8830 PDF

ÇUKUROVA UNIVERSITY
INSTITUTE OF NATURAL AND APPLIED SCIENCES
PhD THESIS
Ahmet Mete VURAL
MODELING OF MULTI-CONVERTER FACTS (FLEXIBLE

ALTERNATING CURRENT TRANSMISSION SYSTEMS)
DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING
ADANA, 2012
MODELING OF MULTI-CONVERTER FACTS (FLEXIBLE ALTERNATING

CURRENT TRANSMISSION SYSTEMS
Ahmet Mete VURAL
PhD THESIS
We certify that the thesis titled above was reviewed and approved for the award of
degree of the Doctor of Philosophy by the board of jury on 24/12/2012.
……………….................................. …………….……........... ……………………….…...

Assoc. Prof. Dr. K. Çağatay BAYINDIR Prof. Dr. Mehmet TÜMAY Prof. Dr. Tankut YALÇINÖZ
SUPERVISOR CO-SUPERVISOR MEMBER
…………………...……... ………...............................
Assoc. Prof. Dr. İlyas EKER Assoc. Prof. Dr. Ulus ÇEVİK
MEMBER MEMBER
This Ph.D. Thesis is written at the Institute of Natural and Applied Sciences of
Çukurova University.
Registration Number:
Prof. Dr. Selahattin SERİN

Director
Institute of Natural and Applied Sciences
Note: The usage of the presented specific declarations, tables, figures and photographs either in this
thesis or in any other reference without citation is subject to "The law of Arts and Intellectual
Products" number of 5846 of Turkish Republic.
ABSTRACT
PhD THESIS
MODELING OF MULTI-CONVERTER FACTS (FLEXIBLE

ALTERNATING CURRENT TRANSMISSION SYSTEMS)
Ahmet Mete VURAL
Supervisor: Assoc. Prof. Dr. K. Çağatay BAYINDIR

Co-Supervisor: Prof. Dr. Mehmet TÜMAY
Year: 2012, Pages: 249
Jury: Prof. Dr. Mehmet TÜMAY
Prof. Dr. Tankut YALÇINÖZ
Assoc. Prof. Dr. İlyas EKER
Assoc. Prof. Dr. Ulus ÇEVİK
Assoc. Prof. Dr. K. Çağatay BAYINDIR
Multi-converter FACTS devices can increase control flexibility of power

systems by providing independent and simultaneous control of multi-power system
parameters. In this thesis, the following multi-converter FACTS devices are modeled
and analyzed at transmission level: Generalized Unified Power Flow Controller
(GUPFC), Interline Power Flow Controller (IPFC), and Back-to-Back Static
Synchronous Compensator (BtB-STATCOM). The steady-state PSCAD models of
these devices are proposed for power flow studies without requiring programming
coding. A quasi multi-pulse voltage source converter with low harmonic content is
designed for converter-level modeling of the aforementioned devices. The magnitude
and the phase angle of the voltage of the designed converter are controlled efficiently
despite triggering the power semiconductors at fundamental system frequency. The
dynamic control characteristics of GUPFC, IPFC, and BtB-STATCOM are
extensively studied and analyzed using the developed PSCAD based converter-level
models in various test power systems. A novel fuzzy-rule based control scheme for
IPFC to decouple real and reactive power flow control loops is designed and tested
through miscellaneous and comparative simulation studies. Using converter-level
models of these devices, transient stability studies of the power systems are also
addressed in this thesis. A novel self-tuning fuzzy damping controller for GUPFC is
designed to damp synchronous generator oscillations and to increase speed stability
of induction generators which are located in a wind farm. The same controller is
adapted for IPFC to suppress inter-area mode of oscillations.
Key Words: FACTS Devices, Modeling, Simulation.
I
ÖZ
DOKTORA TEZİ
ÇOK KONVERTÖRLÜ ESNEK ALTERNATİF AKIM İLETİM

SİSTEMLERİNİN MODELLENMESİ
Ahmet Mete VURAL
ÇUKUROVA ÜNİVERSİTESİ
FEN BİLİMLERİ ENSTİTÜSÜ
ELEKTRİK ELEKTRONİK MÜHENDİSLİĞİ ANABİLİM DALI
1. Danışman: Doç. Dr. K. Çağatay BAYINDIR

2. Danışman: Prof. Dr. Mehmet TÜMAY
Yıl: 2012, Sayfa: 249
Jüri: Prof. Dr. Mehmet TÜMAY
Prof. Dr. Tankut YALÇINÖZ
Doç. Dr. İlyas EKER
Doç. Dr. Ulus ÇEVİK
Doç. Dr. K. Çağatay BAYINDIR
Çok konvertörlü FACTS cihazları, güç sistemlerinin denetim esnekliğini,

çoklu güç sistemi parametrelerinin bağımsız ve eşzamanlı denetimlerini sağlayarak
artırabilirler. Bu tezde, iletim seviyesinde modellenen ve analiz edilen çok
konvertörlü FACTS cihazları şunlardır: Genelleştirilmiş Bütünleşik Güç Akış
Denetleyicisi (GUPFC), Hatlarası Güç Akış Denetleyicisi (IPFC) ve Durgun Senkron
Kompanzatör (STATCOM). Bu cihazların durgun-hal PSCAD modelleri
programlama kodlamasına ihtiyaç duyulmadan güç akış çalışmaları için önerilmiştir.
Düşük harmonik içerikli çok darbeli benzeri bir gerilim kaynaklı konvertör sözü
geçen cihazların konvertör seviyesi modellemeleri için tasarlanmıştır. Tasarlanan
konvertörün geriliminin büyüklüğü ve faz açısı, güç yarıiletkenlerinin temel şebeke
frekansında tetiklenmelerine karşın etkin bir şekilde denetlenmektedir. GUPFC,
IPFC ve BtB-STATCOM’un dinamik denetim nitelikleri, geliştirilen PSCAD tabanlı
konvertör seviyesi modellerin çeşitli örnek güç sistemleri üzerinde kullanılmasıyla
kapsamlı olarak çalışılmış ve analiz edilmiştir. IPFC için aktif ve reaktif güç akış
denetim döngülerini ayrıştırmak amacıyla özgün bulanık-kural tabanlı bir denetim
şeması tasarlanmış, çok yönlü ve karşılaştırmalı benzetim çalışmaları ile test
edilmiştir. Bu cihazların konvertör seviyesi modelleri kullanılarak güç sistemlerinin
geçici kararlılık çalışmaları da bu tezde ele alınmıştır. GUPFC için, senkron jeneratör
salınımlarını söndürmek ve bir rüzgar çiftliğine yerleştirilmiş endüksiyon
jeneratörlerinin hız kararlılıklarını artırmak için özgün kendi kendine ayarlamalı bir
bulanık sönümleme denetleyicisi tasarlanmıştır. Aynı denetleyici alanlar arası
salınım modlarının bastırılması amacıyla IPFC için uyarlanmıştır.
Anahtar Kelimeler: FACTS Cihazları, Modelleme, Simülasyon.
II
ACKNOWLEDGEMENTS
I would like to express my sincere gratitude to my co-supervisor Prof. Dr.

Mehmet Tümay for his guidance, advice, and encouragement throughout my studies,
who is also a role model in my career. I always appreciate his ideas to build my own
academic interests.
I would like to express my sincere gratitude to my supervisor Assoc. Prof. Dr.
Kamil Çağatay Bayındır for his remarkable advice, guidance, criticism, and
encouragement throughout my studies. He has been great sources of inspiration to
me.
I wish to express my special thanks to the rector of Hasan Kalyoncu
University, Prof. Dr. İbrahim Özdemir and engineering faculty staff for their good
wishes and maintaining an excellent academic environment which had a positive
impact on this work.
I would like to thank deeply Electrical and Electronics Engineering
Department staff for their insights and motivation during the time I worked in Atılım
University.
I gratefully thank Prof. Dr. Tankut Yalçınöz, Assoc. Prof. Dr. Ulus Çevik,
and Assoc. Prof. Dr. İlyas Eker for participating in my thesis defense committee.
I would like to express my deepest gratitude to my mother, father and my
brother for their patience, sacrifice, encouragement, and continuous morale support.
They rendered me enormous support during the whole study period.
I would like to express my deepest gratitude to my wife Selma for her
patience, sacrifice, encouragement, and continuous morale support.
Specially, I would like to dedicate this work to my mother with thanks and
appreciations for her great patience and encouragement. Her trust and morale support
inspired me in the most moments of the study period.
III
CONTENTS PAGE
ABSTRACT ............................................................................................................. I
ÖZ ........................................................................................................................... II
ACKNOWLEDGEMENTS .................................................................................... III
CONTENTS ...........................................................................................................IV
LIST OF TABLES..................................................................................................XI
LIST OF FIGURES ............................................................................................. XIII
LIST OF SYMBOLS ......................................................................................... XVII
LIST OF ABBREVIATIONS .............................................................................. XXI
1. INTRODUCTION ............................................................................................... 1
1.1. Motivation for the Thesis .............................................................................. 3
1.2. Objectives of the Thesis ................................................................................ 4
1.3. Contributions of the Thesis ............................................................................ 5
1.4. General Outline ............................................................................................. 5
2. OVERVIEW OF FACTS DEVICES .................................................................... 7
2.1. Background on Alternating Current Power Transmission .............................. 7
2.1.1. Thermal Limit .................................................................................... 7
2.1.2. Maximum Power Transfer.................................................................. 7
2.1.3. Angle Stability ................................................................................... 8
2.1.4. Voltage Stability ................................................................................ 9
2.1.5. Transmission Line Loadability Characteristics ................................. 10
2.2. Classification of FACTS Devices ................................................................ 11
2.2.1. Static VAR Compensator (SVC) ...................................................... 13
2.2.2. Thyristor Controlled Series Capacitor/Compensator (TCSC)............ 15
2.2.3. Thyristor Controlled Phase Angle Regulator (TCPAR) .................... 16
2.2.4. Static Synchronous Compensator (STATCOM) ............................... 17
2.2.5. Static Series Synchronous Compensator (SSSC) .............................. 18
2.2.6. Unified Power Flow Controller (UPFC) ........................................... 19
2.2.7. Interline Power Flow Controller (IPFC) ........................................... 20
2.2.8. Generalized Unified Power Flow Controller (GUPFC) ..................... 21
IV
2.2.9. Back-to-Back STATCOM (BtB-STATCOM) ................................ 22
2.3. More Control Degrees of Freedom .............................................................. 23
2.4. Recent Advances in Power Semiconductors ................................................ 25
2.5. Field Applications of FACTS Devices at Transmission Level ..................... 25
2.6. Summary..................................................................................................... 27
3. STEADY-STATE MODELING......................................................................... 29
3.1. Introduction................................................................................................. 29
3.2. Proposed Steady-state Modeling Approach.................................................. 30
3.2.1. Configurable Multi-Converter FACTS Device................................... 31
3.2.2. Operating Constraints ........................................................................ 33
3.2.3. Control Constraints............................................................................ 35
3.2.3.1. Direct Control Mode .............................................................. 35
3.2.3.2. Indirect Control Mode............................................................ 36
3.3. Modeling in PSCAD ................................................................................... 37
3.3.1. Power Circuit .................................................................................... 37
3.3.2. Control Circuit .................................................................................. 37
3.4. Power Flow Studies ..................................................................................... 39
3.4.1. Test Systems ..................................................................................... 39
3.4.2. WSCC 3-Machine 9-Bus System ....................................................... 42
3.4.2.1. Case 1: STATCOM and SSSC Operations ............................. 42
3.4.2.2. Case 2: UPFC Operation ........................................................ 45
3.4.2.3. Case 3: IPFC Operation ......................................................... 45
3.4.2.4. Case 4: GUPFC Operation ..................................................... 47
3.4.2.5. Discussion of Simulation Results ........................................... 48
3.4.3. IEEE 14-Bus System ......................................................................... 49
3.4.3.1. Case 1: UPFC Operation ........................................................ 49
3.4.3.2. Case 2: IPFC Operation ......................................................... 51
3.4.3.3. Case 3: GUPFC Operation ..................................................... 51
3.4.4. 3-Machine 7-Bus System ................................................................... 52
3.4.4.1. Case 1: Reactive Power-Voltage (Q-V) Characteristics .......... 52
V
3.4.4.2. Case 2: Real Power-Voltage (P-V) Characteristics ................. 54
3.5. Summary..................................................................................................... 57
4. VOLTAGE SOURCE CONVERTER DESIGN ................................................. 59
4.1. Introduction................................................................................................. 59
4.2. Six-pulse VSC ............................................................................................. 60
4.2.1. Circuit Configuration......................................................................... 60
4.2.2. Working Principle ............................................................................. 61
4.2.3. Analysis of Six-pulse VSC ................................................................ 63
4.3. Twelve-pulse VSC ...................................................................................... 65
4.3.2. Analysis of Twelve-pulse VSC .......................................................... 67
4.4. Quasi Multi-pulse VSC ............................................................................... 68
4.4.2. Series Coupling Magnetic Interface ................................................... 72
4.4.3. Control Scheme for Quasi Multi-pulse VSC ...................................... 73
4.4.3.1. 2-angle Control Method ......................................................... 74
4.4.3.2. Pulse-generating Circuit......................................................... 76
4.4.4. Analysis of Quasi Multi-pulse VSC ................................................... 77
4.4.4.1. Quasi 48-pulse Operation....................................................... 77
4.4.4.2. Verification of 2-angle Control Method ................................. 79
4.5. Summary..................................................................................................... 83
5. DYNAMIC MODELING STUDIES .................................................................. 85
5.1. Introduction................................................................................................. 85
5.2. Simplex Optimization Method..................................................................... 87
5.3. Converter-Level Modeling of GUPFC ......................................................... 88
5.3.1. GUPFC Interacting with Power System ............................................. 88
5.3.2. GUPFC Controller Design ................................................................. 89
5.3.3. Finding Optimum Controller Parameters ........................................... 91
5.3.4. Simulation Studies ............................................................................. 93
5.3.4.1. Case 1: Start-up Transients .................................................... 94
VI
5.3.4.2. Case 2: Response to Real Power Flow Step Changes ............. 97
5.3.4.3. Case 3: Response to Reactive Power Flow Step Changes ..... 100
5.3.4.4. Case 4: Single-phase to Ground Fault .................................. 102
5.3.4.5. Case 5: Three-phase to Ground Fault ................................... 106
5.3.4.6. THD Content ....................................................................... 108
5.3.5. Discussion ....................................................................................... 108
5.4. Converter-Level Modeling of IPFC ........................................................... 109
5.4.1. IPFC Interacting with Power System ............................................... 109
5.4.2. IPFC Controller Design ................................................................... 110
5.4.2.1. Decoupled Controller Design ............................................... 111
5.4.2.2. Proposed Hybrid Fuzzy PI (HFPI) Controller ...................... 114
5.4.2.3. FUDE Design ...................................................................... 116
5.4.3. Finding Optimum Controller Parameters ......................................... 119
5.4.4. Simulation Studies ........................................................................... 121
5.4.4.1. Case 1 .................................................................................. 122
5.4.4.2. Case 2 .................................................................................. 124
5.4.4.3. THD Content ....................................................................... 128
5.4.5. Discussion ....................................................................................... 130
5.5. Converter-Level Modeling of BtB-STATCOM ......................................... 130
5.5.1. BtB-STATCOM Interacting with Power System.............................. 130
5.5.2. BtB-STATCOM Controller Design ................................................. 132
5.5.3.1. Case1: Start-up Transients ................................................... 133
5.5.3.2. Case 2: Response to Real Power Transfer Step Changes ...... 136
5.5.3.5. THD Content ....................................................................... 143
5.5.4. Discussion ....................................................................................... 144
5.6. Summary................................................................................................... 144
6. TRANSIENT STABILITY STUDIES ............................................................. 147
6.1. Introduction............................................................................................... 147
VII
6.2. Literature Survey on Transient Stability Studies ........................................ 148
6.3. Transient Stability Improvement Using GUPFC ........................................ 149
6.3.1. Dynamic Equations for Power Generation ....................................... 149
6.3.1.1. Wind Model ........................................................................ 149
6.3.1.2. Blade Dynamics................................................................... 150
6.3.1.3. Self-excited Double Cage Induction Generator .................... 151
6.3.1.4. Salient-Pole Synchronous Generator .................................... 151
6.3.2. Power System Configuration ........................................................... 152
6.3.3. Damping Control Scheme of GUPFC .............................................. 155
6.3.3.1. Fuzzy Damping Controller (FDC)........................................ 156
6.3.3.2. Fuzzified Gain Tuner (FGT) ................................................ 157
6.3.3.3. Tuning of Scaling Factors .................................................... 158
6.3.4.2. Case 2: Three-phase Fault with Longer Duration ................. 165
6.3.4.4. THD Content ....................................................................... 174
6.3.5. Discussion ....................................................................................... 174
6.4. Transient Stability Improvement Using IPFC ............................................ 175
6.4.2. Tuning of Scaling Factors ................................................................ 178
6.4.3.2. Case 2: Two-phase to Ground Fault ..................................... 180
6.4.3.4. THD Content ....................................................................... 191
6.4.4. Discussion ....................................................................................... 192
6.5. Transient Stability Improvement using BtB-STATCOM ........................... 192
6.5.2.1. Case 1: Three-phase to Ground Fault at Generator Bus ........ 195
VIII
6.5.2.2. Case 2: Three-phase to Ground Fault at Infinite Bus ............ 198
6.5.2.3. THD Content ....................................................................... 201
6.5.3. Discussion ....................................................................................... 202
6.6. Summary................................................................................................... 202
7. CONCLUSIONS AND FUTURE WORK........................................................ 205
REFERENCES ..................................................................................................... 211
CIRRICULUM VITAE ........................................................................................ 227
APPENDIX A: Converter Design Data for Power Flow Studies ..... Hata! Yer işareti
tanımlanmamış.
APPENDIX B: Test Systems Data ...................... Hata! Yer işareti tanımlanmamış.
APPENDIX C: PI Controller Parameters ............ Hata! Yer işareti tanımlanmamış.
APPENDIX D: Derivation of Maximum Power Injections for BtB-STATCOM Hata!
Yer işareti tanımlanmamış.
APPENDIX E: Programming Scripts .................. Hata! Yer işareti tanımlanmamış.
IX
X
LIST OF TABLES PAGE
Table 2.1. Overview of major FACTS devices with their attributes ......................... 14
Table 3.1. Flexible configuration of the multi-converter FACTS device .................. 32
Table 3.2. Operating constraints of the multi-converter FACTS device ................... 34
Table 3.3. Power flow results for voltage magnitude regulation @ 1.0 pu ............... 44
Table 3.4. Power flow results for real power regulation of Line 4-5 ........................ 44
Table 3.5. Power flow results for reactive power flow regulation of Line 4-6 .......... 45
Table 3.6. Parameters of the UPFC under different power flow control strategies ... 49
Table 3.7. Parameters of the IPFC under different power flow control strategies ..... 51
Table 3.8. Parameters of the GUPFC under different power flow control strategies 52
Table 3.9. Q-V characteristics of the two converters ............................................... 54
Table 4.1. Number of pulse-generating circuits per multi-converter FACTS device 77
Table 5.1. Simplex optimized controller parameters of GUPFC .............................. 93
Table 5.2. THD values .......................................................................................... 108
Table 5.3. Rule base for ΔVQ................................................................................. 117
Table 5.4. Simplex optimized controller parameters of IPFC ................................ 121
Table 5.5. Quantitative performance analysis of different controllers .................... 129
Table 5.6. THD values in case of three control schemes ........................................ 130
Table 5.7. THD values .......................................................................................... 143
Table 6.1. Optimization results of scaling factors .................................................. 160
Table 6.2. THD values of power system bus voltages ........................................... 174
Table 6.3. Optimization results of scaling factors .................................................. 179
XI
XII
LIST OF FIGURES PAGE
Figure 2.1. Real power transfer between two buses ................................................... 7

