In IEEE Innovative Smart Grid Technologies Conference Europe
(ISGT Europe), pages 1 –8, Oct. 2010.
[98] A. Aabrandt, P.B. Andersen, A.B. Pedersen, Shi You, B. Poulsen,
N. O’Connell, and J. Ostergaard. Prediction and optimization methods for
electric vehicle charging schedules in the edison project. In IEEE Innovative
Smart Grid Technologies (ISGT), pages 1 –7, Jan. 2012.
[99] D.P. Tuttle and R. Baldick. The evolution of plug-in electric vehicle-grid
interactions. IEEE Transactions on Smart Grid, 3(1):500 –505, Mar. 2012.
[100] R.J. Bessa, M.A. Matos, F.J. Soares, and J.A.P. Lopes. Optimized bidding of
a ev aggregation agent in the electricity market. IEEE Transactions on Smart
Grid, 3(1):443 –452, Mar. 2012.
[101] R. J. Bessa and M. A. Matos. Economic and technical manage-
ment of an aggregation agent for electric vehicles: A literature survey.
European Transactions on Electical. Power, 2011. [Online]. Available:
http://onlinelibrary.wiley.com/doi/10.1002/etep.565/abstract.
[102] M. Singh, P. Kumar, and I. Kar. Implementation of vehicle to grid infrastruc-
ture using fuzzy logic controller. IEEE Transactions on Smart Grid, 3(1):565 References 243
–577, Mar. 2012.
[103] Diyun Wu, K.T. Chau, Chunhua Liu, Shuang Gao, and Fuhua Li. Transient
stability analysis of SMES for smart grid with vehicle-to-grid operation. IEEE
Transactions on Applied Superconductivity, 22(3):5701105, June 2012.
[104] K. Capion T. K. Kristoffersen and P. Meibom. Optimal charging of electric
drive vehicles in a market environment. Applied Energy, Vol. 88:1940–1948,
2011.
[105] Di Wu, D.C. Aliprantis, and Lei Ying. Load scheduling and dispatch for
aggregators of plug-in electric vehicles. IEEE Transactions on Smart Grid,
3(1):368 –376, Mar. 2012.
[106] Chenye Wu, H. Mohsenian-Rad, and Jianwei Huang. Vehicle-to-aggregator
interaction game. IEEE Transactions on Smart Grid, 3(1):434 –442, Mar.
2012.
[107] Sekyung Han, Soohee Han, and K. Sezaki. Development of an optimal vehicle-
to-grid aggregator for frequency regulation. IEEE Transactions on Smart
Grid, 1(1):65 –72, June 2010.
[108] J.R. Pillai and B. Bak-Jensen. Integration of vehicle-to-grid in the western
danish power system. IEEE Transactions on Sustainable Energy, 2(1):12 –19,
Jan. 2011.
[109] MOBIE, May 2011. [online] Available: http://www.mobie.pt.
[110] M. Jourabchi. Impact of plug-in hybrid vehicles on northwest power system: References 244
A preliminary assessment. Northwest Power and Conservation Council, 2008.
[111] F. R. Islam and H. R. Pota. Design a PV-AF system using V2G technology
to improve power quality. In 37th Annual Conference on IEEE Industrial
Electronics Society, IECON 2011, pages 861–866, nov. 2011.
[112] F. R. Islam, H. R. Pota, and M. S. Ali. V2G technology for designing active
filter system to improve wind power quality. In 21st Australasian Universities
Power Engineering Conference (AUPEC), 2011, pages 1–6, Sept. 2011.
[113] F. R. Islam and H. R. Pota. V2G technology to improve wind power quality
and stability. In Australian Control Conference (AUCC), 2011, pages 452–457,
Vov. 2011.
[114] Dunbar P. Birnie. Solar-to-vehicle (s2v) systems for powering commuters of
the future. Journal of Power Sources, 186(2):539 – 542, 2009.
[115] Prakash K. Ray, Soumya R. Mohanty, and Nand Kishor. Disturbance detection
in grid-connected distributed generation system using wavelet and s-transform.
Electric Power Systems Research, 81(3):805 – 819, 2011.
[116] Y. Ota, H. Taniguchi, and T. Nakajima. Autonomous distributed V2G satis-
fying scheduled charging. IEEE Transactions on Smart Grid, 3(1):559 –564,
March 2012.
[117] Jianhui Wang, Cong Liu, Dan Ton, Yan Zhou, Jinho Kim, and Anantray
Vyas. Impact of plug-in hybrid electric vehicles on power systems with demand
response and wind power. Energy Policy, 39(7):4016 – 4021, 2011. References 245
[118] Lisa Goransson, Sten Karlsson, and Filip Johnsson. Integration of plug-in hy-
brid electric vehicles in a regional wind-thermal power system. Energy Policy,
38(10):5482 – 5492, 2010.
[119] Scott B. Peterson, J. F. Whitacre, and Jay Apt. Net air emissions from electric
vehicles: The effect of carbon price and charging strategies. Environmental
Science & Technology, 45(5):1792–1797, 2011.
[120] Paulina Jaramillo, Constantine Samaras, Heather Wakeley, and Kyle Meister-
ling. Greenhouse gas implications of using coal for transportation: Life cycle
assessment of coal-to-liquids, plug-in hybrids, and hydrogen pathways. Energy
Policy, 37(7):2689 – 2695, 2009.
[121] P. Jaramillo and C. Samaras. Comparing life cycle ghg emissions from coal-
to-liquids and plug-in hybrids. Technical report, CEIC Working, 2007.