Figure 2.2. Power-angle curve of a transmission line ................................................ 8
Figure 2.3. Voltage-power characteristics of Figure 2.1 (Kundur, 1994) ................... 9
Figure 2.4. Reactive power-voltage curves of Figure 2.1 (Kundur, 1994) ................ 10
Figure 2.5. Loadability characteristics of transmission lines (Zhang et al., 2006) .... 11
Figure 2.6. SVC configuration: (a) typical arrangement (b) VI characteristics ......... 15
Figure 2.7. TCSC configuration: (a) typical arrangement (b) z-α characteristics ...... 16
Figure 2.8. TCPAR configuration: (a) thyrsitor arrangement (b) vector diagrams .... 17
Figure 2.9. STATCOM configuration:(a) arrangement (b) operating modes ............ 18
Figure 2.10. SSSC configuration: (a) typical arrangement (b) operating modes ....... 19
Figure 2.11. UPFC configuration: (a) typical arrangement (b) operating modes ...... 20
Figure 2.12. IPFC configuration: (a) typical arrangement (b) operating mode ......... 21
Figure 2.13. GUPFC configuration: (a) typical arrangement (b) operating modes ... 22
Figure 2.14. BtB-STATCOM configuration with typical arrangement .................... 23
Figure 3.1. Generic multi-converter FACTS device ................................................ 31
Figure 3.2. Voltage source equivalent model of the generic FACTS device ............ 32
Figure 3.3. PSCAD models of shunt and series converters ...................................... 38
Figure 3.4. Control constraint implementation in PSCAD ....................................... 40
Figure 3.5. PSCAD model of the P-Q load connected at high voltage bus ............... 41
Figure 3.6. PSCAD model of WSCC 3-Machine 9-Bus System .............................. 43
Figure 3.7. P-Q Control planes of Line 4-6 obtained with UPFC ............................. 46
Figure 3.8. P-Q Control planes of Line 4-6 obtained with IPFC .............................. 46
Figure 3.9. P-Q control planes of Line 4-5 obtained with GUPFC ........................... 47
Figure 3.10. P-Q control planes of Line 4-6 obtained with GUPFC ......................... 48
Figure 3.11. PSCAD model of IEEE 14-Bus System............................................... 50
Figure 3.12. PSCAD model of 3-Machine 7-Bus System ........................................ 53
Figure 3.13. Comparative P-V curves of Bus 1 ....................................................... 55
Figure 3.14. Comparative P-V curves of Bus 3 ....................................................... 56
XIII
Figure 4.1. Power circuit of three-phase six-pulse VSC......................................... 61
Figure 4.2. Four quadrant VSC operation .............................................................. 62
Figure 4.3. Simulated phase-to-neutral voltage waveforms of six-pulse VSC ........ 63
Figure 4.4. Simulated phase-to-phase voltage waveforms of six-pulse VSC .......... 64
Figure 4.5. Gating signals of GTOs for 180-degrees conduction ........................... 65
Figure 4.6. Harmonic spectrum of VAB for six-pulse VSC ..................................... 65
Figure 4.7. Power circuit of three-phase twelve-pulse VSC ................................... 66
Figure 4.8. Simulated phase-to-phase voltage waveforms of twelve-pulse VSC .... 67
Figure 4.9. Harmonic spectrum of VAB for twelve-pulse VSC ............................... 68
Figure 4.10. Power circuit configuration of three-phase quasi multi-pulse VSC ...... 70
Figure 4.11. PSCAD implementation of ¼ of quasi multi-pulse VSC ...................... 71
Figure 4.12. PSCAD implementation of magnetic interfaces ................................... 72
Figure 4.13. PSCAD implementation of series coupling magnetic interface ............ 73
Figure 4.14. Voltage vectors of converters M and N in rotating reference frame ..... 74
Figure 4.15. PSCAD implementation of equations (4.5) and (4.6) ........................... 75
Figure 4.16. PSCAD implementation of switching logic for six-pulse VSC ............ 76
Figure 4.17. Simulated voltage waveforms of quasi 48-pulse VSC.......................... 78
Figure 4.18. Harmonic spectrum of VAB for quasi 48-pulse operation ..................... 78
Figure 4.19. Four quadrant operation of the proposed quasi multi-pulse VSC ......... 81
Figure 4.20. Flexible magnitude/phase angle controlled quasi multi-pulse VSC ...... 82
Figure 5.1. Flow chart of the simplex optimization method in PSCAD .................. 88
Figure 5.2. WSCC 3-Machine 9-Bus System embedded with GUPFC................... 89
Figure 5.3. Control loops of GUPFC ..................................................................... 90
Figure 5.4. PSCAD implementation of simplex method ........................................ 92
Figure 5.5. Convergence performance of cost function in simplex method ............ 93
Figure 5.6. Simulated waveforms of case 1 ........................................................... 97
Figure 5.7. Simulated waveforms of case 2 ........................................................... 99
Figure 5.8. Simulated waveforms of case 3 ......................................................... 102
Figure 5.11. 4-Machine 4-Bus System embedded with IPFC ................................. 110
XIV
Figure 5.12. PSCAD implementation of PI+DG controllers ................................... 113
Figure 5.13. PSCAD implementation of HFPI controller ........................................ 114
Figure 5.14. PSCAD-MATLAB interface ............................................................. 115
Figure 5.15. PSCAD implementation of SEPOCHDET ......................................... 115
Figure 5.16. Universe of Discourse ....................................................................... 116
Figure 5.17. MFs for FUDE output set .................................................................. 117
Figure 5.18. Control surfaces of the proposed FUDE ............................................ 118
Figure 5.19. Conceptual control configurations for the master VSC ........................ 119
Figure 5.20. Control scheme for the slave VSC ...................................................... 119
Figure 5.21. PSCAD implementation of simplex method ...................................... 120
Figure 5.22. Cost function minimization in simplex method ................................... 120
Figure 5.23. Dynamic performances of real power flow controllers ...................... 123
Figure 5.24. Dynamic performances of reactive power flow controllers ................ 125
Figure 5.25. Dynamic performance of real power flow controller for slave VSC ... 125
Figure 5.26. Dynamic performance of DC voltage controller for slave VSC ......... 126
Figure 5.27. d-q components of master VSC injected current ................................ 126
Figure 5.28. d-q components of master VSC voltage by HFPI controller............... 126
Figure 5.29. Anode-to-cathode voltage of a selected GTO in converter M ............ 127
Figure 5.30. Dynamic performances of real power flow controllers ...................... 127
Figure 5.31. Dynamic performances of reactive power flow controllers ................ 129
Figure 5.32. 3-Machine 7-Bus System embedded with BtB-STATCOM ............... 131
Figure 5.33. Control loops of BtB-STATCOM ..................................................... 132
Figure 6.1. Power system configuration embedded with GUPFC ........................ 153
Figure 6.3. Membership functions and fuzzy rules for STFDC ............................ 157
Figure 6.4. Control surfaces of the proposed STFDC .......................................... 158
Figure 6.5. PSCAD implementation of simplex method ...................................... 159
XV
Figure 6.6. Convergence performance of cost function in simplex method .......... 159
Figure 6.7. Simulated STFDC performance against three-phase fault .................. 164
Figure 6.8. Simulated voltage and current waveforms of GUPFC converters....... 165
Figure 6.9. Simulated phase shift angles (ɸM and ɸN) and one GTO voltage ....... 165
Figure 6.10. Simulated STFDC performance against longer three-phase fault ....... 169
Figure 6.11. Power fluctuations following three-phase fault .................................. 170
Figure 6.12. Simulated STFDC performance against single-phase to ground fault . 174
Figure 6.13. Two-Area System embedded with IPFC and its control scheme ........ 176
Figure 6.15. Cost function minimization for both FACTS devices ........................ 179
Figure 6.16. Simulated STFDC performance following three-phase fault .............. 184
Figure 6.17. Simulated voltage and current waveforms of IPFC converters ........... 184
Figure 6.19. Simulated STFDC performance against two-phase fault .................... 187
Figure 6.20. Simulated STFDC performance against single-phase fault ................ 191
Figure 6.21. Power system configuration embedded with BtB-STATCOM ........... 193
Figure 6.22. Simulated BtB-STATCOM performance in case 1 ............................ 197
Figure 6.23. Simulated voltage and current waveforms of the converters .............. 198
Figure 6.25. Simulated BtB-STATCOM performance in case 2 ............................ 201
XVI
LIST OF SYMBOLS
* : Complex conjugate
° : Degrees
µ(i) : Membership function of the consequent of rule i
A : Blade impact area
a1-3 : Scaling factors of STFDC
bi : Center of membership function of the consequent of rule i
C : DC link capacitance
Cp : Dimensionless power coefficient
D : Damping coefficient of the SG
dX/dt : First order time derivative of the variable X
e : Internal generated voltage of the SG
e(k) : Error at sample instant k
ES : Line-to-line rms voltage of the sending-end side bus
eX : X-axis of the internal generated voltage of the SG
eXN : Line-to-neutral voltage of phase X
H : Henry
iD : D-component of the current
IL : Transmission line current
Iph : Phase current flowing into the converter
iQ : Q-component of the current
J : Inertia of the SEDCIG
j : Square root of -1
k : Number of sampling instant
Kp : Proportional gain of the PI controller
Kw : Damping gain
L : Inductance
Lm : Mutual leakage inductance of the SEDCIG
M : inertia constant of the SG
n : Harmonic order
XVII
P : Number of pole pairs
Pe : Real power flow error
Pinj,m : Injected real power of converter m
Pinj,mref : Injected real power of converter m
Pline : Real power flow of the transmission line
Plineref : Reference value of the real power flow of the transmission line
Ploss,m : Real power loss of converter m
PR : Receiving-end real power
PRMAX : Maximum power transfer of receiving-end side at unity power factor
Ptransfer,m : Transmitted real power of converter m from other converter(s)
PW : Mechanical power extracted from the wind
Qe : Reactive power flow error
Qinj,m : Injected reactive power of converter m
Qinj,mref : Reference value of injected reactive power of converter m
Qline : Reactive power flow of the transmission line
Qlineref : Reference value of the reactive power flow of the transmission line
QR : Receiving-end reactive power
Ra : Armature resistance
RC : Common end-ring resistance of the SEDCIG
RL : Transmission line resistance
RLD : Per-phase resistance of the three-phase load
Rs : Internal resistance of the DC voltage source
s : Laplace operator
Sline : Complex power flow of the transmission line
t : Time
T : Total simulation time
TE : Electrical torque of the SG or SEDCIG
TL : Load torque of the SEDCIG
TM : Mechanical torque of the SG
TX : X-axis time constant of the SG
Vbus : Line-to-line rms voltage of local bus
XVIII
Vbusref : Reference value of line-to-line rms voltage of local bus
Vconv : Line-to-neutral rms voltage of converter
Vconv(max) : Maximum value of line-to-neutral rms voltage of converter
VD : D-component of the voltage
Vdc : DC link voltage
VDref : Reference value of the D-component voltage
VF : Average converter voltage
Vf : Excitation winding voltage of the SG
VM : Voltage phasor of converter M
vn : Peak value of nth voltage harmonic component
VN : Voltage phasor of converter N
VQ : Q-component of the voltage
VQref : Reference value of the Q-component voltage
VR : Line-to-line rms voltage of the receiving-end side bus
VS : Voltage phasor of the selected bus
Vse : Line-to-neutral rms voltage of series converter
Vsh : Line-to-neutral rms voltage of shunt converter
VW : Wind speed
VWB : Base or mean wind speed
VWG : Gust wind component
VWN : Noise wind component
VWR : Ramp wind component
VX : Voltage phasor of the quasi multi-pulse VSC
VXn : Phase X-to-neutral voltage of the converter
VXY : Phase X-to-Phase Y voltage of the converter
w : angular frequency
W : Watt
wa : Speed of the rotating arbitrary reference frame
wB : Blade angular velocity
wi : Speed of generator-i
wiref : Reference value of the speed of generator-i
XIX
XC : Capacitive reactance
XL : Transmission line reactance
XTCR : Impedance of TCR
XTCSC : Impedance of TCSC
XX : X-component of the reactance of the SG
α : Phase angle between VM and VX
β : Gain factor of the FGT
βp : Blade pitch angle
γ : Tip speed ratio
δ : Phase angle between VD and VX
δ : Rotor angle of the SG
ΔX : Rate of change of variable X
ζ : Thyristor firing angle (zeta)
θconv : Phase angle of line-to-neutral rms voltage of converter
θR : Phase angle of receiving-end side bus
θS : Phase angle of sending-end side bus
θse : Phase-angle of line-to-neutral rms voltage of series converter
θsh : Phase-angle of line-to-neutral rms voltage of shunt converter
μ : Micro
ρ : Air density
ΣX : Integral of variable X at sample instant k
τi : Integral time constant of the PI controller
φ : Flux linkage
ΦM : Phase angle of voltage phasor of converter M
ΦN : Phase angle of voltage phasor of converter N
Ω : Ohm
XX
LIST OF ABBREVIATIONS
AC : Alternating Current
BtB-STATCOM : Back-to-Back Static Synchronous Compensator
CPU : Central Processing Unit
DC : Direct Current
DFIG : Doubly Fed Induction Generator
EMTDC : Electromagnetic Transients including Direct Current
FACTS : Flexible Alternating Current Transmission Systems
FDC : Fuzzy Damping Controller
FGT : Fuzzified Gain Tuner
FUDE : Fuzzy Decoupler
GCT : Gate Commutated Thyristor
GTO : Gate Turn-off Thyristor
GUPFC : Generalized Unified Power Flow Controller
HFPI : Hybrid Fuzzy Proportional Integral
HVDC : High Voltage Direct Current
IAE : Integral Absolute Error
IEEE : Institute of Electrical and Electronics Engineers
IGCT : Integrated Gate Commutated Thyristor
IPFC : Interline Power Flow Controller
ISE : Integral Square Error
ITAE : Integral Time Absolute Error
MF : Membership Function
MSC : Mechanically Switched Capacitor
MSR : Mechanically Switched Reactor
MVA : Mega Volt-Ampere
NR : Newton-Raphson
PI : Proportional Integral
PI+DG : Proportional Integral Control with Decoupled Gains
PLL : Phase Lock Loop
XXI
P-Q : Real Power-Reactive Power
PSCAD : Power System Computer Aided Design
PU : Per unit
P-V : Real Power-Voltage
PWM : Pulse Width Modulation
Q-V : Reactive Power-Voltage
SEDCIG : Self-excited Double Cage Induction Generator
SEPOCHDET : Set-Point Change Detector
SG : Salient-Pole Synchronous Generator
SSR : Subsynchronous Resonance
SSSC : Static Series Synchronous Compensator
STATCOM : Static Synchronous Compensator
STFDC : Self-Tuning Fuzzy Damping Controller
SVC : Static Var Compensator
TCPAR : Thyristor Controlled Phase Angle Regulator
TCR : Thyristor Controlled Reactor
TCSC : Thyristor Controlled Series Capacitor/Compensator
THD : Total Harmonic Distortion
TSC : Thyristor Switched Capacitor
TSR : Thyristor Switched Reactor
UPFC : Unified Power Flow Controller
VAR : Volt Ampere Reactive
VI : Voltage-Current
VSC : Voltage Source Converter
WSCC : Western System Coordinated Council
XXII
XXIII
1. INTRODUCTION A. Mete VURAL
1. INTRODUCTION
With fast technological advances and rapid population growth, electrical

power demand has increased substantially over the last decades. This situation has
led to heavily stressed traditional power systems which require either network
expansion or network operation closer to its technical limits. Cost efficient solutions
are usually preferred over network expansion which is very limited due to
environmental reasons and high expenses. In many countries, authorization is hardly
given to build new transmission lines so that existing transmission equipment has to
be enforced to fulfill changing requirements.
On the other hand, deregulation brings the evolution towards a competitive
electricity market in which congestion occurs due to a violation of system operating
limits when the transmission network is unable to put up all of the desired
transactions. Efficient operation and planning of transmission grid are highly
required in case of congestion management which is considerably complex in such
competitive electricity market and necessitates better utilization of available power
system capacities and increasing available transfer capability.
Alternative generation facilities such as wind/solar energy based generation
systems which lead up distributed generation, have become an inevitable trend due to
the critical factors such as limited available primary energy resources used in conventional
power plants, fast increase in fuel prices, and increasing awareness of environmental
problems, such as global warming and pollution. Distributed electrical systems, on
the contrary of traditional power systems where generation side is physically located
apart from consumer side, bring complexity to achieve a larger stability margin and
greater operating flexibility of existing power systems.
Power system operation was not only affected in the past by stability related
problems, leading to unpredictable system behavior, but also in today stability
concerns are getting worse due to the changing operation requirements and
increasing system complexity. Cost constraints have also become much tighter than
in the past, increasing the operational complexity considerably.
1
For the factors described above, it becomes evident that the operation of
power system structure under great changes is a complex and challenging
engineering task which requires efficient use of all power system elements without
disturbing technical operational limits and power systems with increasing complexity
are highly expected to be fast and real-time controlled to fulfill the requirements for
providing uninterrupted and reliable electrical energy to customers in the event of
generation and transmission outages.
Flexible Alternating Current Transmission Systems (FACTS) emerge as
“power electronic based solution” using solid-state switching devices and modern
control algorithms to increase controllability and enhance power transfer capacity of
existing transmission network. FACTS have gained greater interest during the last
decades due to the deregulation and restructuring strategies of power systems.
FACTS concept was originally proposed and conceptualized by Narain G.
Hingorani (Hingorani, 1988) and later defined formally as “alternating current
transmission systems incorporating power electronic-based and other static
controllers to enhance controllability and increase power transfer capability” by the
FACTS Terms & Definitions Task Force FACTS Working Group of the direct
current (DC) and FACTS subcommittee of Institute of Electrical and Electronics
Engineers (IEEE) (IEEE, 1997).
With the increase of voltage and current ratings of solid-state power
semiconductor devices, power electronics technology has penetrated into the area of
high voltage transmission in terms of FACTS devices (controllers) receiving great
attention to enhance power system operation by controlling one or more power
system parameters simultaneously and independently (Hingorani, 2000). During the
last two decades, FACTS devices have been proposed to enhance steady-state (static)
performance of power systems, such as increase of transmission line capacity, real
and reactive power flow control, loop-flow control, load sharing among parallel
corridors, voltage regulation, congestion management, and optimal power flow for
economic power system operation. The examples of dynamic performance
improvement include; enhancement of small signal stability and transient stability of
2
power systems by damping out oscillations, fast reactive power support for dynamic
voltage control, maintaining voltage stability, and power quality improvement.
1.1. Motivation for the Thesis
Since the time when FACTS devices were first proposed, modeling and
control of different FACTS devices have been broadly studied. In particular, there
are extensively research results covering a wide range of applications of single-
converter FACTS devices, such as Static Synchronous Compensator (STATCOM)
and Static Series Synchronous Compensator (SSSC) in literature. Multi-converter
FACTS devices, on the other hand, has emerged as a new opportunity to cope with
the aforementioned power system problems by controlling multiple power system
variables simultaneously and independently. However, the research for multi-
converter FACTS devices is relatively narrow and limited.
There exists a lack in realistic converter models for high power high voltage
applications which takes switching of semiconductor devices into account. Generally
the studies rely on two approaches. In the first approach, a set of linearized equations
are derived using fundamental frequency model of each converter of the FACTS
device. Here the converter is modeled as controllable voltage or current source
operating with fundamental system frequency (50 or 60 Hz), under the assumption
that the harmonics are neglected. This approach may be useful for steady-state or
power flow studies. In the second approach, six-pulse elementary converters
switched at frequencies relatively higher than the system frequency are used to
approximate harmonic content and converter modulation techniques. This approach
can suffer from high switching frequencies and relatively simple converter structure
which are not suitable for high power applications. This situation has motivated to
take a deep glance into the analysis of multi-converter FACTS devices including
more realistic converter models and their advanced controls.
Multi-converter FACTS devices are multi-input multi-output non-linear
systems with operational constraints which require advanced control algorithms. On
the other hand, fuzzy set theory presents good characteristics to address complex
3
control problems and has already proven to be efficient in several planning, control,
and operation problems in power systems. The fact that the need for computational
intelligence based control techniques including optimization methods is
indispensable for high performance control of the multi-converter FACTS devices
has also motivated this work.
The need for efficient utilization of power systems is increasing day by day in
the world and in Turkey. Besides these global conditions, there is not a well-shaped
research background on multi-converter FACTS devices in Turkey. This study will
provide a strong background on this subject.
1.2. Objectives of the Thesis
The objectives of the thesis are as follows:
• To provide appropriate models for both single- and multi-converter