[122] J. Axsen and K. S. Kurani. Anticipating plug-in hybrid vehicle energy impacts
in california: Constructing consumer-informed recharge profiles. Transporta-
tion Research Part D: Trans. and Environ, Vol. 15(5):212 – 219, 2010.
[123] Jonn Axsen, Kenneth S. Kurani, Ryan McCarthy, and Christopher Yang.
Plug-in hybrid vehicle ghg impacts in california: Integrating consumer-
informed recharge profiles with an electricity-dispatch model. Energy Policy,
39(3):1617 – 1629, 2011.
[124] Ramteen Sioshansi, Riccardo Fagiani, and Vincenzo Marano. Cost and emis-
sions impacts of plug-in hybrid vehicles on the ohio power system. Energy References 246
Policy, 38(11):6703 – 6712, 2010.
[125] T. A. Stewart. E-check: A dirty word in ohios clean air debate. Technical
report, Capital University Law Review, vol. 29, pp. 338341, 2001.
[126] Corey D. White and K. Max Zhang. Using vehicle-to-grid technology for
frequency regulation and peak-load reduction. Journal of Power Sources,
196(8):3972 – 3980, 2011.
[127] A. De Los Rios, J. Goentzel, K.E. Nordstrom, and C.W. Siegert. Economic
analysis of vehicle-to-grid (v2g)-enabled fleets participating in the regulation
service market. In IEEE Innovative Smart Grid Technologies (ISGT), pages
1 –8, Jan. 2012.
[128] Chenye Wu, H. Mohsenian-Rad, Jianwei Huang, and J. Jatskevich. Pev-based
combined frequency and voltage regulation for smart grid. In IEEE Innovative
Smart Grid Technologies (ISGT), pages 1 –6, Jan. 2012.
[129] J.A.P. Lopes, F.J. Soares, and P.M.R. Almeida. Identifying management pro-
cedures to deal with connection of electric vehicles in the grid. In IEEE
Bucharest PowerTech, pages 1 –8, july 2009.
[130] Union for the coordination of transmission of electricity (UCTE), operation
handbook, 2008.
[131] A. Oudalov, D. Chartouni, C. Ohler, and G. Linhofer. Value analysis of
battery energy storage applications in power systems. In IEEE Power Systems
Conference and Exposition, PSCE ’06, pages 2206 –2211, Nov. 2006. References 247
[132] A. N. Brooks. Vehicle-to-grid demonstration project: Grid regulation ancillary
service with a battery electric vehicle. Technical report, . San Dimas, CA: AC
Propulsion, Inc., 2002.
[133] K. Mets, T. Verschueren, W. Haerick, C. Develder, and F. De Turck. Optimiz-
ing smart energy control strategies for plug-in hybrid electric vehicle charging.
In IEEE/IFIP Network Operations and Management Symposium Workshops
(NOMS Wksps), pages 293 –299, Apr. 2010.
[134] Lucy Sana. Driving the solution, the plug-in hybrid vehicle. EPRI Journal,
Fall:8–17, 2005.
[135] S.V. Chakraborty, S.K. Shukla, and J. Thorp. A detailed analysis of the
effective-load-carrying-capacity behavior of plug-in electric vehicles in the
power grid. In Innovative Smart Grid Technologies (ISGT), 2012 IEEE PES,
pages 1 –8, Jan. 2012.
[136] IEEE standard for interconnecting distributed resources with electric power
systems, 2003.
[137] E. Sortomme, M.M. Hindi, S.D.J. MacPherson, and S.S. Venkata. Coordinated
charging of plug-in hybrid electric vehicles to minimize distribution system
losses. IEEE Transactions on Smart Grid, 2(1):198 –205, March 2011.
[138] M. Bojrup, P. Karlsson, M. Alakula, and B. Simonsson. A dual purpose
battery charger for electric vehicles. In 29th Annual IEEE Power Electronics
Specialists Conference, 1998, volume 1, pages 565 –570 vol.1, May 1998. References 248
[139] Communication requirements for smart grid technologies. Technical report,
US Dept. Energy,Washington, DC, 2010.
[140] H. Khurana, M. Hadley, Ning Lu, and D.A. Frincke. Smart-grid security issues.
IEEE Security Privacy, 8(1):81 –85, Jan.-Feb. 2010.
[141] J.D. Dogger, B. Roossien, and F.D.J. Nieuwenhout. Characterization of li-
ion batteries for intelligent management of distributed grid-connected storage.
IEEE Transactions on Energy Conversion, 26(1):256 –263, March 2011.
[142] Sekyung Han, Soohee Han, and K. Sezaki. Economic assessment on v2g fre-
quency regulation regarding the battery degradation. In IEEE Innovative
Smart Grid Technologies (ISGT), pages 1 –6, Jan. 2012.
[143] C. Quinn, D. Zimmerle, and T.H. Bradley. An evaluation of state-of-charge
limitations and actuation signal energy content on plug-in hybrid electric ve-
hicle, vehicle-to-grid reliability, and economics. IEEE Transactions on Smart
Grid, 3(1):483 –491, March 2012.
[144] Scott B. Peterson, Jay Apt, and J.F. Whitacre. Lithium-ion battery cell degra-
dation resulting from realistic vehicle and vehicle-to-grid utilization. Journal
of Power Sources, 195(8):2385 – 2392, 2010.
[145] M.J. Rutherford and V. Yousefzadeh. The impact of electric vehicle battery
charging on distribution transformers. In Twenty-Sixth Annual IEEE Applied
Power Electronics Conference and Exposition (APEC), pages 396 –400, Mar.
2011. References 249
[146] C. Desbiens. Electric vehicle model for estimating distribution transformer
load for normal and cold-load pickup conditions. In IEEE Innovative Smart
Grid Technologies (ISGT), pages 1 –6, jan. 2012.
[147] Initiative on plug-in electric vehicles, commonwealth edison company, initial
assessment of the impact of the introduction of plug-in electric vehicles on the
distribution system. Technical report, Illinois Commerce Commission, Dec.
15 2010.
[148] Sungwoo Bae and A. Kwasinski. Spatial and temporal model of electric vehicle
charging demand. IEEE Transactions on Smart Grid, 3(1):394 –403, Mar.
2012.
[149] M. Etezadi-Amoli, K. Choma, and J. Stefani. Rapid-charge electric-vehicle
stations. IEEE Transactions on Power Delivery, 25(3):1883 –1887, July 2010.
[150] R.C. Green, Lingfeng Wang, and M. Alam. The impact of plug-in hybrid
electric vehicles on distribution networks: a review and outlook. In IEEE
Power and Energy Society General Meeting, pages 1 –8, July 2010.
[151] S.S. Raghavan and A. Khaligh. Impact of plug-in hybrid electric vehicle charg-
ing on a distribution network in a smart grid environment. In IEEE Innovative
Smart Grid Technologies (ISGT), pages 1 –7, Jan. 2012.
[152] J. Dowds C. Farmer, P. Hines and S. Blumsack. Modeling the impact of
increasing phev loads on the distribution infrastructure. 2010. References 250
[153] K.J. Dyke, N. Schofield, and M. Barnes. The impact of transport electrifi-
cation on electrical networks. IEEE Transactions on Industrial Electronics,
57(12):3917 –3926, Dec. 2010.
[154] L. Pieltain Fernandez, T. Gomez San Roman, R. Cossent, C.M. Domingo, and
P. Frias. Assessment of the impact of plug-in electric vehicles on distribution
networks. IEEE Transactions on Power Systems, 26(1):206 –213, Feb. 2011.
[155] Ken Huber Kevin Komara Steve Letendre Scott Baker Doug Brun-
ner Nat Pearre Willett Kempton, Victor Udo. A test of vehicle-
to-grid (V2G) for energy storage and frequency regulation in the
pjm system. Technical report, University of Delaware, Npv. 2008.
http://www.udel.edu/V2G/resources/test-v2g-in-pjm-jan09.pdf.
[156] E. Sortomme. Combined bidding of regulation and spinning reserves for uni-
directional vehicle-to-grid. In Innovative Smart Grid Technologies (ISGT),
pages 1–7. IEEE, 2012.
[157] Charles P. Steinmetz. Power control and stability of electric generating sta-
tions. American Institute of Electrical Engineers, Transactions of the, (2):1215
–1287, july 1920.
[158] History of plug-in hybrids. [online] Available: http://en.wikipedia.org.
[159] D M Lemoine, D M Kammen, and A E Farrell. An innovation and policy
agenda for commercially competitive plug-in hybrid electric vehicles. Envi-
ronmental Research Letters, 3(1):014003, 2008. References 251
[160] Hybrid cars. [online] Available: http://www.hybridcars.com.
[161] Minnesota Pollution Control Agency. Air emissions impacts of plug-in hybrid
vehicles in Minnesota’s passenger fleet. 2007.
[162] S.W. Hadley. Evaluating the impact of plug-in hybrid electric vehicles on
regional electricity supplies. In IREP Symposium on Bulk Power System Dy-
namics and Control VII. Revitalizing Operational Reliability, pages 1 –12, aug.
2007.
[163] Electric Power Research Institute (EPRI). Technology primer: the plug-in
hybrid electric vehicle. Technical report, Electric Power Research Institute
(EPRI), 2007.
[164] K. Clement, E. Haesen, and J. Driesen. Coordinated charging of multiple
plug-in hybrid electric vehicles in residential distribution grids. In IEEE Power
Systems Conference and Exposition, 2009, pages 1 –7, March 2009.
[165] M.D. Galus and G. Andersson. Demand management of grid connected plug-in
hybrid electric vehicles (PHEV). In IEEE Energy 2030 Conference, ENERGY
2008, pages 1 –8, Nov. 2008.
[166] A. Ipakchi and F. Albuyeh. Grid of the future. IEEE Power and Energy
Magazine, 7(2):52 –62, march-april 2009.
[167] C. Roe, A.P. Meliopoulos, J. Meisel, and T. Overbye. Power system level
impacts of plug-in hybrid electric vehicles using simulation data. In IEEE
Energy 2030 Conference, pages 1 –6, Nov. 2008. References 252
[168] C. Roe, J. Meisel, A.P. Meliopoulos, F. Evangelos, and T. Overbye. Power
system level impacts of PHEVs. In 42nd Hawaii International Conference on
System Sciences, HICSS ’09, pages 1 –10, Jan. 2009.
[169] Young-Joo Lee and A. Emadi. Integrated bi-directional AC/DC and DC/DC
converter for plug-in hybrid electric vehicle conversion. In IEEE Vehicle Power
and Propulsion Conference, pages 215 –222, Sept. 2007.