FACTS devices for steady-state or power flow studies,
• To design high power quasi multi-pulse voltage source converter for the
simulation and analysis of both single- and multi-converter FACTS
devices,
• To provide converter-level models of the following multi-converter
FACTS devices: Generalized Unified Power Flow Controller (GUPFC),
Interline Power Flow Controller (IPFC), and Back-to-Back STATCOM
(BtB-STATCOM) using high power quasi multi-pulse voltage source
converters with switching and control schemes for dynamic and stability
studies,
• To design real and reactive power flow controllers for IPFC with improved
decoupling function and enhanced dynamic performance,
• To design damping controller for GUPFC to improve the transient stability
of the wind farm integrated power system,
• To design damping controller for IPFC to improve the transient stability of
the power system by damping out inter-area mode of oscillations.
4
1.3. Contributions of the Thesis
The contributions of this work are as follows:

• A new method of representing single- and multi-converter FACTS devices
in simulation program for studying steady-state behavior of power systems
embedded with FACTS devices is proposed, designed, and tested.
• A quasi multi-pulse voltage source converter with two control degrees of
freedom and switching at fundamental system frequency is designed for
converter-level modeling studies of the multi-converter FACTS devices.
• A novel decoupled control scheme for real and reactive power flow control
loops based on a hybrid fuzzy PI controller for IPFC is proposed, designed
and tested.
• A novel simplex optimized self-tuning fuzzy damping controller for
GUPFC for transient stability enhancement of the wind farm integrated
power system by damping synchronous generator oscillations and
increasing speed stability of the induction generators, is proposed,
designed and tested.
• The proposed self-tuning fuzzy damping controller for GUPFC is adapted
for IPFC for transient stability enhancement of the power systems by
suppressing inter-area mode of oscillations.
1.4. General Outline
This thesis is structured in seven chapters. After this introductory chapter,

Chapter 2 provides a summarized review of power system problems, FACTS devices
and their classification, recent advances in power electronics for FACTS concept,
and some FACTS application examples. Chapter 3 addresses steady-state modeling
of converter based FACTS devices for power flow studies in Power System
Computer Aided Design (PSCAD) simulation program with version 4.2.1 (PSCAD,
2010). The modeling approach is verified with simulated cases on WSCC 3-Machine
9-Bus System, 3-Machine 7-Bus System, and IEEE 14-Bus System. Particular
5
steady-state performance comparison is also made between the mentioned FACTS

devices in Chapter 3. The design details of quasi multi-pulse converter including
power circuit, magnetic interface, and pulse generating circuit for the power
semiconductor devices are given in Chapter 4. Chapter 5 presents the dynamic
models of GUPFC, IPFC, and BtB-STATCOM. The dynamic control characteristics
of these devices are investigated using the developed PSCAD based converter-level
models. The simulation studies are carried out and extensively analyzed in various
test power systems. A hybrid fuzzy PI controller is proposed to get high decoupling
performance between real and reactive power flow controllers of the IPFC. The
proposed controller is compared with the parameter optimized PI controllers and the
parameter optimized PI controllers having analytically calculated feed-forward
decoupling gains. The comparative simulation studies are carried out on 4-Machine
4-Bus power system through a number of case studies. Chapter 6 deals with the
transient stability studies of GUPFC, IPFC, and BtB-STATCOM. Different
disturbance scenarios are studied to compare the performance of the proposed
damping scheme of GUPFC with that of conventional approach. The proposed
damping scheme is also adapted for IPFC to damp inter-area mode of oscillations.
The transient stability enhancement of BtB-STATCOM is also investigated in
Chapter 6. In Chapter 7, important conclusions of this work are presented and future
work options on multi-converter FACTS devices are offered. References, curriculum
vitae of the author, and the appendices are given at the end.
6
2. OVERVIEW OF FACTS DEVICES A. Mete VURAL
2. OVERVIEW OF FACTS DEVICES
2.1. Background on Alternating Current Power Transmission
2.1.1. Thermal Limit
Thermal limit of a transmission line is defined in terms of the maximum

current carrying capacity (ampacity). The excess amount of current flowing on the
line produces heat leading to undesirable results, such as annealing and gradual loss
of mechanical strength of the conductor caused by temperature extremes and increase
sag and decreased clearance to ground due to conductor expansion at higher
temperatures (Kundur, 1994). So the transmission line can be utilized best only if it
is loaded up to its thermal limit which cannot be done normally without line
compensation.
2.1.2. Maximum Power Transfer
Electrical power transfer from generation side to consumer side is preferred in

alternating current (AC) form through overhead transmission lines due to flexibility
and cost. By referring to Figure 2.1, the real (active) power flow between two
arbitrary buses in a power system can be expressed in equation (2.1) (Kundur, 1994).
Figure 2.1. Real power transfer between two buses
E S VR
PR = sin(θ S − θ R ) (2.1)
XL
7
where PR and QR are the real and reactive power flow into Bus 2, respectively. ES and
VR are the voltage magnitudes of Buses 1 and 2, respectively. θS and θR denote phase
angles of Buses 1 and 2, respectively. XL is the reactance of the transmission line
having negligible resistance and capacitance. From equation (2.1) a non-linear
power-angle relationship can be obtained as in Figure 2.2 assuming fixed XL and
fixed bus voltage magnitudes. There is a maximum limit of transmitted power when
phase shift is 90°. Under fixed bus voltages, a suitable FACTS device can increase
maximum limit of the transmitted power further by line compensation, i.e., reducing
XL effectively.
Figure 2.2. Power-angle curve of a transmission line
2.1.3. Angle Stability
Power-angle curve in Figure 2.2 can be used to describe roughly the angle
stability of the generators in a power system without making classification. Under
steady-state conditions, there is equilibrium between input mechanical power and
output electrical power of each synchronous generator in an interconnected system
which leads to constant speed operation. When the system is perturbed for instance a
fault occurs, this equilibrium is upset, resulting in accelerating or decelerating of the
8
rotors of the machines. If the machine connected at Bus 1 runs faster transiently than
the other one connected at Bus 2, phase shift in equation (2.1) increases which result
in an increase of real power transfer from Bus 1 to Bus 2 acting to reduce speed error
between the machines. When phase shift increases further beyond a certain limit, real
power transfer decreases which can lead to unstable operation. Fast and robust
control algorithms for FACTS devices can solve the stability problem by real-time
control of XL and/or phase shift.
2.1.4. Voltage Stability
Voltage stability is defined as “the ability of a power system to maintain

steady acceptable voltages at all buses in the system under normal operating
conditions and after being subjected to a disturbance” (Kundur, 1994). In another
words, voltage stability is the ability of a power system to meet its reactive power
demand. The simple system in Figure 2.1 can be used to describe voltage stability
problem in its simplest form assuming Bus 1 represents a large system that transfers
real and reactive power through a transmission line to a load area, represented by Bus
2. The curves of the relationship between PR and VR for various load power factors
are shown in Figure 2.3.
Figure 2.3. Voltage-power characteristics of Figure 2.1 (Kundur, 1994)
9
Due to the fact that voltage drop in the transmission line is a function of real
as well as reactive power flow, load power factor is prominent on the power-voltage
curves of the system. For a given power factor, real power can be transferred at two
different voltage levels. The voltage stable operation is above the dashed line
denoting locus of critical points. In another words, the system is voltage stable only
if the load bus voltage VR is near to 1.0 per-unit (pu).
Figure 2.4 shows QR-VR curves at Bus 2 for a fixed value of PR. Voltage
stability limit is reached at the critical point where dQR/dVR reaches zero. The system
is voltage stable at the right side of the locus of critical points where dQR/dVR is
positive. Stable operation at the left side of the locus of critical points can be
achieved effectively using reactive power compensation with a FACTS device
having sufficiently control range.
Figure 2.4. Reactive power-voltage curves of Figure 2.1 (Kundur, 1994)
2.1.5. Transmission Line Loadability Characteristics
Reliable and efficient power transfer capability of a transmission line is

directly influenced by ampacity, voltage and angle stability which define loadability
characteristics of the line. Figure 2.5 shows operational characteristics of the
10
transmission lines for different voltage levels. In ideal, the usage of the transmission
line for real power transmission is up to its thermal limit. As line length increases,
voltage and angle stability limits determine line loading. These limits can be shifted
upward, up to the thermal limit by means of utilizing appropriate FACTS devices. It
is clear that the more line length, the more opportunity for the utilization of FACTS
devices. Needs, benefits, and the practical requirements should be examined together
to justify the investment into the appropriate FACTS device.
Figure 2.5. Loadability characteristics of transmission lines (Zhang et al., 2006)
2.2. Classification of FACTS Devices
Inherent limitations on AC power transmission mentioned in the previous

sections point to the problems of maintaining economic and secure operation of large
interconnected systems. These problems can be overcome with sufficient operation
margins in the power transfer that can be maintained by the introduction of fast
dynamic control of both real and reactive power using different types of FACTS
devices.
11
Depending on the connection type to the power network, FACTS devices can
be divided into five categories:
• Shunt FACTS devices

• Series FACTS devices
• Combined Shunt-Shunt FACTS devices
• Combined Shunt-Series FACTS devices
• Combined Series-Series FACTS devices
Depending on the switching properties of the power semiconductor devices,

FACTS devices can be categorized into two generations:
• First generation or conventional FACTS devices with response times of

about 2-3 cycles using thyristor with only ignition controlled by a gate
• Second generation or converter based FACTS devices with faster response
times of about 1-2 cycles using power semiconductor with both ignition
and extinction controlled by a gate
Second generation FACTS devices usually employ voltage source converter

(VSC) based configurations which form the basis of converter based FACTS devices
due to economy and performance (Gyugyi, 2000). Moreover, second generation
FACTS devices have two fundamental advantages over first generation FACTS
devices:
• They employ self-commutated inverters (converters) as synchronous

voltage sources which can internally generate or absorb reactive power
without requiring bulky AC capacitors or reactors.
• They can manage both real and reactive power independently.
Second generation FACTS devices can be further classified according to the

number of converters being utilized:
12
• Single-converter FACTS devices

• Multi-converter FACTS devices
Combining the above mentioned classification configurations, major FACTS

devices applied at the transmission level and their main attributes are listed in Table
2.1 ignoring relative performance comparison and applications for power quality
improvement and those for the distribution level. With the installation of energy
storage units (super capacitor, battery, fuel cell, superconducting magnetic energy
storage, etc.) in parallel to the DC link capacitors, depending on the storage size,
converter based FACTS devices gain real power generation/absorption abilities
which can further perform frequency regulation (Divya et al., 2009).
2.2.1. Static VAR Compensator (SVC)
SVC is a static VAR generator or absorber by injecting capacitive or

inductive current to maintain or control bus voltage or other power system variables
(IEEE, 1994). Although SVC has different configurations in detail, Figure 2.6 shows
a typical SVC configuration along with VI characteristics with thyristor
controlled/switched reactor (TCR/TSR), thyristor switched capacitor (TCS),
mechanically switched reactor (MSR), and mechanically switched capacitor (MSC)
(Hingorani et al., 2000). TCR/TSR is utilized for absorbing reactive power and TSC
is utilized for generating reactive power. The required reactive power is varied by the
coordinated control of the combination of these branches. SVC gives a smoother and
more precise response when compared with the mechanically switched
compensation.
13
Table 2.1. Overview of major FACTS devices with their attributes

FACTS device Connection Attributes
First generation
Voltage control, VAR
Static VAR compensator (SVC) Shunt compensation, voltage stability,
power oscillation damping
Power flow control, voltage
control, voltage stability, series
Thyristor controlled series impedance control, power
Series
capacitor/compensator (TCSC) oscillation damping,
subsynchronous resonance (SSR)
mitigation
Power flow control, phase angle
Thyristor controlled phase angle Combined control, voltage control, power
regulator (TCPAR) Shunt-Series oscillation damping, mitigation of
(SSR)
Second generation
Voltage control, VAR
Single-converter
Static synchronous
Shunt compensation, voltage stability,
compensator (STATCOM)
power oscillation damping
Static series synchronous control, voltage stability, VAR
Series
compensator (SSSC) compensation, power oscillation
damping, SSR mitigation
Unified power flow Combined control, voltage stability, VAR
controller (UPFC) Shunt-Series compensation, power oscillation
damping, SSR mitigation
Multi-line power flow control,
voltage control, voltage stability,
Interline power flow Combined
VAR compensation, power
controller (IPFC) Series-Series
Multi-converter
oscillation damping, SSR

mitigation
Multi-line power flow control,
voltage control, voltage stability,
Generalized unified power Combined
VAR compensation, power
flow controller (GUPFC) Shunt-Series
oscillation damping, SSR
mitigation
Real power transfer, voltage

Back-to-Back STATCOM Combined control, VAR compensation,
(BtB-STATCOM) Shunt-Shunt voltage stability, power
oscillation damping
14
Figure 2.6. SVC configuration: (a) typical arrangement (b) VI characteristics
The required rating and the specification of the SVC are determined
according to the VI characteristics of the SVC shown in Figure 2.6b. Since the
reactive power of the capacitor is directly proportional to the system voltage, a sharp
reduction of reactive power support at large voltage drops is observed during some
severe contingencies (Hingorani et al., 2000). This situation is the major drawback of
SVC applications for voltage support in power systems.
2.2.2. Thyristor Controlled Series Capacitor/Compensator (TCSC)
TCSC typically consists of a TCR in parallel with a capacitor to vary

effectively transmission line reactance XL in equation (2.1) mainly for power flow
control and power oscillation damping. TCSC configuration is shown in Figure 2.7
where it is located in series with the transmission line (Hingorani et al., 2000).
Operation principle of TCSC is to provide a variable capacitor in a continuous
manner by means of controlling the effective reactance of the TCR with the thyristor
firing angle ζ (zeta) which is measured from the zero crossing of the line current. At
fundamental system frequency TCR becomes variable reactive impedance whose
equivalent is given in equation (2.2) (Hingorani et al., 2000).
15
Figure 2.7. TCSC configuration: (a) typical arrangement (b) z-α characteristics
π
X TCR (ς ) = X TCR (2.2)
π − 2ς − sin ς
where XTCR=wL, and XTCR≤ XTCR(ζ)≤∞. The controllable steady-state impedance of the
TCSC at system fundamental frequency is obtained as
X C X TCR (ς )
X TCSC (ς ) = (2.3)
X TCR (ς ) − X C
where XC=1/wC. From Figure 2.7b, the resonance region is inhibited for ζ1≤ ζ≤ ζ2
where XTCR(ζ) = XC. XTCSC is generally kept below XL to avoid over-compensation of
the transmission line.
2.2.3. Thyristor Controlled Phase Angle Regulator (TCPAR)
TCPAR, also known either as thyristor controlled phase shifting transformer or

static phase shifting transformer is a replacement of phase shifting transformer for
power flow control in a transmission line in steady-state conditions (Padiyar, 2007).
The use of thyristors enables fast action to obtain a rapidly varying phase angle for
dynamic regulation of power flow and stability improvement. TCPAR configuration is
shown in Figure 2.8 (Hingorani et al., 2000). Phase shifting is attained by adding
relatively small voltage vector Δv having ±90° phase shift relative to the system
16
voltage v as illustrated in Figure 2.8b. This small voltage vector resulted from the other
two phases via shunt transformers is inserted in series with the transmission line. With
this addition phase angle of the system voltage is varied. Thyristor switching enables
relatively small angular adjustments making resultant angular change approximately
proportional to the injected voltage, while the magnitude of system voltage remains
almost constant (Hingorani et al., 2000).
Figure 2.8. TCPAR configuration: (a) thyrsitor arrangement (b) vector diagrams
2.2.4. Static Synchronous Compensator (STATCOM)
STATCOM is an advanced SVC with a VSC instead of controllable reactors

and switched capacitors. STATCOM has many advantages over SVC that
STATCOM has faster response, requires less space, being modular, and it can be
interfaced with energy storage units. STATCOM configuration is shown in Figure
2.9 (Singh et al., 2009). During low voltage conditions STATCOM shows its
superiority as the magnitude of the supplied reactive current is independent of the
system voltage. However in SVC, the capacitive current drops linearly as system
voltage reduces when high capacitive current is highly required. AC output of the
converter, VSTATCOM is proportional to the DC link voltage Vdc. By varying VSTATCOM
the reactive current is can be varied. Phase angle between VSTATCOM and VB is zero
when neglecting losses (Padiyar, 2007).
17
Figure 2.9. STATCOM configuration:(a) arrangement (b) operating modes
2.2.5. Static Series Synchronous Compensator (SSSC)
SSSC is a series VSC based FACTS device to produce a controllable voltage

in quadrature with the line current for power flow or bus voltage control (Gyugyi,
1997). SSSC has many advantages over TCSC that bulky passive components such
as capacitors and reactors are eliminated. SSSC improves technical characteristics
including symmetric capability in both inductive and capacitive operating modes.
Moreover connecting an energy storage unit on the DC link is possible to exchange
real power with the power system. Series injected voltage is controlled independent
of the line current to effectively change the overall reactive voltage drop across the
transmission line. Figure 2.10a depicts a typical SSSC arrangement (Sen, 1998). If
VSSSC lags line current by 90°, VSSSC becomes a capacitive voltage, and if VSSSC leads
line current by 90°, VSSSC becomes an inductive voltage. These two possible
18
operating modes emulate equivalent inductive or capacitive reactance in series with

the line as denoted in Figure 2.10b while ignoring losses (Sen, 1998). By controlling
the magnitude of VSSSC up to its maximum allowable limit, the amount of series
compensation can be fully adjusted.
Figure 2.10. SSSC configuration: (a) typical arrangement (b) operating modes
2.2.6. Unified Power Flow Controller (UPFC)
UPFC is a versatile multi-converter FACTS device which can be regarded as

the first example to this class. UPFC has been the most researched subject of all
multi-converter FACTS devices to meet various control objectives of the electric
power industry. UPFC configuration is shown in Figure 2.11 which was proposed by
Gyugyi as the combination of STATCOM and SSSC (Gyugyi, 1992). In UPFC, the
two VSCs are coupled through a common DC link to allow bi-directional real power
transfer between the VSCs. The capability of real power exchange enables
multifunctional flexibility so that the series voltage generated by the series VSC can be
injected into the transmission line with controllable magnitude (0≤Vpq≤Vpqmax) and
desired phase angle (0°≤θ≤360°) without requiring any external power source. This
feature provides simultaneous or selective control of terminal voltage regulation,
series compensation, and phase shifting for independent real and reactive power flow
control. For instance, terminal voltage regulation can be attained if Vpq is inserted
19
having phase angle equal to that of VS. Series compensation can be applied if Vpq is
inserted having phase angle that leads or lags 90° by IL. Transmission angle can be
shifted if Vpq is inserted such that desired phase shift is obtained without any change
in magnitude. Flexible operation mode shown in Figure 2.11b yields independent and
simultaneous control of real and reactive power flow on the line which cannot be
attained by single-converter FACTS devices. UPFC can also control bus voltage
where its shunt VSC is connected by reactive power injection (Gyugyi, 1995).
Figure 2.11. UPFC configuration: (a) typical arrangement (b) operating modes
2.2.7. Interline Power Flow Controller (IPFC)
IPFC employs a number of series VSCs with a common DC link, each of