[170] S.H. Berisha, G.G. Karady, R. Ahmad, R. Hobbs, and D. Karner. Current
harmonics generated by electric vehicle battery chargers. In Power Electronics,
Drives and Energy Systems for Industrial Growth, 1996., Proceedings of the
1996 International Conference on, volume 1, pages 584 –589, jan 1996.
[171] M. Basu, K. Gaughan, and E. Coyle. Harmonic distortion caused by ev battery
chargers in the distribution systems network and its remedy. In 39th Interna-
tional Universities Power Engineering Conference, volume 2, pages 869 –873
vol. 1, sept. 2004.
[172] T. Das and D.C. Aliprantis. Small-signal stability analysis of power system
integrated with PHEVs. In IEEE Energy 2030 Conference, pages 1 –4, Nov.
2008.
[173] M. Pavella and P. G. Murthy. Transient Stability of Power Systems: Theory
and Practice. John Wiley and Sons, Chichester, 1994.
[174] G. Rogers. Power System Oscillations. Kluwer Academic Publishers, Boston,
2000. References 253
[175] H. K. Khalil. Nonlinear Systems. Prentice-Hall, New York, 1992.
[176] P. Kundur. Power System Stability and Control,. New York: McGraw-Hill,
1994.
[177] Fordescape. Avaiable: http://avt.inel.gov/pdf/phev/FordEscapePHEVfact.pdf.
[178] M. Ceraolo. New dynamical models of lead-acid batteries. IEEE Transactions
on Power Systems, 15(4):1184 –1190, nov 2000.
[179] S. Barsali and M. Ceraolo. Dynamical models of lead-acid batteries: imple-
mentation issues. IEEE Transactions on Energy Conversion, 17(1):16 –23,
Mar 2002.
[180] P. Menga R. Giglioli, A. Buonarota and M. Ceraolo. Charge and discharge
fourth order dynamic model of the leadacid battery. In 10th International
Electric Vehicle Symposium, Hong-Kong, Dec. 1990.
[181] P. Romano M. Ceraolo, D. Prattichizzo and F. Smargrasse. Experiences on
residual-range estimation of electric vehicles powered by leadacid batteries. In
15th International Electric Vehicle Symposium,Brussels, Belgium, 29th Oct.
1998.
[182] H. L. N.Wiegman and R. D. Lorenz. High efficiency battery state control
and power capability prediction. In 15th International Electric Vehicle Sym-
posium,Brussels, Belgium, 29th Oct. 1998.
[183] A. R. Bergen. Voltage Stability of Electric Power Systems,. New Jersey,
Prentice-Hall, 1986. References 254
[184] C. W. Taylor. Power System Voltage Stability. New York, McGraw-Hill, 1994.
[185] T. V. Cutsem and C. D. Vournas. Voltage Stability of Electric Power Systems.
Norwell, MA: Kluwer Academic, 1998.
[186] J.V. Milanovic and I.A. Hiskens. Effects of load dynamics on power system
damping. IEEE Transactions on Power Systems, 10(2):1022 –1028, May 1995.
[187] http://solarbuzz.com/facts-and-figures/market-facts/global-pv-market.
[188] D. Sera, T. Kerekes, R. Teodorescu, and F. Blaabjerg. Improved MPPT
method for rapidly changing environmental conditions. In IEEE International
Symposium on Industrial Electronics, volume 2, pages 1420 –1425, July 2006.
[189] T. Esram and P.L. Chapman. Comparison of photovoltaic array maximum
power point tracking techniques. IEEE Transactions on Energy Conversion,
22(2):439 –449, June 2007.
[190] F. Bouchafaa, D. Beriber, and M.S. Boucherit. Modeling and control of a
gird connected PV generation system. In 18th Mediterranean Conference on
Control Automation (MED), pages 315 –320, June 2010.
[191] N. Hamrouni and A. cherif. Modeling and control of a gird connected photo-
voltaic system. International Journal of Electrical and Power Engineering, 1
(3):307–313, 2007.
[192] Yun Tiam Tan and D.S. Kirschen. Impact on the power system of a large
penetration of photovoltaic generation. In IEEE Power Engineering Society
General Meeting, pages 1 –8, June 2007. References 255
[193] D. Gautam, V. Vittal, and T. Harbour. Impact of increased penetration of
dfig-based wind turbine generators on transient and small signal stability of
power systems. IEEE Transactions on Power Systems, 24(3):1426 –1434, Aug.
2009.
[194] S. Civanlar, J.J. Grainger, H. Yin, and S.S.H. Lee. Distribution feeder recon-
figuration for loss reduction. IEEE Transactions on Power Delivery, 3(3):1217
–1223, Jul 1988.
[195] S. Dahal, N. Mithulananthan, and T. Saha. Investigation of small signal
stability of a renewable energy based electricity distribution system. In IEEE
Power and Energy Society General Meeting, 2010, pages 1 –8, July 2010.
[196] W. Freitas, J.C.M. Vieira, A. Morelato, L.C.P. da Silva, V.F. da Costa, and
F.A.B. Lemos. Comparative analysis between synchronous and induction ma-
chines for distributed generation applications. IEEE Transactions on Power
Systems, 21(1):301 – 311, Feb. 2006.
[197] M.J. Hossain, H.R. Pota, V. Ugrinovskii, and R.A. Ramos. A robust STAT-
COM control to augment LVRT capability of fixed speed wind turbines. In
The 48th IEEE Conference on Decision and Control, held jointly with the 28th
Chinese Control Conference. CDC/CCC 2009, pages 7843 –7848, Dec. 2009.
[198] T. Ackermann. Wind Power in Power System”. England: John Wiley and
Sons Ltd., 2005.
[199] M.J. Hossain, H.R. Pota, V.A. Ugrinovskii, and R.A. Ramos. Simultaneous References 256
STATCOM and pitch angle control for improved LVRT capability of fixed-
speed wind turbines. IEEE Transactions on Sustainable Energy, 1(3):142 –151,
Oct. 2010.
[200] F. Milano. Power Syatem Modelling and Scripting. New York: Springer-
Verlag, 2010.
[201] D M Lemoine, D M Kammen, and A E Farrell. An innovation and policy
agenda for commercially competitive plug-in hybrid electric vehicles. Envi-
ronmental Research Letters, 3(1):014003, 2008.
[202] Y. Kobayashi, N. Kiyama, H. Aoshima, and M. Kashiyama. A route search
method for electric vehicles in consideration of range and locations of charging
stations. In IEEE Intelligent Vehicles Symposium (IV), pages 920 –925, June
2011.
[203] Wang Meng and Liu Kai. Optimization of electric vehicle charging station
location based on game theory. In International Conference on Transportation,
Mechanical, and Electrical Engineering (TMEE), pages 809 –812, Dec. 2011.
[204] G. Preetham and W. Shireen. Photovoltaic charging station for plug-in hybrid
electric vehicles in a smart grid environment. In IEEE Innovative Smart Grid
Technologies (ISGT), pages 1 –8, Jan. 2012.
[205] Long Jia, Zechun Hu, Yonghua Song, and Zhuowei Luo. Optimal siting and
sizing of electric vehicle charging stations. In Electric Vehicle Conference
(IEVC), pages 1 –6, Mar. 2012. References 257
[206] Willett Kempton and Jasna Tomi´c.Vehicle-to-grid power fundamentals: Cal-
culating capacity and net revenue. Journal of Power Sources, 144(1):268–279,
2005.
[207] C. Hutson, G.K. Venayagamoorthy, and K.A. Corzine. Intelligent scheduling
of hybrid and electric vehicle storage capacity in a parking lot for profit max-
imization in grid power transactions. In Energy 2030 Conference, pages 1–8,
Nov. 2008.
[208] H. Akagi, E. Watanabe, and M. Aredes. Instantaneous Power Theory and
Applications to Power Conditioning. Willey-Interscience, 2007.
[209] Fang Zheng Peng and Jih-Sheng Lai. Generalized instantaneous reactive power
theory for three-phase power systems. IEEE Transaction on Instrumentation
and Measurement, 45(1):293–297, Feb 1996.
[210] Gunnarsson S., L. Jiang, and Petersson A. Active filters in HVDC transmis-
sion. 2009. [online] Available: www.abb.com/hvdc.
[211] E. W. Kimbark. Direct Current Transmission. Willey-Interscience, 1971.
[212] D. A. Bradley J. Arrilaga and P. S. Bodger. Power System Harmonics. John
Willy & Sons, 1985.
[213] A. Sarshar, M.R. Iravani, and J. Li. Calculation of HVDC converter non-
characteristic harmonics using digital time-domain simulation method. IEEE
Transactions on Power Delivery, 11(1):335 –344, Jan 1996.
[214] M. Etezadi-Amoli and T. Florence. Voltage and current harmonic content of References 258
a utility system-a summary of 1120 test measurements. IEEE Transactions
on Power Delivery, 5(3):1552–1557, 1990.
[215] TC Shuter, HT Vollkommer Jr, and TL Kirkpatrick. Survey of harmonic
levels on the american electric power distribution system. IEEE Transactions
on Power Delivery, 4(4):2204–2213, 1989.
[216] B. K. Bose. Modern Power Electronics - Evolution, Technology and Applica-
tion. Jhon Willey & Sons, 1992.
[217] H. Sasaki and T. Machida. Transient analysis of harmonic current elimina-
tion method by magnetic flux compensation. IEEE Transactions on Power
Apparatus and Systems, PAS-93(2):669 –675, Mar. 1974.
[218] J. Arrilaga. High Voltage Direct Current Transmission. Peter Peregrinus Ltd.,
1983.
[219] L. Gyugyi and E. C. Strycula. Active AC power filters. In IEEE Industry
Applications Society Annual Meeting, pages 529– 535, 1976.
[220] W. F. Long G. R. Dreifuerst N. Mohan, H. A. Peterson and J. J. Vithayathil.
Active filters for AC harmonic suppression. IEEE Transactions on Power
Apparatus and Systems, 96(4), July/Aug 1977.
[221] C. Wong, N. Mohan, S.E. Wright, and K.N. Mortensen. Feasibility study of
ac- and dc-side active filters for hvdc converter terminals. IEEE Transactions
on Power Delivery, 4(4):2067 –2075, Oct 1989. References 259
[222] F.Z. Peng, H. Akagi, and A. Nabae. A new approach to harmonic compen-
sation in power systems a combined system of shunt passive and series active
filters. IEEE Transactions on Industry Applications, 26(6):983 –990, Nov/Dec
1990.
[223] H. Fujita and H. Akagi. A practical approach to harmonic compensation in
power systems-series connection of passive and active filters. IEEE Transac-
tions on Industry Applications, 27(6):1020 –1025, Nov/Dec 1991.
[224] Isao Takahashi and Yasushi Omura. High-power active filter using LC-tuned
filter. Electrical Engineering in Japan, 113(3):135–143, 1993.