which is connected to the transmission line via series coupling transformer to provide
series compensation for a selected line of a multi-line substation (Gyugyi et al.,
1999). The simplest IPFC configuration having two VSCs is shown in Figure 2.12a.
Generally one VSC, for example lower VSC is assigned for flexible compensation,
as shown in Figure 2.12b, by injecting full controllable voltage into the line for
independent real and reactive power flow control.
20
Figure 2.12. IPFC configuration: (a) typical arrangement (b) operating mode
This feature is not possible either in TCSC or SSSC in which only real power
flow can be controlled. Upper VSC regulates DC link voltage by balancing real
power between VSCs and at the same time it can regulate real or reactive power flow
on the line where it is being coupled. Moreover IPFC can serve like a virtual
transmission line so that an overloaded line can be relieved by forwarding real power
flow to the underloaded line. Although IPFC and UPFC have the same number of
control degrees of freedom, IPFC has received less attention generally in literature
when compared with the UPFC based studies.
2.2.8. Generalized Unified Power Flow Controller (GUPFC)
GUPFC is the extended version of UPFC with the addition of one or more
series VSC to increase power system controllability (Fardanesh et al., 2000). GUPFC
extends the concept of power flow and voltage control beyond that is achievable with
either UPFC or IPFC. GUPFC having the simplest structure consists of one shunt
VSC and two series VSCs as shown in Figure 2.13. Series VSCs can exchange real
21
power with two transmission lines to inject controllable voltages Vpq1 and Vpq2 with
full angle control (0°≤θ1≤360°, 0°≤θ2≤360°) that cannot be attained either UPFC or
IPFC. Shunt VSC both supports real power requirements of the series VCSs via
common DC link and provide voltage support at the bus where it is being connected.
GUPFC can control real and reactive power flows of the two parallel transmission
lines as well as bus voltage simultaneously and independently, hence it has stronger
control capabilities than UPFC. To add extra control degrees of freedom, the number
of series VSCs can be increased to control more power flows at the same time. To
relieve congestions, GUPFC may be installed in a substation to manage power flows
of multi-lines or a group of lines and provide voltage support as well. Although the
concept is not new, GUPFC has not gained much interest in literature.
Figure 2.13. GUPFC configuration: (a) typical arrangement (b) operating modes
2.2.9. Back-to-Back STATCOM (BtB-STATCOM)
BtB-STATCOM is a multi-converter FACTS device constructed by joining

two separate STATCOMs at their DC link (Larsson et al., 2001), (Reed et al., 2003).
22
In literature, instead of “BtB-STATCOM” as the compensator name, “BtB DC link”

or “VSC based BtB high voltage DC (HVDC) transmission link” are the alternative
definitions (Tyagi et al., 2006), (Parkhidehet al., 2009), (Xinghao et al., 2009), (Liu
et al., 2010). The concept can also be described as VSC based HVDC without long
transmission network. As shown in Figure 2.14, each STATCOM is connected in
parallel to the system bus via shunt coupling transformer. Common DC link provides
bi-directional real power transfer between two AC grids (synchronous or
asynchronous or even with different frequencies). In addition each STATCOM can
provide independent reactive power support for dynamic voltage control. It is
observed that there are few papers regarding BtB-STATCOM in literature. In this
research, the capabilities of BtB-STATCOM is investigated for dynamic bus voltage
control, real power transfer capability, and damping out oscillations caused by severe
disturbances.
Figure 2.14. BtB-STATCOM configuration with typical arrangement
2.3. More Control Degrees of Freedom
The reactive power flow on the line in Figure 2.1 can be written in equation
(2.4). The effective change in either or the combination of the line impedance and
23
phase angles by the applied compensation does not only change real power flow
(equation (2.1)) but also reactive power flow is varied as well (Kundur, 1994).
Operational characteristics of UPFC, IPFC, and GUPFC can overcome this natural
real power-reactive power (P-Q) coupling phenomenon so that independent real and
reactive power flow control can be provided simultaneously. This feature cannot be
attained either by a conventional or a single-converter FACTS device.
E S VR
QR = (1 − cos(θ S − θ R )) (2.4)
XL
Independent P-Q control feature also regulates X/R ratio of the transmission
line indirectly in which conventional series compensators such as fixed capacitor,
TCSC, or SSSC only controls real power flow by varying line reactance.
Conventional series compensation which are unable to control reactive power flow
reduces only X, thus, X/R ratio is distorted significantly in which excessive amounts
of reactive power flows are observed on the compensated lines which increase line
losses significantly. Multi-converter FACTS devices, on the other hand, compensate
against resistive line voltage drop so that effective value of R is also controlled to get
a balanced X/R ratio (Hingorani et al., 2000).
Control degrees of freedom of the FACTS devices can be defined as 2n-1
where n represents number of converters being utilized. For instance, the simplest
IPFC and the simplest GUPFC can control three and five power system parameters in
a simultaneous manner, respectively.
Total MVA rating, roughly the sum of individual ratings of hardware
elements such as high power converters and coupling magnetic interface, is
effectively used in a multi-converter FACTS device. For instance, when the
operation of individual STATCOM plus SSSC is compared with that of UPFC with
the same MVA rating, the latter FACTS device provides an additional control
capability, that is the capability to control the reactive flow on the transmission line.
The independent operation of STATCOM and SSSC cannot provide this flexibility
24
even the sum of MVA ratings of STATCOM and SSSC is equal to that of UPFC. In
this regard, UPFC provides more control capabilities to the power system.
2.4. Recent Advances in Power Semiconductors
The development of FACTS devices has continued to grow with the growing
capabilities of power semiconductors. Silicon based thyristors is widely used in first
generation FACTS devices which have been present for several decades with voltage
rating up to 11.0 kV (Chakraborty, 2011). The applications of second generation
FACTS devices have been implemented using high power converters ranging from
10 MVA to 250 MVA. Gate turn-off thyristor (GTO), gate commutated thyristor
(GCT), and integrated gate commutated thyristor (IGCT) are the common options
with switching frequencies up to a few kHz. Silicon based GTOs have current rating
up to 10 kA with voltage rating up to 9.0 kV (Chakraborty, 2011). GCT with ratings
6 kV and 6 kA has proven itself for high power converter applications for
STATCOM (Reed et al., 2001). The performance and electrical rating of IGCT has
increased dramatically in recent years. IGCTs with ratings 4.5-10 kV and 4.0 kA-6.5
kA are available in the market (Yongsug et al., 2009). IGCT does not require snubber
circuits and has better turn-off characteristics, lower conducting and switching loss,
and simpler gate control compared with GTO. It finds an application area of high
power converters for wind power now, and seems to be a future option for extensive
application prospect, including FACTS devices (Chengsheng et al., 2009). Ongoing
semiconductor research for the next decade seems to cover mainly silicon carbide
and gallium nitride materials to increase suitability and broadened applications of
semiconductor devices in mega-watt range systems (Vobecky, 2011).
2.5. Field Applications of FACTS Devices at Transmission Level
Although there are numerous successful FACTS installation examples

including first generation, some application examples of the second generation
25
FACTS devices applied at the transmission level are listed below for the sake of
highlight:
• ± 50 MVAR STATCOM has been put in service for a couple of years at

154 kV Sincan transformer substation in Ankara, Turkey for reactive
power compensation, terminal voltage regulation, and stability
enhancement (Gultekin et al., 2012).
• World’s first UPFC application is comprised of two ± 160 MVA
converters at Inez Substation in Kentucky, USA and has been in service
since 1998 for voltage support and power flow control (Renz et al., 1999).
• ± 100 MVAR STATCOM at Sullivan substation in North-Eastern
Tennessee, USA has been in service since 1995 (Schauder et al., 1997).
• 80 MVA-154 kV UPFC comprised of two ± 40 MVA converters at
Gangjin substation in Korea, has contributed to voltage stability
enhancement and power flow control since 2003 (Im et al., 2005).
• ± 200 MVA convertible static compensator, which can be configured as
either STATCOM, SSSC, UPFC, or IPFC, has been installed in 2001 at
345 kV Marcy substation in New York, USA for maximizing the use of
existing transmission network and improving voltage and power flow
control capabilities (Zelingher et al., 2000).
• +133/-41 MVA STATCOM has been in service at 115 kV Essex
substation near Burlington, USA since 2001 for dynamic reactive power
compensation for smooth voltage control over a wide range of operating
conditions (Reed et al., 2001.
• World’s largest STATCOM which has been planned at Toshin substation
in Japan in 2010 rated at 450MVA for stability improvement and dynamic
voltage control (Fujii et al., 2010).
• ±100 MVA STATCOM has been commissioned at Talega 138 kV
substation in California, USA since 2003 for dynamic reactive power
support during peak load conditions (Reed et al., 2002).
26
• ±72 MVA BtB-STATCOM consisting of two ± 36 MVA STATCOMs

have been installed for the first time in Eagle Pass substation and in
Piedras Negras substation to enable bi-directional real power transfer
between USA and Mexico as well as reactive power support for dynamic
voltage control at two distinct buses (Larsson et al., 2001).
2.6. Summary
This chapter has surveyed main technical limitations of AC power

transmission and the FACTS devices with their operational characteristics as a
solution option. In addition, recent advances in semiconductor technology
development and some real-world FACTS installation examples are reviewed and
summarized. It is emphasized that the multi-converter FACTS devices have more
control capabilities than single-converter FACTS devices. With more control degrees
of freedom by employing more converters, more control objectives can be achieved.
Reactive power flow control is only attainable with more converters coupled through
a common DC link. However GUPFC, IPFC, and BtB-STATCOM have received
less attention in literature when compared with STATCOM, SSSC, and UPFC. The
following next chapters will focus on modeling of these multi-converter FACTS
devices in both steady and dynamic states. It is evident that these FACTS devices
due to their special characteristics will play an essential role in the design and the
operation of the future electric power systems.
27
28
3. STEADY-STATE MODELING A. Mete VURAL
3. STEADY-STATE MODELING
3.1. Introduction
Power flow (load flow) studies which require steady-state modeling are
crucial for the design and performance analysis phases of the FACTS devices
embedded in power systems. The decisions and the future expansion options are
specified based on the results obtained from power flow studies. In this chapter, an
approach for the steady-state modeling of GUPFC, IPFC, and BtB-STATCOM for
power flow studies is presented.
In power flow studies of the IPFC, each VSC is generally modeled as pure
sinusoidal three-phase balanced voltage source whose magnitude and phase angle are
controlled. This voltage source is assumed to operate at fundamental system
frequency of 50 Hz or 60 Hz and connected to the transmission line in series with a
series reactance/impedance, which represents coupling transformer. Each coupling
transformer is modeled as pure inductive reactance if converter losses are neglected
(Xuan et al., 2004), (Xia et al., 2008), (Vasquez-Arnez et al., 2008). When the losses
of the converters are taken into account, each coupling transformer is modeled as
impedance having both resistive and reactive components (Zhang, 2003),
(Bhowmick et al., 2009), (Yankui et al., 2006), (Vinkovic et al., 2011), (Natália et
al., 2012). Alternatively, series voltage source is decomposed into direct and
quadrature components which facilitates the control of the source (Vasquez-Arnez et
al., 2008). Each series converter of the IPFC is modeled as controllable impedance
inserted into the compensated line in series (Fardanesh et al., 2004).
Power flow studies of the GUPFC assumes that the shunt and the series
converters can be represented as ideal voltage sources with series
reactances/impedances when losses are ignored (Padhy et al., 2005), (Vasquez-Arnez
et al., 2008). When losses are taken into account, each series transformer is modeled
as impedance having resistive and reactive components (Zhang et al., 2001), (Zhang
et al., 2004). Discrepantly, the shunt converter of the GUPFC is modeled as current
source with series connected reactance (Vasquez-Arnez et al., 2008).
29
Power flow studies of the BtB-STATCOM is limited and have been studied
under the concept of VSC based HVDC in which shunt converters are modeled as
voltage sources with series impedances including converter losses into account
(Zhang et al., 2004), (Pizano-Martinez et al., 2007).
The inclusion of voltage/current sources provides real and reactive power
exchange between the FACTS device and the power grid at newly added ghost buses
which forms the basis of power injection models of the FACTS devices.
Conventional Newton-Raphson (NR) power flow algorithm is then modified by the
user using the set of real and reactive power injections at the buses where the FACTS
devices are located. Generally the structure of the Jacobian matrix is preserved but
Jacobian matrix formation for IPFC and GUPFC is different by taking derivatives
with respect to real and imaginary parts of the line current as opposed to
conventional approach (Vinkovic et al., 2011).
Alternatively, NR solution algorithm is accomplished by a power system
analysis software package instead of writing the codes by the user. The power
injection model of the FACTS device is then defined in a user-defined model for
power flow studies of the power systems embedded with the FACTS devices
(Tümay et al., 2004), (Vural et al., 2007). In this chapter a different steady-state
modeling approach for power flow studies of GUPFC, IPFC, and BtB-STATCOM is
proposed in PSCAD neither requires a power injection based user-defined model nor
modification of NR codes by the user is required due to real and reactive power
injections caused by the FACTS device.
3.2. Proposed Steady-state Modeling Approach
The aforementioned modeling approaches in literature are powerful even for

relatively large power systems having many buses but suffer from the complexities
of the programming codes in NR power flow algorithm. However, the proposed
modeling approach is graphically implemented in PSCAD without editing NR codes
by the user. Instead, Electromagnetic Transients including Direct Current (EMTDC),
the simulation engine of PSCAD, handles the interaction of the FACTS device and
30
the power system and solves both operating and control constraint equations required
for the power flow study.
3.2.1. Configurable Multi-Converter FACTS Device
For the purpose of the study a generic configurable multi-converter FACTS

device having four converters with a common DC link is presented in Figure 3.1. By
turning on/off appropriate switches (S1-4) six types of FACTS devices in nine
different configurations can be obtained as listed in Table 3.1. VSC1 and VSC4 are
connected at Buses i and j via shunt coupling transformers Tr1 and Tr4, respectively.
Series converters VSC2 and VSC3 are connected to the transmission lines Lines m
and n, via series coupling transformers Tr2 and Tr3, respectively. Each VSC can
synthesize three-phase controllable AC voltage so that each VSC is modeled as a
sinusoidal voltage source with controllable magnitude (Vsh1-2,Vse1-2) and phase angle
(θsh1-2,θse1-2) as shown in Figure 3.2.
Figure 3.1. Generic multi-converter FACTS device
31
Table 3.1. Flexible configuration of the multi-converter FACTS device

Switch positions FACTS
Mode Control attribute
S1 S2 S3 S4 device
1 ON ON ON OFF STATCOM-1 Bus i voltage
2 ON ON OFF ON STATCOM-2 Bus j voltage
3 OFF OFF ON OFF SSSC-1 Line m real or reactive power flow
4 OFF ON OFF OFF SSSC-2 Line n real or reactive power flow
Buses i and j voltage + real power
5 ON ON ON ON BtB-STATCOM
transfer
Bus i voltage + Line m real and
6 ON OFF ON OFF UPFC-1
reactive power flows
Bus i voltage + Line n real and
7 ON ON OFF OFF UPFC-2
Line m real power flow + Line n
8 OFF OFF OFF OFF IPFC
real and reactive power flows
Bus i voltage + Line m/n real and
9 ON OFF OFF OFF GUPFC
Each voltage source is assumed to have the capabilities of independent

reactive power injection (Qinj1-4) and dependent real power injection (Pinj1-4) through
the transformers to the power system. DC link enables real power exchange between
converters (Ptransfer1-4) so that the sum of real power injections into the power system
is zero. Ploss1-4 is the sum of switching loss plus coupling transformer loss of each
related VSC, respectively.
Figure 3.2. Voltage source equivalent model of the generic FACTS device
32
3.2.2. Operating Constraints
Operating constraint of the FACTS device is mainly related with the

maximum voltage injection capability and the MVA rating of each commissioned
converter. A set of equations for voltage constraint in polar form of each converter
can be written in equation (3.1). Vconv is the generic symbol representing line-to-
neutral rms voltage of shunt/series converter (Vsh1-2, Vse1-2). θconv is the respective
phase angle of Vconv (θsh1-2, θse1-2).
0 ≤ Vconv ≤ Vconv(max) 
  (3.1)
 0 ≤ θ conv ≤ 2π 
Each converter should be fed from a constant DC link voltage for “voltage
source” based operation. DC link voltage Vdc is defined in equation (3.2) for single-
converter operation (Mode 1-4) and should be kept constant in steady-state. This
constraint is established by regulating Vdc to its reference that can be succeeded by a
closed-loop control scheme. In steady-state, time derivative of Vdc becomes zero and
equation (3.2) reduces to equation (3.3) for converter m.
dVdc
Pinj ,m + Ploss ,m = CV dc (3.2)
dt
Pinj ,m + Ploss ,m = 0
(3.3)
In multi-converter operation (Mode 5-9), equation (3.3) should be modified

since real power transfer occurs between converters. For converter m the constraint is
updated as given in equation (3.4).
Pinj , m − Ptransfer, m + Ploss,m = 0 (3.4)
33
Steady-state operating requirements of each configuration are listed in Table

3.2 based on real and reactive power interactions given in Figure 3.2 and equations
(3.3) and (3.4). Maximum amount of real and reactive power related with each
converter is a constraint and defined in terms of MVA converter rating in pu. Real
power constraint is derived on the fact that conservation of power. Loss meeting
function of the FACTS device can be theoretically assigned to any converter in a
number of ways. In single-VSC operation, each converter should meet losses itself
from the power system. In multi-VSC operation, overall losses can be met by only
single converter or any converter combination. For the sake of simplicity, only one
converter is assigned to meet overall losses of the multi-converter FACTS device.
Table 3.2. Operating constraints of the multi-converter FACTS device