[225] M.C. Kisacikoglu, B. Ozpineci, and L.M. Tolbert. Examination of a PHEV
bidirectional charger system for V2G reactive power compensation. In Applied
Power Electronics Conference and Exposition (APEC), pages 458–465, Feb.
2010.
[226] SAFT hybrid electric military vehicle data sheet.
[227] Saft hybride electric military vehicle data sheet. [online] Avialable:
http://www.saftbatteries.com/.
[228] O. Tremblay, L. A. Dessaint, and A. I. Dekkiche. A generic battery model
for the dynamic simulation of hybrid electric vehicles. In IEEE Vehicle Power
and Propulsion Conference, 2007, pages 284 –289, Sept. 2007.
[229] M. Yilmaz and P. Krein. Review of the impact of vehicle-to-grid technologies
on distribution systems and utility interfaces. IEEE Transactions on Power References 260
Electronics, PP(99):1, 2012.
[230] Yu Du, S. Lukic, B. Jacobson, and A. Huang. Review of high power isolated
bi-directional DC-DC converters for PHEV/EV DC charging infrastructure.
In Energy Conversion Congress and Exposition (ECCE), pages 553 –560, Sept.
2011.
[231] Hao Chen, Xiaochen Wang, and A. Khaligh. A single stage integrated bidirec-
tional ac/dc and dc/dc converter for plug-in hybrid electric vehicles. In IEEE
Vehicle Power and Propulsion Conference (VPPC), 2011, pages 1 –6, Sept.
2011.
[232] Gyu-Yeong Choe, Jong-Soo Kim, Byoung-Kuk Lee, Chung-Yuen Won, and
Tea-Won Lee. A bi-directional battery charger for electric vehicles using pho-
tovoltaic PCS systems. In Vehicle Power and Propulsion Conference (VPPC),
2010 IEEE, pages 1 –6, Sept. 2010.
[233] M. A. Guerrero-Martnez J. G. Lozano, M. I. Milans-Montero and E. Romero-
Cadaval. Three-phase bidirectional battery charger for smart electric vehicles.
In in Proceding of Compatibility and Power Electronics Conference and Work-
shop, pages 371 – 376, 2011.
[234] E. Sortomme and M.A. El-Sharkawi. Optimal combined bidding of vehicle-to-
grid ancillary services. IEEE Transactions on Smart Grid, 3(1):70 –79, Mar.
2012. References 261
[235] A. Brooks and S. H. Thesen. PG & E and tesla motors: Vehicle to grid demon-
stration and evaluation program. In in Proc. Electrical Vehicle Symposium,
pages 1–10, Dec. 2007.
[236] J. Voelcker. Lithium batteries take to the road. IEEE Spectrum, 44(9):26 –31,
Sept. 2007.
[237] M.F.M. Elias, K.M. Nor, N.A. Rahim, and A.K. Arof. Lithium-ion battery
charger for high energy application. In Power Engineering Conference, pages
283 – 288, Dec. 2003.
[238] P. V. D. Bossche. The electric vehicle: raising the standards. PhD thesis, Vrije
Universteit Brussel, 2003.
[239] Reductive charging. Technical report. [online] Available:
http://www.acpropulsion.com.
[240] W. Korthof. Level 2+: Economical Fast Charging for EVs. In 17th Electric
Vehicle Symposium, Montreal, Canada, October 2000.
[241] Lixin Tang and Gui-Jia Su. Control scheme optimization for a low-cost,
digitally-controlled charger for plug-in hybrid electric vehicles. In IEEE En-
ergy Conversion Congress and Exposition (ECCE), 2010, pages 3604 –3610,
Sept. 2010.
[242] Edson H. Watanabe, MaurAcio˜ Aredes, and Hirofumi Akagi. The p-q theory
for active filter control: some problems and solutions. Transaction on Revista
Control and Automation. References 262
[243] A. Tilli and A. Tonielli. Sequential design of hysteresis current controller for
three-phase inverter. IEEE Transactions on Industrial Electronics, 45(5):771–
781, Oct 1998.
[244] M. A. Laughton and D.F. Warne., editors. Electrical Engineer’s Reference
Book (16th Edition). Elsevier, 2003.
[245] IEEE guide for the evaluation of the reliability of hvdc converter stations.
IEEE Std 1240-2000, page i, feb. 2001.
[246] V.G. Agelidis, G.D. Demetriades, and N. Flourentzou. Recent advances in
high-voltage direct-current power transmission systems. In IEEE International
Conference on Industrial Technology, pages 206 –213, Dec. 2006.
[247] N. Flourentzou, V.G. Agelidis, and G.D. Demetriades. VSC-based HVDC
power transmissions systems: An overview. IEEE Transaction on Power Elec-
tronics, 24(3):592 –602, Mar. 2009.
[248] A. Singh and B. Singh. Performance evaluation of power converters with
distribution static compensator. In Joint International Conference on Power
Electronics, Drives and Energy Systems (PEDES), India, pages 1 –5, Dec.
2010.
[249] J. D. Ainsworth. Proposed benchmark model for study of HVDC controls by
simulator or digital computer. In Procedings, CIGRE SC-14 Colloquium on
HVDC With Weak AC Systems, Maidstone, Sep. 1985.
[250] M. Szechtman, T. Wess, and C.V. Thio. First benchmark model for HVDC References 263
control studies. Electra, No. 135, page 5467, Apr. 1991.
[251] M.O. Faruque, Yuyan Zhang, and V. Dinavahi. Detailed modeling of CIGRE
HVDC benchmark system using PSCAD/EMTDC and PSB/SIMULINK.
IEEE Transactions on Power Delivery, 21(1):378–387, Jan. 2006.
[252] IEEE Std 519-1992, IEEE Recommended Practices and Requirements for Har-
monic Control in Electric Power Systems ⃝c Institute of Electrical and Elec-
tronics Engineers, Inc, 1993.
[253] Po-Tai Cheng, S. Bhattacharya, and D.M. Divan. Application of dominant
harmonic active filter system with 12 pulse nonlinear loads. IEEE Transactions
on Power Delivery, 14(2):642–647, Apr 1999.
[254] Jlio Martins Joao Afonso, Carlos Couto. Active filters with control based on
the p-q theory. IEEE Industrial Electronics Society Newsletter, 47(3):5–10,
2000.
[255] Vladislav Akhmatov and Hans Knudsen. An aggregate model of a grid-
connected, large-scale, offshore wind farm for power stability investigations-
importance of windmill mechanical system. International Journal of Electrical
Power & Energy Systems, 24(9):709–717, Nov. 2002.
[256] Akhmatov, V. Knudsen, and H. Nielsen. Advanced simulation of windmills
in the electrical power supply. International Journal of Electrical Power &
Energy Systems, Elsevier, 22(6):421–434, Aug. 2000.
[257] A. Feijo, J. Cidrs, and C. Carrillo. A third order model for the doubly-fed References 264
induction machine. Electric Power Systems Research, 56(2):121–127, Nov.
2000.
[258] K. Nandigam and B. H. Chowdhury. Power flow and stability models for in-
duction generators used in wind turbines. In IEEE Power Engineering Society
General Meeting, pages 2012–2016, Denver, CO, June 10–10, 2004.
[259] J. G. Slootweg, S. W. H. de Haan, H. Polinder, and W. L. Kling. General
model for representing variable speed wind turbines in power system dynamics
simulations. IEEE Transactions on Power Systems, 18(1):144–151, Feb. 2003.
[260] P. Ledesma, J. Usaola, and J. L. Rodriguez. Transient stability of a fixed
speed wind farm. Renewable Energy, 28(9):1341–1355, July 2003.
[261] S. A. Papathanassiou; M.P. Papadopoulos. Harmonic analysis in a power sys-
tem with wind generation. IEEE Transactions on Power Delivery, 21(4):2006–
20016, 2006.
[262] George J. Wakileh. Power System Harmonics. New York: Springer-Verlag,
2001.
[263] F.S. dos Reis, J.A.V. Ale, F.D. Adegas, R. Tonkoski, Slan S, and K. Tan.
Active shunt filter for harmonic mitigation in wind turbines generators. In
37th IEEE Power Electronics Specialists Conference, 2006, pages 1 –6, June
2006.
[264] Yong Hua Song and Allan T Johns, editors. Flexible ac transmission system
(FACTS). The Institution of Electrical Engineers, London, 1999. References 265
[265] Arthit Sode-Yome, Nadarajah Mithulananthan, and Kwang Y Lee. A compre-
hensive comparison of FACTS devices for enhancing static voltage stability.
In IEEE Power Engineering Society General Meeting, pages 1–8. IEEE, 2007.
[266] Nadarajah Mithulananthan and Naresh Acharya. A proposal for investment
recovery of facts devices in deregulated electricity markets. Electric Power
Systems Research, 77(5):695–703, 2007.
[267] R. Mathur and R. Varma. Thyristor-Based FACTS Controllers for Electrical
Transmission Systems. IEEE Press Series on Power Engineering. 2002.
[268] E. W. Kimbark. Direct Current Transmission. Willey-Interscience, 1971.
[269] M.C. Kisacikoglu, B. Ozpineci, and L.M. Tolbert. Examination of a PHEV
bidirectional charger system for V2G reactive power compensation. In Twenty-
Fifth Annual IEEE Applied Power Electronics Conference and Exposition
(APEC), pages 458 –465, Feb. 2010.
[270] L. Gyugyi N. G. Higorani. Understanding FACTS Devices. IEEE Press, 2000.
[271] A. Edris, A.S. Mehraban, M. Rahman, L. Gyugyi, S. Arabi, and T. Reitman.
Controlling the flow of real and reactive power. IEEE Computer Applications
in Power, 11(1):20 –25, Jan 1998.
[272] L. Gyugyi. Unified power-flow control concept for flexible ac transmission
systems. IEE Proceedings on Generation, Transmission and Distribution,
139(4):323 –331, July 1992.
[273] M. Noroozian, L. Angquist, M. Ghandhari, and G. Andersson. Use of UPFC References 266
for optimal power flow control. IEEE Transactions on Power Delivery,
12(4):1629 –1634, Oct 1997.
[274] A. Nabavi-Niaki and M.R. Iravani. Steady-state and dynamic models of unified
power flow controller (upfc) for power system studies. IEEE Transactions on
Power Systems, 11(4):1937 –1943, Nov 1996.
[275] C.R. Fuerte-Esquivel and E. Acha. Newton-raphson algorithm for the reliable
solution of large power networks with embedded facts devices. IEE Proceedings
on Generation, Transmission and Distribution,, 143(5):447 –454, Sept. 1996.
[276] H. Ambriz-Perez, E. Acha, C.R. Fuerte-Esquivel, and A. de la Torre. Incorpo-
ration of a upfc model in an optimal power flow using newton’s method. IEE
Proceedings on Generation, Transmission and Distribution,, 145(3):336 –344,
May 1998.