Loss Constraint equations in pu
Mode
meeting Real power Reactive power Apparent power
1 VSC1 Pinj1 + Ploss1 = 0 -1.0 ≤ Qinj1 ≤ 1.0 (Pinj12 + Qinj12) ≤ 1.0
2 VSC4 Pinj4+ Ploss4 = 0 -1.0 ≤ Qinj4 ≤ 1.0 (Pinj42 + Qinj42) ≤ 1.0
Pinj1 – Ptransfer1 + ...
... + Ploss1 + Ploss4 = 0 -1.0 ≤ Qinj1 ≤ 1.0 (Pinj12 + Qinj12) ≤ 1.0
5 VSC1
Pinj4 – Ptransfer4 = 0 -1.0 ≤ Qinj4 ≤ 1.0 (Pinj42 + Qinj42) ≤ 1.0
Ptransfer1 + Ptransfer4 = 0
Pinj1 - Ptransfer1 + ...
6 VSC1
Pinj2 - Ptransfer2 = 0 -1.0 ≤ Qinj2 ≤ 1.0 (Pinj22 + Qinj22) ≤ 1.0
7 VSC1
8 VSC2
... + Ploss1 + Ploss2 + ... ...
+ Ploss3= 0 -1.0 ≤ Qinj1 ≤ 1.0 (Pinj12 + Qinj12) ≤ 1.0
9 VSC1 Pinj2 – Ptransfer2 = 0 -1.0 ≤ Qinj2 ≤ 1.0 (Pinj22 + Qinj22) ≤ 1.0
Ptransfer1 + Ptransfer2 +...
... + Ptransfer3 = 0
34
3.2.3. Control Constraints
3.2.3.1. Direct Control Mode
In direct control mode, the user sets the reference values of the real and
reactive power injections for each VSC directly in any operating mode of the FACTS
device. In steady-state, equation (3.5) can be written as a control constraint for
reactive power injection for converter m. This case is valid for all modes (Mode 1-9).
Qinj ,m − Qinj ,m ref = 0 (3.5)
When revealing real power injection constraint, reference value of the real power
injection by converter m should be equal to the power loss of the converter in steady-
state, as written in equation (3.6). This case is valid for single-converter operation
(Mode 1-4). In multi-converter operation (Mode 5-9), reference values of the desired
real power injections become dependent upon each other. For example for UPFC-1,
equation (3.7) is derived and written as real power injection constraints for the two
converters. Equation (3.7) can be modified as equations (3.8) and (3.9) for IPFC and
GUPFC, respectively. In direct control mode, the effects of real and reactive power
injections on power system variables, such as, real and reactive power flows, real and
reactive transmission losses, bus voltage profile, can be investigated based on power
injection concept.
Ploss ,m + Pinj ,m ref = 0 (3.6)
Ploss1 + Ploss 2 + Pinj ,1ref + P inj , 2 ref = 0 (3.7)
Ploss 2 + Ploss 3 + Pinj , 2 ref + P inj ,3 ref = 0 (3.8)
Ploss1 + Ploss 2 + Ploss 3 + Pinj ,1 ref + Pinj , 2 ref + P inj ,3 ref = 0 (3.9)
35
3.2.3.2. Indirect Control Mode
In indirect control mode, the user sets reference values of the power system
parameters such as, bus voltage and real and reactive power flows instead of direct
real and reactive power injections by VSCs. The bus voltage control constraint is
given in equation (3.10) and can be solved generally by the FACTS device having a
shunt VSC (Mode 1,2,5,6,7,9). Vbus is the voltage magnitude of the local bus, to
which shunt VSC is connected. Vbusref is the reference value of the voltage magnitude
of the local bus. In a similar manner, power flow control constraint pair given in
equation (3.11) is solved by the FACTS device having multi-converters (Mode 5-9).
Alternatively, only real or only reactive power flow constraint is required to be
solved merely, this can be established by the FACTS device having series converters
(Mode 3,4,6-9).
Vbus − Vbus ref = 0 (3.10)
 Pline − Pline ref = 0 

  (3.11)
Qline − Qline ref = 0
Apparent power constraint of any mode, given in the last column of Table 3.2, is
not supposed to be a control constraint either in direct/indirect control mode. It is not
come up to a reference value, instead it is observed explicitly and expected to be in
the limits of FACTS device rating. It can be observed that under which operating
conditions, violation of apparent power rating occurs. Real power constraints, given
in equations (3.6) and (3.9), should also be provided in indirect control mode. Either
in direct or indirect control mode, for a given control objective, required voltage
magnitude and phase angle of each converter are iteratively found in PSCAD by
updating the solution at each solution time step. Depending on the control
requirements these two control modes can be operated simultaneously.
36
3.3. Modeling in PSCAD
3.3.1. Power Circuit
PSCAD is well-accepted graphical based electromagnetic time domain

transient simulation environment and principally suited for simulating time domain
instantaneous responses of electrical systems. Power circuit of the FACTS device,
consisted of shunt/series converters and the coupling transformers is modeled using
standard available components of the PSCAD master library which is shown in
Figure 3.3. Each shunt/series converter is constructed using three-phase voltage
source model-2, whose AC voltage magnitude and phase angle are controlled
through external signals. Three-phase two-winding transformer is connected as
delta/wye which couples AC output of shunt converter with the high voltage of
system bus. Three identical single-phase two-winding transformers are used to inject
AC voltage of the series converter into the transmission line. The shunt and the series
converters are combined together to realize the respective FACTS device together
with its control circuit mentioned in the next section. Design data of the shunt and
series converters are given in Appendix A. The switches presented in Figure 3.1 are
implemented using three-phase breakers with single line view, shown in Figure 3.3.
3.3.2. Control Circuit
Each control constraint, defined either in direct/indirect control mode, is

treated as a closed-loop control problem and solved via a simple PI controller. So it
is guaranteed that each controlled variable is equal to its reference value in steady-
state by choosing appropriate controller parameters. Control circuits for both
direct/indirect control modes are implemented in PSCAD and depicted in Figure 3.4.
The error signal obtained by differencing the actual and the reference value of
the controlled variable drives the controller to produce suitable adjustments of
control inputs of the converter. In order to bring clarity and to make the solution
37
appropriate for power flow studies, the voltage and the phase angle of the respective
converter are used as control inputs in steady-state conditions.
Figure 3.3. PSCAD models of shunt and series converters
38
In direct control mode, reactive and real power injections by the converter m
of the FACTS device are controlled by the voltage magnitude, Vcontrolm and the
phase angle, ph_vscm, respectively as shown in Figure 3.4. After the power flow
problem has reached to a solution, PSCAD variable Vcontrolm (m=1,2,3,4) reaches
to its steady-state values of Vsh1-2, Vse1-2 and PSCAD variable ph_vscm (m=1,2,3,4)
becomes equal to the steady-state values of θsh1-2, θse1-2, respectively.
In indirect control mode, external power system parameters such as line real
and reactive power flows and/or bus voltage magnitudes are regulated at their desired
values by the control inputs of the converters. In both modes, PI controller is also
used as constraint provider, so it holds voltage magnitude and phase angle of the
converter within allowed limits. Operating constraints given in equation (3.1) are
satisfied by this means. All controlled variables are graphically displayed using
multimeter blocks in PSCAD master library.
3.4. Power Flow Studies
3.4.1. Test Systems
The proposed steady-state modeling approach for multi-converter FACTS

devices is tested and verified through case studies in different test systems whose
branch and line data are given in Appendix B. These systems are Western System
Coordinated Council (WSCC) 3-Machine 9-Bus System (Sauer et al., 1997), IEEE
14-Bus System (Washington, 2012), and 3-Machine 7-Bus System (Fardanesh, 2004)
with 100 MVA base each. The components of the test systems are modeled using
standard components of the PSCAD master library. All generators and the condenser
are modeled using three-phase voltage source model-2, (for condenser, voltage
source is phase angle controlled), transmission line is modeled using coupled pi
section transmission line, and the transformer is modeled using three-phase two-
winding transformer. P-Q load which is connected at high voltage bus is modeled as
a custom module using PSCAD master library components in PSCAD, as shown in
Figure 3.5.
39
Figure 3.4. Control constraint implementation in PSCAD
40
Figure 3.5. PSCAD model of the P-Q load connected at high voltage bus
41
Duration of the simulation run is always set to a relatively long time to

observe and ensure that all the variables approach to their steady-state values after
the simulation is completed. The key to the simulation cases is to guarantee that
operating limits are not violated and stable solutions of the control constraints are
always held. Regulator parameters of each mode in each case study are listed in
Appendix C. Total loss of the single-converter FACTS device is assumed to be 0.015
pu on a 100 MVA base (Lee et al., 2003) so that the multi-converter operation yields
an operational loss that is integer multiples of this value. With this respect, UPFC or
IPFC has a total loss of 0.030 pu, and GUPFC has a total loss of 0.045 pu. Solution
speed is kept as fast as possible by setting solution time step relatively long (100 µs)
while keeping the stability of PSCAD.
3.4.2. WSCC 3-Machine 9-Bus System
3.4.2.1. Case 1: STATCOM and SSSC Operations
PSCAD model of WSCC 3-Machine 9-Bus System is shown in Figure 3.6.

Without any compensation, power flow solution of the system is found as:
Receiving-end Line 4-5 flow = 0.7983+j0.1528 pu, receiving-end Line 4-6 flow =
0.5349+j0.0573. At first, non-real power-voltage (P-V) remote/local bus voltages are
regulated to 1.0 pu by Modes 1,3,4 (indirect control mode). STATCOM-1 (Mode 1)
is positioned at Bus 4 (VSC1) and SSSC1-2 (Mode 3,4) are positioned on Lines 4-5
and 4-6, respectively. Simulation results listed in Table 3.3 have proven that
STATCOM-1 and SSSC1-2 are able to regulate bus voltages effectively although the
primary function of SSSC is known as power flow regulation. Bus 9 is regulated by
SSSC-1 at the expense of device rating constraint violation.
42
Figure 3.6. PSCAD model of WSCC 3-Machine 9-Bus System
43
Table 3.3. Power flow results for voltage magnitude regulation @ 1.0 pu
Bus Uncompensated STATCOM-1 SSSC-1 SSSC-2
No, i Vi (pu) Qinj1 (pu) Qinj2 (pu) Qinj3 (pu)
4 0.9930 + 0.1272 + 0.1654 + 0.0762
5 0.9665 + 0.6423 + 0.0168 + 0.7550
6 0.9830 + 0.3359 + 0.4907 + 0.0014
7 1.0010 + 0.0050 - 0.0002 + 0.2782
8 0.9911 + 0.3893 + 0.0230 + 0.0118
9 1.0010 - 0.3434 + 2.2900 + 0.1584
Secondly, real power flow on Line 4-5 is regulated by STATCOM-1 and

SSSC-1 at a sequence of set points, defined in terms of arbitrary chosen
positive/negative percentage changes of the uncompensated real power flow (indirect
control mode). Results are listed in Table 3.4 including resultant VSC output voltage
and injected reactive power to the system. Without device rating constraint violation,
SSSC-1 is able to regulate real power flow at its desired values. However, Line 4-5 is
regulated by STATCOM-1 at the expense of device rating violation. Specifically,
real power flow that cannot be forced to zero by STATCOM-1 is easily regulated to
zero by SSSC-1 which completes the regulation task better than STATCOM-1 with a
slightly smaller device rating.
Table 3.4. Power flow results for real power regulation of Line 4-5
STATCOM-1 SSSC-1 STATCOM-1 SSSC-1
P4-5ref P4-5+jQ4-5 P4-5+jQ4-5 Qinj1, Vsh1 Qinj3, Vse1
(pu) (pu) (pu, kV) (pu, kV)
- 25% 0.5987-j0.2546 0.5987+j0.1795 -3.1370, 0.10 -0.0510, 1.84
- 50% 0.3991-j0.3794 0.3991+j0.1720 -3.1720, 8.17 -0.0734, 3.81
- 75% 0.1995-j0.3955 0.1995+j0.1182 -2.4140, 4.66 -0.0621, 6.07
- 100% 0.0320-j0.1642 0.0000-j0.0050 -0.0030, 0.05 0.0013, 9.05
+ 2% 0.8142+j0.2382 0.8142+j0.1453 0.6433, 21.6 0.0050, 0.14
+ 5% 0.8382+j0.3775 0.8382+j0.1390 1.6410, 22.93 0.0133, 0.35
+ 7% 0.8541+j0.4773 0.8541+j0.1345 2.3520, 23.80 0.0189, 0.49
+ 10% 0.8781+j0.6384 0.8781+j0.1274 3.4920, 25.07 0.0278, 0.70
Thirdly, reactive power flow on Line 4-6 is regulated by STATCOM-1 and

SSSC-2 according to an arbitrary sequence of set points for reactive power flow,
similarly in the above task (indirect control mode). Results are listed in Table 3.5
including resultant VSC output voltage and injected reactive power to the system.
STATCOM-1 and SSSC-2 are able to regulate reactive power flow at zero. Without
device rating constraint violation SSSC-2 is able to regulate reactive power flow at
44
its desired values. However, Line 4-6 is regulated by STATCOM-1 at the expense of
device rating violation.
Table 3.5. Power flow results for reactive power flow regulation of Line 4-6
SSSC-2 STATCOM-1 SSSC-2
STATCOM-1
Q4-6ref P4-6+jQ4-6 Qinj1, Vsh1 Qinj3, Vse2
P4-6+jQ4-6 (pu)
(pu) (pu, kV) (pu, kV)
- 25% 0.5556+j0.0429 0.5458+j0.0429 0.8256, 21.87 0.0058, 1.20
- 50% 0.5527+j0.0286 0.5441+j0.0286 0.7109, 21.71 0.0043, 1.03
-75% 0.5498+j0.0143 0.2640+j0.0143 0.5961, 21.55 0.0305, 2.59
- 100% 0.5468+j0.0000 0.3487+j0.0000 0.4810, 21.38 0.0276, 1.78
+ 25% 0.5613+j0.0716 0.5493+j0.0716 1.0560, 22.18 0.0095, 1.55
+ 50% 0.5641+j0.0859 0.5510+j0.0859 1.1700, 22.33 0.0117, 1.72
+75% 0.5669+j0.1002 0.5528+j0.1002 1.2850, 22.48 0.0142, 1.89
+ 100% 0.5696+j0.1146 0.5547+j0.1146 1.4000, 22.63 0.0168, 2.07
3.4.2.2. Case 2: UPFC Operation
UPFC-2 (Mode 7) is positioned at Bus 4 (VSC1) and on Line 4-6 (VSC3).

Bus 4 voltage, V4 is regulated at 1.0 pu by VSC1 while Pinj1 is regulated at -0.030 pu
to meet the losses of the converters and ensuring real power balance between them
(indirect control mode). At the same time Qinj3 is regulated at values of 0.1 pu, 0.2
pu, and 0.3 pu, respectively by VSC3 (direct control mode). Since UPFC has two
converters, control degree of freedom is three, so that Vsh1, Vse3, and θse3 are the
independent control parameters. However, θsh1 should be regulated to ensure real
power balance among the converters. For each regulated value of Qinj3, θse3 is altered
from 0º to 360º in small degrees to obtain P-Q control planes of Line 4-6 as shown in
Figure 3.7.
3.4.2.3. Case 3: IPFC Operation
IPFC (Mode 8) is positioned on Line 4-5 (VSC2) and on Line 4-6 (VSC3).
Bus 4 voltage, V4 is regulated at 1.0 pu by VSC2 while Pinj2 is regulated at -0.030 pu
to meet the losses of the converters and ensuring real power balance between them
(indirect control mode). At the same time Qinj3 is regulated at values of 0.1 pu, 0.2
pu, and 0.3 pu, respectively by VSC3 (direct control mode). Since IPFC is a two-
45
VSC FACTS device, control degree of freedom is three, so that Vse2, Vse3, and θse3 are
independent control parameters. However, θse2 should be regulated to ensure real
power balance among the converters. For each regulated value of Qinj3, θse3 is altered
from 0º to 360º in small degrees to obtain P-Q control planes of Line 4-6 as shown in
Figure 3.8.
Figure 3.7. P-Q Control planes of Line 4-6 obtained with UPFC
Figure 3.8. P-Q Control planes of Line 4-6 obtained with IPFC
46
3.4.2.4. Case 4: GUPFC Operation
GUPFC (Mode 9) is positioned at Bus 4 (VSC1), on Line 4-5 (VSC2), and on

Line 4-6 (VSC3). Bus 4 voltage, V4 is regulated at 1.0 pu by VSC1 while Pinj1 is
regulated at -0.045 pu to meet the losses of the converters and ensuring real power
balance between the three converters (indirect control mode). At the same time Qinj2
and Qinj3 are regulated concurrently at values of 0.05 pu, 0.1 pu, and 0.12 pu,
respectively (direct control mode). Since GUPFC is a three-VSC FACTS device,
control degree of freedom is five, so that Vsh1, Vse2, Vse3, θse2, and θse3 are independent
control parameters. However, θsh1 should be regulated to ensure real power balance
among the converters. For each set of regulated reactive power injections, θse3 is
altered from 0º to 360º in small degrees while keeping θse2 lags θse3 by 30º arbitrarily
for the sake of simplicity, although VSC2 and VSC3 can operate independently.
Obtained P-Q control planes of Lines 4-5 and 4-6 are presented in Figures 3.9 and
3.10, respectively.
Figure 3.9. P-Q control planes of Line 4-5 obtained with GUPFC
47
Figure 3.10. P-Q control planes of Line 4-6 obtained with GUPFC
3.4.2.5. Discussion of Simulation Results
By examining Tables 3.4 and 3.5, undeterministic real and reactive power
flows are observed in both of the above tasks. This is due to the utilization of single-
converter FACTS devices which are mentioned in Section 2.3. This might necessitate
using multi-converter topologies providing multiple control degrees of freedom, if
independent real and reactive power flow regulation is required on a specific line. It
is generally concluded that STATCOM is practical for voltage regulation and SSSC
exhibits superior real and reactive power flow regulation performance than
STATCOM with smaller device rating of SSSC. It is also concluded from the
obtained P-Q circles that UPFC, IPFC, or GUPFC is able to increase/decrease real
and reactive power flows as well as reverse the direction of flow. Zero reactive
power flow can also be achieved to decrease transmission losses. The higher reactive
compensation level which means higher reactive power injection, the larger P-Q
control area is attained by the FACTS device. These results have been verified from
literature (Gyugyi et al., 1995), (Gyugyi et al., 1999). Maximum attainable reactive
compensation level for each converter is always observed under 1.0 pu because of
non-zero Ptransfer and converter losses.
48
3.4.3. IEEE 14-Bus System
3.4.3.1. Case 1: UPFC Operation
PSCAD model of IEEE 14-Bus System is shown in Figure 3.11. Power flow
control capabilities of UPFC-1 and UPFC-2 are investigated in this case study.
Uncompensated parameters of the test system are as follows: V5 = 1.020 pu, P52
+jQ52 = - 0.4074 + j0.1667 pu, and P54 + jQ54 = 0.6293 + j0.0276 pu.
First, UPFC-1 (Mode 6) is positioned at Bus 5 (VSC1) and on Line 5-2
(VSC2) in indirect control mode. Bus 5 voltage, V5 is regulated at different set points
by VSC1 while Pinj1 is regulated at -0.030 pu to meet the losses of the converters and
ensuring real power balance between them. For the given set of reference values,
power flow solution is obtained with the internal parameters of UPFC-1. The results
are listed in Table 3.6.
Secondly, UPFC-2 (Mode 7) is positioned at Bus 5 (VSC1) and on Line 5-4
(VSC3) in indirect control mode. Similarly V5 is regulated at different set points by
VSC1 while Pinj1 is regulated at -0.030 pu. For the given set of reference values,
power flow solution is obtained with the internal parameters of UPFC-2. The results
are listed in Table 3.6.
Table 3.6. Parameters of the UPFC under different power flow control strategies
UPFC-1
scheduled system variables VSC1 output voltage VSC2 output voltage
P52ref + jQ52ref V5ref magnitude (kV) phase angle (º) magnitude (kV) phase angle (º)
0.15+j0.40 1.00 12.21 -41.46 2.35 4.05
-0.20+j0.05 1.01 12.13 -39.78 0.81 17.62
-0.35+j0.15 1.01 12.16 -38.80 0.57 -35.74
-0.50-j0.15 0.95 10.39 -37.16 0.79 -82.73
UPFC-2
P54ref + jQ54ref V5ref magnitude (kV) phase angle (º) magnitude (kV) phase angle (º)
0.50+j0.02 1.00 12.01 -38.34 0.73 -71.10
0.80+j0.05 1.05 12.10 -39.89 0.99 -8.09
-0.20+j0.01 1.00 11.87 -35.07 2.50 -122.74
0.15-j0.025 1.015 12.25 -36.75 1.42 -122.49
49
Figure 3.11. PSCAD model of IEEE 14-Bus System
50
3.4.3.2. Case 2: IPFC Operation
IPFC (Mode 8) is positioned on Line 5-2 (VSC2) and on Line 5-4 (VSC3) to
control reactive power flow of Line 5-2 and real and reactive power flows of Line 5-
4 at their desired values in indirect control mode. Pinj2 is regulated at -0.030 pu to
meet the losses of the converters and ensuring real power balance between them. For
the given reference values of the real and reactive power flows, power flow solutions
including internal parameters of IPFC are listed in Table 3.7.
Table 3.7. Parameters of the IPFC under different power flow control strategies
IPFC
P54ref + jQ54ref Q52ref magnitude (kV) phase angle (º) magnitude (kV) phase angle (º)
0.50+j0.10 j0.01 3.66 -141.41 1.16 29.68
0.65+j0.008 j0.08 0.68 -109.69 0.35 68.06
-0.20+j0.005 -j0.08 1.83 -112.24 0.48 100.03
-0.40-j0.10 j0.015 2.12 -127.56 2.31 199.48
3.4.3.3. Case 3: GUPFC Operation
GUPFC (Mode 9) is positioned at Bus 5 (VSC1), on Line 5-2 (VSC2) and on

Line 5-4 (VSC3) to control real and reactive power flows of Line 5-2 and real and
reactive power flows of Line 5-4 simultaneously in indirect control mode. At the
same time V5 is regulated at various set points independently from the controlled
flows and Pinj1 is regulated at -0.045 pu to meet the losses of the three converters and
ensuring real power balance between them. Table 3.8 indicates the results of power
flow with the internal parameters of the GUPFC for the given set of reference values.
The simulation results prove that the proposed FACTS device model in
various operating modes is capable of handling the scheduled real and reactive power
flows and bus voltage in indirect control mode. Operational and control constraints
are satisfied with a stable solution and an acceptable simulation time. It is shown that
51
well-known control functions, such as real and reactive power flow increase/decrease
as well as reversing real power flow are all implemented in this case study. Multi-
control objectives are met with the multi-converter FACTS devices in steady-state.
Table 3.8. Parameters of the GUPFC under different power flow control strategies
GUPFC
scheduled
system VSC1 output voltage VSC2 output voltage VSC3 output voltage
variables
P52ref + jQ52ref
magnitude phase angle magnitude phase angle magnitude phase angle
P54ref + jQ54ref
(kV) (º) (kV) (º) (kV) (º)
V5ref
-0.20+j0.10
0.50+j0.01 13.08 -40.10 0.74 42.53 0.59 67.02
1.04
0.40+j0.01
0.40-j0.01 12.80 -45.72 3.59 38.70 0.93 37.71
1.03
0.30-j0.01
0.65-j0.01 12.80 -47.02 3.67 40.23 1.71 35.26
1.03
-0.035-j0.20
0.08+j0.025 13.20 -38.63 1.22 89.46 1.29 207.44
1.065
3.4.4. 3-Machine 7-Bus System
Power flow studies of BtB-STATCOM in 3-Machine 7-Bus System

(including generator buses), shown in Figure 3.12, are conducted. BtB-STATCOM
installation is implemented at two neighboring buses, namely, Bus 1 (VSC1) and Bus
3 (VSC4) which is different from a BtB DC link that connects two remote-end buses
via long HVDC transmission.
3.4.4.1. Case 1: Reactive Power-Voltage (Q-V) Characteristics
Q-V characteristics of the two converters of BtB-STATCOM as two

independent STATCOMs are investigated by decoupling the DC links of the
converters. Qinjref is changed in steps for each VSC independently. θsh1 and θsh2 are
regulated individually to meet losses (0.015 pu) of STATCOM-1 and STATCOM-2,
52
Figure 3.12. PSCAD model of 3-Machine 7-Bus System
53
respectively with zero real power transfer between them. Qinjref max is
calculated according to operating constraints in Table 3.2 for Modes 1 and 2.
Numerical results are illustrated in Table 3.9.
Table 3.9. Q-V characteristics of the two converters

VSC1 Qinj1 VSC4 Qinj4
V1 (pu) V3 (pu)
(kV, deg) (pu) (kV, deg) (pu)
14.06, -12.41° 0.999 0.9955 14.10, -25.71° 0.999 0.9990
13.94, -12.49° 0.8 0.9896 13.99, -25.89° 0.8 0.9931
Capacitive
13.82, -12.56° 0.6 0.9836 13.87, -26.07° 0.6 0.9871
Compensation
13.70, -12.64° 0.4 0.9774 13.75, -26.25° 0.4 0.9810
13.58, -12.73° 0.2 0.9712 13.63, -26.45° 0.2 0.9748
No FACTS 0.0, 0.0° 0.0 0.9652 0.0, 0.0° 0.0 0.9688
13.33, -12.90° -0.2 0.9584 13.38, -26.86° -0.2 0.9620
13.20, -12.99° -0.4 0.9517 13.25, -27.08° -0.4 0.9554
Inductive
13.07, -13.09° -0.6 0.9450 13.12, -27.31° -0.6 0.9487
Compensation
12.94, -13.19° -0.8 0.9380 12.99, -27.55° -0.8 0.9419
12.80, -13.30° -0.999 0.9309 12.85, -27.80° -0.999 0.9349
3.4.4.2. Case 2: Real Power-Voltage (P-V) Characteristics
BtB-STATCOM and STATCOM are compared when improving P-V curves

of Buses 1 and 3 under steady-state conditions when demanded real power (load) at
these buses is increased in steps. At first, STATCOM-1 and STATCOM-2 are
commanded simultaneously to inject pre-defined Qinjref at Buses 1 and 3,
respectively. Next, BtB-STATCOM is operated to hold voltage levels against to
increasing demanded power when max. possible real power transfer occurs between
the two VSCs. The derivation of Pinjref max and Qinjref max for these operations are
derived in Appendix D. Comparative P-V curves of Buses 1 and 3 for two different
directions of real power transfer are shown in Figures 3.13 and 3.14, respectively.
54
(a) Bus 1 voltage profile when real power transfer is from VSC1 to VSC4
(b) Bus 1 voltage profile when real power transfer is from VSC4 to VSC1
Figure 3.13. Comparative P-V curves of Bus 1
55
(a) Bus 3 voltage profile when real power transfer is from VSC1 to VSC4
(b) Bus 3 voltage profile when real power transfer is from VSC4 to VSC1
Figure 3.14. Comparative P-V curves of Bus 3
Results show that either STATCOM or BtB-STATCOM can control bus

voltages in steady-state and improve voltage stability of the power system by
improving P-V curves of the local buses. Coupling the DC links of the two
STATCOMs yields real power transfer from one neighboring bus into another in
BtB-STATCOM operation. Real power transfer, which brings an extra control
56
degree of freedom in power flow studies, changes real and reactive power flow
distributions on transmission lines, hence changes voltage profiles indirectly. This
situation should be considered for BtB-STATCOM design in practical applications.
3.5. Summary
A new modeling approach for power flow studies of the power systems
embedded with single- and multi-converter FACTS devices is presented in PSCAD
which is based on the regulation of magnitude and phase angle of the converters in
steady-state conditions. Operational and control constraints defined for each FACTS
device are solved in PSCAD using simple PI regulators. Direct and indirect control
modes for each FACTS device are tested and verified with various case studies in
different test systems. Graphical interface of PSCAD removes programming burden
such as coding or Jacobian matrix modification of NR method due to contributions of
shunt/series converters in terms of power injections. Also it contributes to a clear,
flexible, and understandable modeling approach but at the expense of PI regulator
tuning for each mode. The model is expandable so that the number of converters can
be increased with simple modifications to the constraints. Converter losses can be
explicitly defined and modeled. The proposed approach is also beneficial for large
scale systems if sufficient computing power and large memory are available.
57
58
4. VOLTAGE SOURCE CONVERTER DESIGN A. Mete VURAL
4. VOLTAGE SOURCE CONVERTER DESIGN
4.1. Introduction
High power voltage source converter (VSC), developed from the applications
of low and medium power levels in industrial applications, is the building block of
the second generation FACTS devices having single or multiple converter
arrangements (Kazerani et al., 2002), (Tan et al., 2006). VSC can also be regarded as
a self-commutating converter, built from power semiconductors having turn-off
capabilities, such as GTO, GCT, or IGCT, and has the capability of both consuming
and generating reactive power which provides independent and simultaneous control
of real and reactive power flows. This property makes VSC superior when compared
with the line commutating converter which is only able to consume reactive power
from the power system and suffers from commutation failures of conventional
thyristors having only turn-on capabilities. VSC based topologies are generally
preferred over current source converters for FACTS applications at transmission
level due to higher losses and more complicated control (Hingorani et al., 2000),
(Kazerani et al., 2002), (Bahrman et al., 2003).
Converter-level modeling of the multi-converter FACTS devices requires
realistic high power VSC design for dynamic performance analysis and transient
stability studies if realistic time domain simulated responses are required to be
observed.
In converter design, the objective is to minimize switching frequency of the
power semiconductors hence minimize losses and to produce high quality quasi-
sinusoidal voltage waveform at transmission level with minimum or no filtering
requirements. Multi-pulse converter topology can be preferred over multi-level one
when back-to-back operation of two or more VSCs fed from a common DC link is
considered. Since DC link voltage control is easy due to a single DC voltage level as
opposed to multi-level structures for back-to-back VSCs in multi-converter FACTS
device applications (Soto et al., 2002), (Lee et al., 2003). On the other hand, quasi
multi-pulse topology can be preferred over true multi-pulse one due to: i) simple
59
design with ordinary transformers without employing phase-shifting zig-zag

transformers which can bring extra cost when practical aspects are considered (Lee et
al., 2003), ii) total harmonic distortion (THD) similar to that of true multi-pulse one
(Cavaliere et al., 2002), (Lee et al., 2003).
The objective of this chapter is to give design details of high power quasi
multi-pulse VSC design for converter-level models of GUPFC, IPFC, and BtB-
STATCOM suitable for dynamic performance analysis and transient stability studies.
First, basic circuits and operating principles of six-pulse and twelve-pulse converters,
which are the basic building blocks of the proposed quasi multi-pulse VSC are
discussed and summarized. Next, power circuit design and control scheme of the
quasi multi-pulse VSC are given with the details including pulse-generating circuit
for the power semiconductors. PSCAD is used to simulate the output voltage
waveforms of the converters for evaluation together with their harmonic content.
4.2. Six-pulse VSC
4.2.1. Circuit Configuration
Three-phase two-level six-pulse VSC consists of six valves as shown in

Figure 4.1. There are three arms having two valves each. Each valve is comprised of
turn-off capable power semiconductor, such as GTO, and a reverse-parallel diode.
For self-commutating converter operation, each GTO should be turn on and off at
controlled time instants to shape AC output voltage from a fixed DC voltage,
maintained by a capacitor which is large enough to retain changes in DC current
without changes in DC voltage. Diode is used to provide a path for inductive current
whenever the GTO in the same valve is turned off. In VSC, DC voltage always has
one polarity while the power reversal is achieved by reversal of DC current.
Therefore, power semiconductor only requires unidirectional voltage blocking
capability.
Under constant DC voltage, three-phase output of the six-pulse VSC can be
controlled both in magnitude and phase angle, normally accomplished by pulse width
60
modulation (PWM) control. There are different PWM methods in literature to reduce
harmonic content as much as possible to make AC output voltage resembling a pure
sinusoid. Unfortunately high frequency PWM control is considered uneconomical for
high power applications due to high switching losses, thus resulting both in
decreased conversion efficiency and in bulk cooling equipment.
Figure 4.1. Power circuit of three-phase six-pulse VSC
4.2.2. Working Principle
To investigate the interaction of six-pulse VSC with the power system,

switches S1 and S3 are turned off while S2 is turned on in Figure 4.1. In this case,
VSC is able to exchange real and/or reactive power with the three-phase system
through interface reactor L according to the four quadrant operation which is
presented in Figure 4.2 (Sood, 2004). Often, leakage impedance of the coupling
magnetic interface (shunt for STATCOM, series for SSSC) can serve as the inductive
impedance with or without L, which electrically separates sinusoidal three-phase
system voltages and three-phase voltages of the six-pulse VSC containing harmonics.
DC side voltage is kept constant by rectifier and inverter operations which are
61
illustrated in Figures 4.2a and 4.2b, respectively. In rectifier operation capacitor is

charged by taking real power from the three-phase system. In inverter operation
capacitor is discharged by injecting real power into the three-phase system. Reactive
power compensation is accomplished by making converter current Iph purely
reactive. If Iph leads three-phase system voltage, eAN by 90°, as shown in Figure 4.2c,
reactive power is injected into the three-phase system. Alternatively, if Iph lags eAN
by 90°, as shown in Figure 4.2d, reactive power is taken from the three-phase
system. Vector control of the VSC provides required magnitude and the phase angle
of the VSC output voltage in either one of or the combinations of the operations
available in four quadrant operation. VSC output voltage is synchronized with the
voltage of the three-phase system by a phase lock loop (PLL) to match required
phase angle shift. The operation principle and the interaction with the power system
is the same for more complicated VSC configurations regardless of the converter
topology to meet high voltage/current ratings and to reduce harmonic content which
is acceptable for FACTS applications at transmission level.
Figure 4.2. Four quadrant VSC operation
62
4.2.3. Analysis of Six-pulse VSC
For the sake of clarity and to highlight basic operation, six-pulse VSC is
isolated from the three-phase system by turning on switches S1 and S3, and turning
off switch S2, shown in Figure 4.1. An external DC voltage source is connected at
the DC terminals and a three-phase resistive load RLD is connected at the AC
terminals. The circuit is simulated in PSCAD for 4 cycles with Vdc is set to 3.0 kV,
C=2000 µF, RLD and Rs is set to 1.0 MΩ and 0.001 Ω, respectively. GTO and diode
have turn on/off resistances of 0.005 Ω and 1.0E8 Ω, respectively. Snubber circuit
elements are ignored. The simulated phase-to-neutral and phase-to-phase voltage
waveforms of six-pulse VSC are shown in Figures 4.3 and 4.4, respectively. Each
phase is shifted by ± 120° with respect to other phases for balanced three-phase
operation. A simple square-wave switching scheme is applied with a switching
frequency of 50 Hz so that each GTO conducts only for 180° or 10 ms duration, as
shown in Figure 4.5. This type of switching is called 180-degree conduction in which
only three GTOs remain on at any time instant.
Figure 4.3. Simulated phase-to-neutral voltage waveforms of six-pulse VSC
Fourier series expansion of the phase-to-phase voltage waveform, VAB is

given by the equation (4.1) to have an idea about the harmonic content, generated by
63
the six- pulse operation in an analytical approach (Dávalos et al., 2005). The voltages
VBC and VCA exhibit similar patterns except phase shifts of -120° and +120°,
respectively.
∞
 nπ 
V AB (t ) = ∑ v n sin  nwt +  (4.1)
n =1  6 
The peak value of nth voltage harmonic component is given in equation (4.2). Noting
that n=6r±1, (r=0,1,2,…). Even and triplen harmonics are zero.
4 nπ
vn = Vdc cos (4.2)
nπ 6
Figure 4.4. Simulated phase-to-phase voltage waveforms of six-pulse VSC
Figure 4.6 illustrates dominant harmonics in the harmonic spectrum of phase-

to-phase voltage VAB by applying fast Fourier transform in PSCAD where the voltage
waveform is broken down into a spectrum of sinusoidal frequencies up to the 61st
harmonic. The dominant harmonics have an index of m=6r±1 for six-pulse VSC
where r is positive integer (m=5,7,11,13,17,19,…). These significant harmonic
components are illustrated in percentage of the fundamental component.
64
Figure 4.5. Gating signals of GTOs for 180-degrees conduction
Figure 4.6. Harmonic spectrum of VAB for six-pulse VSC
4.3. Twelve-pulse VSC
In six-pulse VSC operation, THD is relatively high as a consequence of

availability of low order harmonic components (5,7,11,13,…), which are not suitable
for high power FACTS applications. One way to decrease the level of harmonics is
65
to increase the number of six-pulse units. Twelve-pulse operation can be achieved by

summing phase-to-phase voltage of one converter (upper one) with the phase-to-
neutral voltage of the other converter (lower one) by means of a magnetic interface
as shown in Figure 4.7.
Figure 4.7. Power circuit of three-phase twelve-pulse VSC
To correct inherent phase shift between the phase-to-phase and phase-to-

neutral voltages, the gating patterns of wye-connected converter should require a
phase shift of +30°, or the gating patterns of delta-connected converter should
require a phase shift of -30°. Magnetic interface is generally designed either using
two three-phase-two-winding transformer banks having a total of twelve windings or
six single-phase transformers having a total of twelve windings. However in the
proposed design which is shown in Figure 4.7, three single-phase-three-winding
transformers are used, having a total of nine windings only. Since phase-to-phase
66
voltage is √3 times the phase-to-neutral voltage, turns ratio of delta and wye
windings are selected such that the ratio of the secondary side of delta-winding to
that of wye-winding becomes √3.
4.3.2. Analysis of Twelve-pulse VSC
Phase-to-phase voltage waveforms in Figure 4.8 are obtained by simulating

the circuit in Figure 4.1 by replacing six-pulse VSC with twelve-pulse one. Three
25/3 MVA rated (5.0kV/3.0kV/1.732kV) transformers with leakage reactance of j0.1
pu are used. The resultant voltage signal resembles more closely to a perfect
sinusoidal waveform than the one obtained from six-pulse operation.
Figure 4.8. Simulated phase-to-phase voltage waveforms of twelve-pulse VSC
Fourier series expansion of the twelve-pulse waveform of phase-to-phase

voltage VAB is given by the equation (4.3) (Dávalos et al., 2005). The voltages VBC
and VCA exhibit similar patterns except phase shifts of -120° and +120°, respectively.
∞
 nπ 
V AB (t ) = ∑ v n sin  nwt +  (4.3)
n =1  6 
67
The peak value of nth voltage harmonic component is given in equation (4.4).
Noting that n=12r±1, (r=0,1,2,…).
8 nπ
vn = Vdc cos (4.4)
nπ 6
Figure 4.9 illustrates dominant harmonics in the harmonic spectrum of phase-

to-phase voltage VAB. The dominant harmonics have an index of m=12r±1 for
twelve-pulse VSC where r is positive integer (m=11,13,23,25,35,37…). Although
THD is decreased and twelve-pulse arrangement effectively reduces 5th and 7th
harmonics, 11th and 13th harmonics still exist which require further process.
Figure 4.9. Harmonic spectrum of VAB for twelve-pulse VSC
4.4. Quasi Multi-pulse VSC
The proposed quasi multi-pulse VSC is designed using four phase-shifted