[277] E. Uzunovic, C.A. Canizares, and J. Reeve. Fundamental frequency model of
unified power flow controller. In North Amirican Power Symposium (NAPS),
1998, pages 294–299, Oct. 1998.
[278] K.S. Smith, L. Ran, and J. Penman. Dynamic modelling of a unified power flow
controller. IEE Proceedings on Generation, Transmission and Distribution,,
144(1):7 –12, Jan 1997.
[279] Z. Huang, Y. Ni, CM Shen, F.F. Wu, S. Chen, and B. Zhang. Application of
unified power flow controller in interconnected power systems-modeling, inter-
face, control strategy, and case study. IEEE Transactions on Power Systems, References 267
15(2):817–824, 2000.
[280] C. Schauder and H. Mehta. Vector analysis and control of advanced static var
compensators. In IEE Proceedings on Generation, Transmission and Distri-
bution, volume 140, pages 299–306. IET, 1993.
[281] KR Padiyar and AM Kulkarni. Control design and simulation of unified power
flow controller. IEEE Transactions on Power Delivery, 13(4):1348–1354, 1998.
[282] H.F. Wang. Damping function of unified power flow controller. In IEE Pro-
ceeding on Generation, Transmission and Distribution,, volume 146, pages
81–87. IET, 1999.
[283] HF Wang. Applications of modelling UPFC into multi machine power systems.
In IEE Proceedings on Generation, Transmission and Distribution, volume
146, pages 306–312. IET, 1999.
[284] E. Uzunovic, C.A. Canizares, and J. Reeve. EMTP studies of UPFC power
oscillation damping. In Proceedings of the North American Power Symposium
(NAPS), pages 405–410, 1999.
[285] P. Kumkratug and MH Haque. Versatile model of a unified power flow con-
troller in a simple power system. In IEE Proceedings on Generation, Trans-
mission and Distribution, volume 150, pages 155–161. IET, 2003.
[286] ZJ Meng and PL So. A current injection UPFC model for enhancing power
system dynamic performance. In IEEE Power Engineering Society Winter
Meeting, volume 2, pages 1544–1549. IEEE, 2000. References 268
[287] M. Noroozian, L. Angquist, M. Ghandhari, and G. Andersson. Improving
power system dynamics by series-connected facts devices. IEEE Transactions
on Power Delivery, 12(4):1635–1641, 1997.
[288] F. R. Islam and H. R. Pota. Design a PV-AF system using V2G technology
to improve power quality. In 37th Annual Conference on IEEE Industrial
Electronics Society, IECON 2011, pages 861 –866, nov. 2011.
[289] F. R. Islam, H. R. Pota, and M. S. Ali. V2G technology for designing active
filter system to improve wind power quality. In 21st Australasian Universities
Power Engineering Conference (AUPEC), 2011, pages 1–6, Sept. 2011.
[290] F. R. Islam and H. R. Pota. V2G technology to improve wind power quality
and stability. In Australian Control Conference (AUCC), 2011, pages 452
–457, Nov. 2011.
[291] H. Akagi and H. Fujita. A new power line conditioner for harmonic compen-
sation in power systems. IEEE Transactions on Power Delivery, 10(3):1570–
1575, 1995.
[292] H. Fujita and H. Akagi. The unified power quality conditioner: the integration
of series and shunt-active filters. IEEE Transactions on Power Electronics,
13(2):315–322, 1998.
[293] H. Akagi, E. Watanabe, and M. Aredes. Instantaneous Power Theory and
Applications to Power Conditioning. Willey-Interscience, 2007.
[294] E. Acha and M. Madrigal. Power System Harmonics, Computer Modelling References 269
and Analysis. John Wiley & Sons, Chichester, England, 2001.
[295] M.H.J. Bollen. Understanding power quality problems. IEEE press New York,
2000.
[296] EN 61000-4-30, 2002 (draft). Electromagnetic compatibility: Testing and
measuring techniques power quality measurements method.
[297] R. C. Dugan, M. F. Mcgranaghan, S. Santoso, and H. W. Beaty. Electrical
Power Systems Quality (2nd edition). McGraw-Hill, New York,.
[298] B. Buchholz, I. Huebel, D. Povh, and M. Weinhold. Medium-voltage direct-
current (MVDC) coupling in the liberalized market. In CIGRE Session, pages
2002–14, 2002.
[299] B. Buchholz, C. Schwaegerl, T. Stephanblome, H. Frey, N. Lewald, and Z. Sty-
czynski. Advanced planning and operation of dispersed generation ensuring
power quality, security and efficiency in distribution systems.. In Beitrag
CIGRE-Konferenz Paris, 2004.
[300] R. DeBlasio and C. Tom. Standards for the smart grid. In IEEE Energy 2030
Conference, pages 1–7. IEEE, 2008.
[301] K Rudion, A Orths, ZA Styczynski, and K Strunz. Design of benchmark of
medium voltage distribution network for investigation of DG integration. In
IEEE Power Engineering Society General Meeting, 2006, pages 6–12. IEEE,
2006.
[302] P. Mitra, G.K. Venayagamoorthy, and K.A. Corzine. Smartpark as a virtual References 270
STATCOM. IEEE Transactions on Smart Grid, 2(3):445 –455, Sept. 2011.
[303] Paisan Boonchiam and Nadarajah Mithulananthan. Understanding of dy-
namic voltage restorers through matlab simulation. Thammasat Int. J. Sc.
Tech, 11(3):1–6, 2006.