twelve-pulse converter units (1,2,3,4) that are connected in series in line side and
connected in parallel in the DC link side, as shown in Figure 4.10. Based on required
68
blocking voltage and peak current ratings, parallel and/or series combinations of
GTOs can also be employed for reasons of economy and easy availability of switches
with lower ratings (Sun et al., 2004). In this design, only one GTO with a reverse-
parallel diode per valve is utilized as the switching device to increase simulation
speed.
For the purpose of VSC control which will be discussed later, the overall
circuit is decomposed into two main parts, namely converters M and N, respectively.
Phase-A of twelve-pulse converter unit 1 is coupled to the phase-A of twelve-pulse
converter unit 2 with a single-phase transformer A1. Similarly, phase-B and phase-C
are coupled using transformers B1 and C1, respectively to make a quasi 24-pulse
converter (converter M). The second quasi 24-pulse converter (Converter N) is built
up using the other twelve-pulse converter units (3-4) and single-phase transformers
A2, B2, and C2. Phase-A of converter M and that of converter N are electro-
magnetically added using transformers A1 and A2, since the primaries are connected
in series. In a similar fashion, transformers B1 and B2 are used to sum phase-B of
converter M with that of converter N. Transformers C1 and C2 are used to sum
phase-C of converter M with that of converter N. Summing and interfacing
magnetics also couples VSC output voltage with the transmission level with no
requirement to an extra shunt coupling transformer by adjusting the voltage ratings
of primaries of A1, A2, B1, B2, C1, C2.
Phase shift angle between two adjacent twelve-pulse converters should be
7.5° (Singh et al., 2009). So, 7.5º, 0.0º, -7.5º, and -15º phase shifts are applied to the
gating signals of each upper six-pulse converter of twelve-pulse unit 1,3,2,4,
correspondingly. This arrangement also satisfies that twelve pulse units 1 and 3 and
units 2 and 4 can operate as two independent quasi 24-pulse converters, respectively.
On the other hand, gating signals of each lower six-pulse converter of four twelve-
pulse units are shifted by 30° one by one with respect to each upper side VSC for
proper twelve-pulse operation. Figure 4.11 shows only ¼ of the PSCAD
implementation of the quasi multi-pulse VSC, which is designed using PSCAD
master library components. Figure 4.12 shows PSCAD implementation of two
different magnetic interfaces required for quasi multi-pulse operation.
69
Figure 4.10. Power circuit configuration of three-phase quasi multi-pulse VSC
70
Figure 4.11. PSCAD implementation of ¼ of quasi multi-pulse VSC
71
More specifically, Figure 4.12a illustrates the details of the magnetic interface
for twelve-pulse operation, which is modeled as a PSCAD default module. Summing
and magnetic interface for quasi multi-pulse VSC is presented in Figure 4.12b, which
is modeled directly on the main project page of PSCAD.
Figure 4.12. PSCAD implementation of magnetic interfaces
4.4.2. Series Coupling Magnetic Interface
Power circuit configuration in Figure 4.10 can be used as shunt connection by

adjusting the parameters of the summing and magnetic interface, shown in Figure
4.12b. For GUPFC and IPFC, series connection is also required. In this context, a
72
series coupling magnetic interface shown in Figure 4.13 can be designed to directly
inject three-phase AC voltages of the series VSCs to three-phase transmission line.
Figure 4.13. PSCAD implementation of series coupling magnetic interface
4.4.3. Control Scheme for Quasi Multi-pulse VSC
Each quasi multi-pulse VSC of the multi-converter FACTS device should be

controlled both in magnitude and phase angle to meet the required control objectives.
As mentioned before, high frequency PWM methods which can easily control both
converter voltage and phase angle are not useful for high power applications due to
high switching losses. This type of work is rich in literature in which approximated
or simple converter models (six-pulse operation) are employed, not realistic for high
power applications. Line frequency switching at 50 Hz or 60 Hz can be preferred
alternatively, where each GTO is switched only once per cycle. This brings a
difficulty due to the fact that the rms value of the fundamental component of the
converter voltage becomes only a function of DC link voltage, which is strictly
regulated to a constant value for proper VSC operation. So the only control
73
parameter for six-pulse operation and hence quasi multi-pulse VSC, regardless of the
FACTS device and the multi-pulse configuration is the phase shift applied to the gate
pulse pattern of the GTOs. In order to bring an extra control degree, the “2-angle
control” method is adopted from literature (Hagiwara et al., 2003). In this approach,
quasi multi-pulse VSC can be controlled both in magnitude and phase angle by
appropriate two kinds of phase shifts even GTOs are switched at line frequency. The
calculation procedure of these two shift angles are given in the next section.
4.4.3.1. 2-angle Control Method
2-angle control method can be described by the phasor diagram shown in

Figure 4.14 (Hagiwara et al., 2003). VM and VN respectively denote output voltage
vectors of quasi 24-pulse converters (converters M and N in Figure 4.10), described
by the equation (4.5). VX is the resultant summation vector of the quasi multi-pulse
VSC comprised of leading converter M and lagging converter N, respectively.
Subscript X describes the relevant VSC in the multi-converter FACTS device
(sh1,sh2,se1,se2 in Figure 3.2). VS is the reference voltage vector from a selected bus
in the power system whose phase angle is measured by a PLL for axis
synchronization. Since AC output voltage magnitude of each quasi 24-pulse VSC is
equal and constant, the voltage vector VX can only be controlled both in magnitude
and phase angle by choosing appropriate phase angles (ΦM, ΦN) using equation (4.6).
Figure 4.14. Voltage vectors of converters M and N in rotating reference frame
74
r
 VM = VM ∠φ M = V M ∠(α − δ )° 
r  (4.5)
V N = V N ∠φ N = V N ∠ − (α + δ )°
(2VF )2 − (VD ref ) ( )

2 2
−1
− VQ ref
α = tan
(V ) + (V )
,
ref 2 ref 2
D Q
−1
VQ ref
δ = tan (4.6)
VD ref
VF = 0.5(VM+VN) is the measured average converter voltage to minimize

measurement variations, VDref and VQref are required d-axis (direct axis) and q-axis
(quadrature axis) components of VX, computed from control loops of the multi-
converter FACTS device, which will be discussed in Chapters 5 and 6.
Synchronization with the d-axis is ensured with the following set of relations where θS
is the phase angle of the selected bus, measured using PLL. Equations (4.6) and (4.7)
are implemented to compute the required phase angles (ΦM, ΦN) for converters M and
N in real-time using blocks of continuous system model functions in PSCAD master
library, as represented in Figure 4.15.
phM = θ S + (α − δ )
phN = θ S − (α + δ ) (4.7)
Figure 4.15. PSCAD implementation of equations (4.5) and (4.6)
75
4.4.3.2. Pulse-generating Circuit
Pulse-generating circuit is used to generate square type waveforms for GTO

switching. In PSCAD, logic level one is used to turn on GTO, while logic level zero
is used to turn off GTO. PSCAD implementation of the pulse generating scheme for
six-pulse VSC is shown in Figure 4.16.
Figure 4.16. PSCAD implementation of switching logic for six-pulse VSC
In the pulse-generating scheme, 50 Hz or 60 Hz sinusoidal signal with

amplitude one is compared with zero. Hence, for the first half-cycle of fundamental
frequency the comparator output becomes logic one, and for the second half-cycle
the comparator output becomes logic zero. The produced square-wave is phase shift
controlled since the phase shift of sinusoidal signal can be externally controlled. One
comparator should produce two complementary pulses firing upper and lower GTO
of a given arm to prevent short circuit of DC link. In total, three comparators that are
±120° out of phase with each other are needed to produce six pulses for the three-arm
76
for balanced three-phase operation. The required number of pulse-generating circuits

for each multi-converter FACTS device aimed in this research is listed in Table 4.1.
Table 4.1. Number of pulse-generating circuits per multi-converter FACTS device

Number of
FACTS device Number of VSC Number of GTO pulse-generating
circuit
IPFC 2 2x6x8=96 2x8=16
BtB-STATCOM 2 2x6x8=96 2x8=16
GUPFC 3 3x6x8=144 3x8=24
4.4.4. Analysis of Quasi Multi-pulse VSC
4.4.4.1. Quasi 48-pulse Operation
The circuit shown in Figure 4.1 is simulated for a fundamental frequency of

60 Hz by replacing six-pulse VSC with the proposed quasi multi-pulse VSC with the
same simulation parameters for six-pulse and twelve-pulse topologies. Summing and
magnetic interface in Figure 4.12b is designed using six 16.67 MVA
(10.0kV/10.0kV) rated single-phase transformers with a leakage reactance of j0.1 pu
each. The primary side is rated at 10 kV for convenience, although is set to higher
values generally to couple with high voltage transmission level in FACTS
applications. Quasi 48-pulse voltage waveforms are simulated by setting ΦM and ΦN
both to zero. Phase-to-neutral and phase-to-phase voltages are presented in Figure
4.17. The resultant voltage signals resembles more closely to a perfect sinusoidal
waveform than previously mentioned converter topologies.
Figure 4.18 plots harmonic spectrum of phase-to-phase phase-A voltage,
plotted in Figure 4.17. Although, 48-pulse operation does not cancel all the
harmonics of orders 6n±1 (n=1,2,3,…), their amplitudes are strongly decreased. For
example, 11th and 13th order harmonics are effectively reduced when compared with
either six-pulse or twelve-pulse topology. As a result, THD is effectively reduced to
3.76 %. Except 11th and 13th order harmonics, dominant harmonics have an index of
77
48r±1 (r=1,2,3,…) almost obtained by true 48-pulse configuration (Lee et al., 2003),
(Geethalakshmi et al., 2007).
Figure 4.17. Simulated voltage waveforms of quasi 48-pulse VSC
Figure 4.18. Harmonic spectrum of VAB for quasi 48-pulse operation
Fourier series expansion of true 48-pulse waveform of phase-to-phase voltage

VAB can be given by the equation (4.8) for the sake of clarity (Geethalakshmi et al.,
2007). The voltages VBC and VCA exhibit similar patterns except phase shifts of -120°
and +120°, respectively.
∞
V AB (t ) = ∑ v n sin (nwt + 18.75°n + 11.25°i ) (4.8)
n =1
78
The peak value of nth voltage harmonic component is given in equation (4.9). Noting
that n=48r±1, (r=0,1,2,…) and i= -1 for n=47,95,… and i=1 for n=49,97,…,
respectively.
32 nπ
vn = Vdc cos (4.9)
nπ 6
Although theoretically possible, the converter topology can become more and
more complicated if the pulse numbers above 48 are applied that can be rarely
justified in practical applications. It is shown that a quasi 48-pulse VSC is sufficient
for 100 MVA STATCOM application (Schauder et al., 1995) and a true 48-pulse
topology is designed for 80 MVA SVC (Mori et al., 1993).
The proposed quasi multi-pulse VSC is designed such that the AC outputs of
two quasi 24-pulse converters are magnetically added, each of which is
independently and externally phase shift controlled. On the other hand, four twelve-
pulse converter units are designed together with appropriate phase shifts to operate as
quasi 48-pulse converter if external phase shifts are set to zero. In this sense, at the
best case, the proposed VSC can show the harmonic performance of quasi 48-pulse
topology and thereby named as “quasi multi-pulse”. In Chapters 5 and 6, the actual
THD content of the proposed quasi multi-pulse VSC will be evaluated and compared
with the IEEE standards when its shunt/series combinations are utilized in GUPFC,
IPFC, and BtB-STATCOM configurations.
4.4.4.2. Verification of 2-angle Control Method
2-angle control method is verified by open-loop control of the quasi multi-

pulse VSC when the phase shifts of quasi-24 pulse converters (ΦM, ΦN) are
externally commanded to show how the magnitude and phase angle of VX in Figure
4.14 can be freely controlled. The converter is isolated from the power system as the
previous analysis cases and it is connected with a wye-connected resistive load in its
AC side and DC side is fed from a constant DC voltage source. The closed-loop
79
control requires the calculation of VDref and VQref from external control loops for each
VSC of the multi-converter FACTS device. This situation is thoroughly studied and
discussed in detail in Chapters 5 and 6. In the verification procedure, at first the
output of the quasi multi-pulse VSC is forced to align on four different quadrants on
the graph of polar plot, as shown in Figure 4.19. This case study shows how the
phase angle of VX can be controlled in the range between 0° and 360°. Secondly, it is
demonstrated how both the magnitude and phase angle of the output of quasi multi-
pulse VSC can be freely controlled using randomly selected simulation results under
the condition that the output of each quasi 24-pulse converter is fixed and equal to
7.75 kV. The cases are shown in Figure 4.20.
80
Voltage VM Voltage VN Voltage VX
ΦM (ref)=0.0°
ΦN (ref)=0.0°
D D D
7.75677 -0.1203 7.75677 -0.1203 15.5135 -0.1203
(kV) (°) (kV) (°) (kV) (°)
VX in
first quadrant
ΦM (ref)=90.0°
ΦN (ref)=0.0°
D D D
7.74924 90.04 7.7533 0.01782 10.9686 45.03
(kV) (°) (kV) (°) (kV) (°)
VX in
second quadrant
ΦM (ref)=180.0°
ΦN (ref)=90.0°
D D D
7.75677 179.9 7.74439 90.18 10.9589 135.1
(kV) (°) (kV) (°) (kV) (°)
VX in
third quadrant
ΦM (ref)=270.0°
ΦN (ref)=180.0°
D D D
7.74925 -89.97 7.7533 -180 10.9686 -135.2
(kV) (°) (kV) (°) (kV) (°)
VX in
fourth quadrant
ΦM (ref)=0.0°
ΦN (ref)=270.0°
D D D
7.7533 0.0736 7.74925 -90.12 10.9553 -44.95
(kV) (°) (kV) (°) (kV) (°)
Figure 4.19. Four quadrant operation of the proposed quasi multi-pulse VSC
81
ΦM (ref)=30.0°
ΦN (ref)=60.0°
D D D
7.75497 30.02 7.74879 59.89 14.9759 44.94
(kV) (°) (kV) (°) (kV) (°)
ΦM (ref)=174.0°
ΦN (ref)=320.0°
D D D
7.74935 174 7.75164 -40.17 4.61606 -113.9
(kV) (°) (kV) (°) (kV) (°)
ΦM (ref)=190.0°
ΦN (ref)=350.0°
D D D
7.75272 -169.9 7.75164 -10.12 2.86963 -91.23
(kV) (°) (kV) (°) (kV) (°)
ΦM (ref)=250.0°
ΦN (ref)=12.0°
D D D
7.75174 -110 7.75243 12.02 7.54778 -49.05
(kV) (°) (kV) (°) (kV) (°)
ΦM (ref)=180.0°
ΦN (ref)=0.0°
D D D
7.75331 -180 7.75331 0.01782 8.75729e-... 0
(kV) (°) (kV) (°) (kV) (°)
Figure 4.20. Flexible magnitude/phase angle controlled quasi multi-pulse VSC
82
4.5. Summary
A realistic and high power quasi multi-pulse VSC for multi-converter FACTS
devices is proposed and designed with given all details down to the GTO level,
including magnetic interface and control scheme. First, working principles of
elementary two-level six-pulse and twelve-pulse converter topologies are discussed.
Later on, quasi multi-pulse VSC is designed using four twelve-pulse converter units
which is more accurate than existing low-order or average models. Line frequency
switching scheme is applied to minimize converter losses and voltage/current
stresses on each GTO valve are fairly decreased using multi-converter structure. 2-
angle control method is adapted from literature to gain an extra control degree to the
proposed VSC without changing the magnitude of the DC link voltage. Harmonic
content is quantitatively evaluated in terms of individual harmonic voltages and
THD. Power and voltage rating are flexible so that quasi multi-pulse VSC can be
simulated with regard to different operating requirements. For the next two chapters,
the designed quasi multi-pulse converter topology will be used as the building
element for GUPFC, IPFC, and BtB-STATCOM where the test and the verification
of the model will be done digitally in PSCAD in one sense.
83
84
5. DYNAMIC MODELING STUDIES A. Mete VURAL
5. DYNAMIC MODELING STUDIES
5.1. Introduction
Dynamic modeling of the multi-converter FACTS devices aims to investigate

the instantaneous time-domain responses of these devices when controlling one or
more power system parameters. Generally, regardless of the FACTS device type and
application, dynamic modeling is divided into two main categories: “average model”
and “converter-level model”. In average modeling approach, the converter dynamics
are represented using linear time-domain differential equations in dq synchronous
rotating frame. In this modeling approach, discrete-time nature of converter
switching and the effects of harmonics are neglected. Converter output voltage is
usually described by the expression which is the function of a constant multiplication
factor (usually modulation index of PWM method), an angle (required phase shift of
converter voltage), and a conceptual DC link voltage. Modulation index and phase
shift of each converter are the two control inputs of each VSC of the FACTS device
to perform the given control task where the DC link dynamics are generally modeled
as a power balance equation in terms of dq components of voltage and current. In
this regard, IPFC is modeled using a set of linearized differential equations in
rotating dq frame (Menniti et al., 2002), (Moghadam et al., 2010), (Moghadam et al.,
2011), (Ajami et al., 2009). Voltage source representing each VSC of IPFC is
controlled by a duty-cycle vector which is applied to conceptual DC link voltage to
approximate converter switching (Strzelecki, et al., 2005).
In converter-level modeling approach, the studies are varied according to
converter structures. For example, each VSC of IPFC is modeled as PWM triggered
two-level six-pulse converter with a DC voltage source which approximates DC link
(Ye et al., 2006), with a DC link capacitor (Muruganandham, et al., 2012). Each VSC
of IPFC is modeled using three-level twelve-valve neutral-point-clamped converter
having two DC capacitors (Karthik, 2007). Switching frequencies are in the range
between 1-2 kHz. 48-pulse VSC consisting of four three-level converters is used in
IPFC configuration, with line real and reactive power flow control (Aali, et al.,
85
2010), with only bus voltage control (Bharathi et al., 2011). IPFC based on multi-
output sparse matrix converter switching at relatively high frequency of 10 kHz is
modeled (Hosseini et al., 2011).
Dynamic modeling studies of GUPFC are rather limited in literature when
compared with IPFC based studies although there are plenty of UPFC based dynamic
studies. For instance, average converter models are used for each VSC of GUPFC
(Lubis 2011a). Alternatively in average modeling approach, the controlled voltage
source representing each VSC of GUPFC is embedded into the power system model
directly and the simulation engine iteratively solves the system equations (Fardanesh
et al., 2000), (Sun et al., 2003).
Converter-level modeling studies of GUPFC have recently been published.
Elementary two-level six-pulse converter topology is the most common power circuit
scheme (Prakash et al., 2007), (Sujin et al., 2012), (Abdul et al., 2012). Sixty-pulse
converter model of GUPFC comprised of five three-phase three-level converters and
five phase-shifting transformers are presented (Lubis 2011b).
Dynamic modeling of BtB-STATCOM is approximated using average
modeling approach where each output of VSC is modeled as controllable ideal
voltage source (Tyagi et al., 2006), (Xinghao et al., 2009), (Lee et al., 2011),
(Parkhideh et al., 2009). On the other hand, converter-level models of BtB-
STATCOM are more detailed. For instance, two elementary two-level six-pulse
converters are used in BtB-STATCOM configuration (Ruihua et al., 2005), (Jovcic et
al., 2007), (Liu et al., 2010). Converter structure is pretty simple that does not reflect
realistic BtB-STATCOM operation completely. More detailed converter topologies
are alternatively considered. For instance quasi multi-pulse converter topology
consisting of sixteen six-pulse units are combined to build each VSC of BtB-
STATCOM (Hagiwara et al., 2003). The BtB system consists of two sets of four
three-phase neutral-point-clamped converter units each having twelve GTOs driven
by PWM (Hagiwara et al., 2005), (Hagiwara et al., 2008). 24-pulse three-level
voltage source converters with fundamental frequency switching for HVDC system
is proposed (Madhan Mohan, et al., 2009).
86
5.2. Simplex Optimization Method
Due to nonlinear nature of converter switching and the interactions among the
controllers of the multi-converter FACTS device, finding optimum parameters of the
control scheme while satisfying stable operation of the multi-converter FACTS
device is not an easy task. It is presented that the inherent dynamic interactions
between individual controllers of the UPFC can lead to unstable operation even
though each controller of UPFC itself is designed satisfactorily (Wang et al., 2000).
For a GUPFC, the situation can become desperately as more control functions are
attributed to GUPFC. One solution depends on analytical approaches such as ziegler-
nichols oscillation method, smith predictor, and pole assignment methods, which
require exact mathematical model of the system which is difficult to obtain without
simplification or averaging (Goodwin et al., 2000). Another solution may suffer from
the long computing time to find the optimum parameters where several simulation
runs exist to select the best parameters (PSCAD, 2005). Alternatively, a direct search
algorithm, which is called “simplex method”, is used in this research (Neider et al.,
1965) that is integrated into the PSCAD (Gole et al., 2005). This method does not
rely on gradient information and applicable for highly-nonlinear multi-input multi-
output systems without obtaining mathematical models and hence suitable for
finding the minimum of an objective or a cost function defined by several variables.
In this research, simplex method is executed not only to find the optimum multi-
controller parameters but also to find the best parameters for a specific designed
component in PSCAD.
Simplex is the name of a geometric figure whose vertices are defined by
variable numbers. For example, for two-variable optimization, simplex is a triangle,
for three-variable optimization, simplex is a tetrahedron. The problem becomes a
pattern search that compares function values at all vertices. The worst vertex, where
the cost function is the largest, is rejected and replaced by a new vertex. A new
simplex is formed until the function values at the vertices are the smallest. Simplex
size is then reduced iteratively and the coordinates of the minimum point are found.
The flow chart of the simplex optimization method is shown in Figure 5.1.
87
Set initial Evaluate objective

Start parameter set function by
simulating power
system embedded
with FACTS device
Update No Execute
parameter set Convergence simplex algorithm
?
Yes
Output
End
parameter set in a
file
Figure 5.1. Flow chart of the simplex optimization method in PSCAD
5.3. Converter-Level Modeling of GUPFC
5.3.1. GUPFC Interacting with Power System
Three quasi multi-pulse VSCs based GUPFC is positioned on WSCC 3-

Machine 9-Bus System as shown in Figure 5.2. It is aimed to control five power
system parameters simultaneously. These are real and reactive power flows of lines
L-45 and those of L-46, and Bus 4 voltage. GUPFC is in operation while switch, sw1
is closed, and switches, sw2-3 are opened. The AC terminal of shunt VSC (VSC1) is
connected to the Bus 4 through a magnetic interface which is conceptually drawn and
named as “tr1”. Similarly, the AC terminals of series VSCs (VSC2-3) are connected
to the neighboring transmission lines, namely, L-45 and L-46 through magnetic
interfaces, named as “tr2-3”. The DC terminals of each VSC are joined in a DC link,
which is represented by a capacitor C (0.2 F), provides real power exchange between
the three VSCs.
Each VSC of GUPFC is rated at 100 MVA. Each single-phase three-winding
transformer in twelve-pulse converter unit is rated at 60 Hz, 8.33 MVA, 20/2/1.154
kV with a leakage reactance of j0.025 pu. Each single-phase transformer of summing
and magnetic interface is rated at 60 Hz, 16.67 MVA, 137.92/38 kV, j0.1 pu for
88
shunt VSC. For series VSC, the rating is modified as 36/36 kV, j0.025 pu. Each
single-phase transformer of series coupling magnetic interface is rated at 60 Hz,
33.33 MVA, 104.13/40 kV, j0.01 pu.
Figure 5.2. WSCC 3-Machine 9-Bus System embedded with GUPFC
5.3.2. GUPFC Controller Design
The control of the quasi multi-pulse VSC depends on the decomposition of

the converter voltage into its d- and q-axis components as mentioned in Chapter 4. In
this regard, control loops of GUPFC are designed as in Figure 5.3. Once the
reference value of each controlled power system parameter is decided, an error signal
is generated by comparing the reference value with the actual or measured value of
the controlled variable. The error signal is then used as an input to the PI controller to
89
determine the required axis components of shunt and series converter voltages. The
control loops are designed according to the fact that the quadrature component (q-
axis) of the series injected voltage mainly controls real power flow, while the direct
component (d-axis) of the series injected voltage controls reactive power flow (Ye et
al., 2006), (Mishra, 2006), (Liming et al., 2007), (Xia et al., 2010). Since GUPFC’s
losses are met by VSC1, the shunt converter control is the DC link voltage control, E
and the voltage magnitude control of Bus 4, V4, achieved by VshD (Figure 5.3a) and
VshQ (Figure 5.3b), respectively. On the other hand, series VSC2 controls real and
reactive power flows of Line L-45, achieved by Vse2Q (Figure 5.3c) and Vse2D (Figure
5.3d), respectively. Series VSC3 controls real and reactive power flows of Line L-46,
achieved by Vse3Q (Figure 5.3e) and Vse3D (Figure 5.3f), respectively. A total of
twelve parameters of GUPFC’s control scheme (6xproportional gain, Kp and
6xintegration time constant τi) is optimized using simplex method to alleviate
controller interaction.
Figure 5.3. Control loops of GUPFC
90
5.3.3. Finding Optimum Controller Parameters
Optimal parameter set of GUPFC’s controllers given in equation (5.1) is

investigated iteratively by minimizing the cost function described in equation (5.2).
The derived cost function is based on the sum of integral square error (ISE) of
individual controllers. The PSCAD configuration for this task is shown in Figure 5.4.
{
p = K p1 , τ i1 , K p 2 , τ i 2 , K p 3 , τ i 3, K p 4 , τ i 4 , K p 5 , τ i 5 , K p 6 , τ i 6 } (5.1)
(
T
F ( p ) = ∫ (V 4 ref − V 4 ) 2 + ( E ref − E ) 2 + (Q 45 − Q45 ) 2 + ...
ref
t =0
... + ( P45 ref − P45 ) 2 + (Q46 ref − Q46 ) 2 + ( P46 ref − P46 ) 2 dt ) (5.2)
The total simulation time T in equation (5.2) is chosen much longer than the
settling time of the whole control system. Reference values of real and reactive
power flows are chosen same as in case 1 in the next section. F(p) is plotted against
iteration number in Figure 5.5 and the optimum parameters are listed in Table 5.1.
The algorithm is converged in 504 iterations for a tolerance of 1.0E-6. Due to
interaction between converters, the individual converter design is not preferred and
only one cost function is identified to obtain stable and reasonably dynamic
performance.
91
Figure 5.4. PSCAD implementation of simplex method
92
Figure 5.5. Convergence performance of cost function in simplex method
Table 5.1. Simplex optimized controller parameters of GUPFC

Shunt VSC1
AC voltage
DC voltage Kp1 Ti1 Kp2 Ti2
controller controller
initial initial
0.800000 0.010000 0.800000 0.010000
parameters parameters
optimized optimized
39.159948 0.007580 15.240716 0.000988
Series VSC2
Real power
Reactive power Kp3 Ti3 Kp4 Ti4
flow controller flow controller
initial initial
0.100000 0.001000 0.100000 0.001000
optimized optimized
15.227331 0.001005 0.795768 0.001003
Series VSC3
Real power
Reactive power Kp5 Ti5 Kp6 Ti6
flow controller flow controller
initial initial
0.100000 0.001000 0.200000 0.001000
optimized optimized
14.966115 0.001010 29.389397 0.001006
5.3.4. Simulation Studies
PSCAD environment is used to simulate the multi-machine power system

embedded with GUPFC as shown in Figure 5.2. The following case studies are
undertaken for evaluating the dynamic performance of the proposed GUPFC in
controlling real and reactive power flows of the lines along with bus voltage control
93
and under two types of faults. The simplex optimized controller parameters listed in
Table 5.1 are used in the simulations and the nominal DC link voltage of the GUPFC
is controlled at 2.0 kV throughout all simulation cases. Solution time-step is set to
100 µs in PSCAD.
5.3.4.1. Case 1: Start-up Transients
The start-up transients of the GUPFC are investigated when it is commanded

to increase L-45 real power flow from 0.7983 pu (uncompensated flow) to 1.0 pu
(20% increase) and L-46 real power flow from 0.5349 pu (uncompensated flow) to
0.74 pu (28% increase) at t=0+. At the same time, GUPFC holds both of the reactive
power flows of L-45 and L-46 at zero. Note that the uncompensated reactive power
flows of the lines are j0.1548 pu and j0.0573 pu, respectively. The simulation results
for this case study are shown in Figures 5.6a-l. No overshoot is observed in the
tracking signals of the real power flows in Figures 5.6a and 5.6c. P45 and P46 come to
their desired values with no steady-state error within around 1.0 s, respectively. The
maximum overshot/undershoot with the PI controllers for Q45 and Q46 is about 30%
(Figures 5.6b and 5.6d). The DC link voltage is a very important factor for successful
operation of shunt and series converters of GUPFC. On account of this, E is tightly
regulated at 2.0 kV (Figure 5.6e). Actual DC link voltage settles on the 2.0 kV line
within 1.0 s. Figure 5.6f shows the response of the power system to a step change in
V4 reference at 1.0 pu based on 230 kV transmission level. Note that the
uncompensated magnitude of V4 is 0.9828 pu and it is regulated by the shunt
converter of the GUPFC with no overshoot and with a reasonable transient
performance. The tracking performances of all individual power system parameters
are stable in this case study. The settling times of the control loops are generally in
the order of 1.0 s due to the fact that the time constant of the power system with the
GUPFC is relatively large.
94
(a) Real power flow control of Line L-45 at start-up
(b) Reactive power flow control of Line L-45 at start-up
(c) Real power flow control of Line L-46 at start-up
(d) Reactive power flow control of Line L-46 at start-up
95
(e) DC link voltage control of GUPFC at start-up
(f) Bus 4 voltage control at start-up
(g) Phase shifts for converters M and N of VSC1 (ΦM and ΦN) during start-up
(h) Phase shifts for converters M and N of VSC2 (ΦM and ΦN) during start-up
96
(i) Phase shifts for converters M and N of VSC3 (ΦM and ΦN) during start-up
(j) Series inserted voltage by VSC2 during start-up
(k) Series inserted voltage by VSC3 during start-up
(l) GTO anode to cathode voltage of VSC1 during start-up

Figure 5.6. Simulated waveforms of case 1
5.3.4.2. Case 2: Response to Real Power Flow Step Changes
Real power flow references for both lines (L-45, L-46) are decreased from 1.0
pu to 0.85 pu and from 0.74 pu to 0.60 pu at 10.5 s, simultaneously. The other
references of GUPFC control loops are kept unchanged as in case 1. Figures 5.7a-f
97
show the traces of real and reactive power flows of Line L-45 and L-46, GUPFC DC
link voltage, and V4, respectively. Real power traces reach to their reference values
within about 0.5 s after the step change command is applied with no steady-state
error. The transmission line reactive power flows stay constant as zero after
following a 6 % undershoot and 10 % overshoot in Q45 and Q46, respectively. PI
controller could keep GUPFC DC link voltage in its reference so that the trace has
almost no change towards step change command to real power flows. The response
of V4 after the step change command in real power flow references is almost constant
on 1.0 pu line. PI control scheme for each control loop in this case study exhibits
stable performance.
(a) Real power flow of Line L-45 for a step-change at 10.5 s
(b) Reactive power flow of Line L-45 in response to step-changes
98
(c) Real power flow of Line L-46 for a step-change at 10.5 s
(d) Reactive power flow of Line L-46 in response to step-changes
(e) Bus 4 line-to-line rms voltage voltage in response to step-changes
(f) DC link voltage of GUPFC in response to step-changes

99
5.3.4.3. Case 3: Response to Reactive Power Flow Step Changes
In this case, at 10.5 s, reactive power flow references for Line L-45 and L-46
are changed from 0.0 pu to -0.15 pu and from 0.0 pu to -0.10 pu, simultaneously. The
other references in GUPFC control loops are kept constant as in case 1. Figures 5.8a-
f show the traces of real and reactive power flows of Line L-45 and L-46, GUPFC
DC link voltage, and V4, respectively. When comparing cases 2 and 3 by examining
Figures 5.7a and 5.8b, the settling time of the reactive power control loop is about
0.5 s longer than that of real power control loop. The same amount of delay in the
response of reactive power control loop in Figure 5.7c is also observed when
compared with the response of real power control loop in Figure 5.8d. The real
power flow controllers response to the step change command of reactive power flows
with a 7% and 2.6 % overshoot, respectively. After following these transients,
GUPFC could bring real and reactive power flows to their desired values with no
steady-state error. GUPFC DC link voltage is controlled tightly so that the trace has
almost no change towards step change command to reactive power flows. When
comparing Figures 5.7f and 5.8f, the response of V4 after the step change in reactive
power flow reference is more ludic than the case after the step change in real power
flow reference. The sluggish response of GUPFC is due to the fact that voltage
magnitude is sensitive to reactive power.
(a) Real power flow of Line L-45 in response to step-changes
100
(b) Reactive power flow control of Line L-45 for a step-change at 10.5 s
(c) Real power flow of Line L-46 in response to step-changes
(d) Reactive power flow control of Line L-46 for a step-change at 10.5 s
e) Bus 4 line-to-line rms voltage in response to step-changes of reactive power flows
101
f) DC link voltage of GUPFC in response to step-changes of reactive power flows

5.3.4.4. Case 4: Single-phase to Ground Fault
The dynamic performance of the GUPFC is evaluated by applying phase-A to

ground fault at Bus 8 (Figure 5.2) at 10.5 s for 100 ms through an impedance of
j0.285 pu and cleared without any change in the network configuration. Probability
of realization of such a disturbance is high as the ratio of single-phase to ground
faults is about 80-81 % of fault types (Heine et al., 2003), (Bordalo et al., 2006). The
reference values of real and reactive power flows of Line L-45 and L-46 along with
V4 are controlled by GUPFC as in case 1. Figures 5.9a-f show comparative simulated
responses to the fault. In Figures 5.9a and 5.9b, the real and reactive power flows on
Line L-45 return to their controlled values within 1.0 s after the fault is applied with
and overshoot of about 10% and undershoot of about 5%. In Figure 5.9c, even real
power flow response of Line L-46 is more ludic than that of Line L-45, it shows less
oscillation and comes to its controlled value within 1.5 s. Reactive power flow on
Line L-46 has similar overshoot transients compared with the one on Line L-45, but
having smaller settling time. For fault condition, V4 drops to not greater than 2% of
its nominal value and shows a transient increase, not greater than 1% of its nominal
value. The disturbance lasts for 0.5 s. GUPFC DC link voltage oscillation is not
sustained following after the fault and the test system with GUPFC survive through
the fault and returns to stability smoothly in maximum 1.5 s in this case study.
102
Figures 5.9g-l also show the simulated voltage and current waveforms of the GUPFC
converters under single-phase to ground fault condition.
(a) Line L-45 real power flow response to phase-A to ground fault at 10.5 s
(b) Line L-45 reactive power flow response to phase-A to ground fault at 10.5 s
(c) Line L-46 real power flow response to phase-A to ground fault at 10.5 s
103
(d) Line L-46 reactive power flow response to phase-A to ground fault at 10.5 s
(e) Bus 4 line-to-line rms voltage response to phase-A to ground fault at 10.5 s
(f) GUPFC DC link voltage response to phase-A to ground fault at 10.5 s
(g) VSC1 injected voltage (shunt)
104
(h) VSC1 injected current (shunt)
(i) VSC2 injected voltage (primary side of series coupling transformer)
(j) VSC2 injected current (primary side of series coupling transformer)
(k) VSC3 injected voltage (primary side of series coupling transformer)
(l) VSC3 injected current (primary side of series coupling transformer)

105
5.3.4.5. Case 5: Three-phase to Ground Fault
A three-phase to ground fault with a duration of 100 ms is applied at Bus 9

(Figure 5.2) through an impedance of j0.285 pu at time 10.5 s. This disturbance is the
most severe disturbance with an occurrence probability of 1.5-3.0 % (Heine et al.,
2003), (Bordalo et al., 2006). The reference values of real and reactive power flows
of Line L-45 and Line L-46 along with V4 are controlled by GUPFC as in case 1.
Figures 5.10a-f show comparative simulated responses to the short circuit. Real
power flows on Line L-45 and L-46 are severely disturbed since their overshoot
magnitude is two times bigger than the cases in previous fault. But GUPFC retrieves
the flows to their controlled values in around 1.0 s. The waveforms in Figures 5.10b
and 5.10d indicate that for the same fault duration with the previous case, reactive
power flows on Line L-45 and L-46 result in more oscillation within approximately
same recovery time without losing stability. The transient in V4 during the
disturbance occurs between around 0.95 pu and 1.025 pu which is few wider than in
previous case due to influential fault. Combining cases 4 and 5, the voltage
excursions in DC link are avoided which is essential for the successful GUPFC
operation. Eventually it can be deduced that the GUPFC is able to mitigate the two
fault types within 1.0-1.5 s without losing stability.
(a) Line L-45 real power flow response to three-phase short circuit at 10.5 s
106
(b) Line L-45 reactive power flow response to three-phase short circuit at 10.5 s
(c) Line L-46 real power flow response to three-phase short circuit at 10.5 s
(d) Line L-46 reactive power flow response to three-phase short circuit at 10.5 s
(e) Bus 4 line-to-line rms voltage response to three-phase short circuit at 10.5 s
107
(f) GUPFC DC link voltage response to three-phase short circuit at 10.5 s

5.3.4.6. THD Content
Table 5.2 lists the average recorded THD values of the voltages measured
from three common coupling points (Buses 4-5-6) between GUPFC and the power
system when GUPFC is commanded to control real and reactive power flows of lines
L-45 and L-46 with the reference values given in case 1. Records of the simulation
run lasting for 12.5 s show that GUPFC switching at fundamental frequency of 60 Hz
does not cause the violation of the THD upper limit for 230 kV transmission level
(IEEE, 1993). It is seen that voltage distortions at common coupling points are within
the acceptable limits. Consequently, filtering is not required even GTOs are
switching at fundamental system frequency.
Table 5.2. THD values

Common coupling point Bus 4 Bus 5 Bus 6
THD (%) 1.50 0.52 0.43
5.3.5. Discussion
GUPFC is built using two series and one shunt quasi multi-pulse VSC
designed in Chapter 4. GUPFC dynamic performance on the control of real and
reactive power flows of two neighboring transmission lines and bus voltage control
is evaluated through different simulation scenarios including faults on WSCC 3-
108
Machine 9-Bus System. The robustness of the GUPFC controllers is ensured by

tuning of controller parameters using simplex method which does not rely on
mathematical model of the system. It is concluded that conventional PI controllers
are stable and yet practically applicable. This study verifies the control scheme of the
quasi multi-pulse VSC mentioned in Chapter 4 in one respect. It is also noted that the
quasi multi-pulse VSCs do not inject any harmonics with no more than THD=1.5%,
which complies with the IEEE regulations. In this regard, high switching frequency
or filter circuit is not required.
5.4. Converter-Level Modeling of IPFC
5.4.1. IPFC Interacting with Power System
Two quasi multi-pulse VSCs based IPFC is located on 4-Machine 4-Bus

System as shown in Figure 5.11 whose data are given in Appendix B. A total of three
power system parameters are aimed to control simultaneously in this simulation
configuration. These are real and reactive power flows of Line-2 and real power flow
of Line-1. IPFC is activated while switches sw1 and sw2 are opened. The AC
terminal of VSC1 and that of VSC2 are connected to two neighboring transmission
lines (Lines 1 and 2) through magnetic interfaces, named as “tr1-2”, respectively.
The DC terminals of each VSC are joined in a DC link, which is represented by
capacitor C (0.2 F), provides real power exchange between the two VSCs. Each VSC
of IPFC is rated at 100 MVA. Each single-phase three-winding transformer in
twelve-pulse converter unit is rated at 50 Hz, 8.33 MVA, 10/1/0.5774 kV with a
leakage reactance of j0.1 pu. Each single-phase transformer of summing and
magnetic interface is rated at 50 Hz, 16.67 MVA, 23.42/23.42 kV, j0.1 pu. Each
33.33 MVA, 23.42/9.0 kV, j0.01 pu.
109
Figure 5.11. 4-Machine 4-Bus System embedded with IPFC
5.4.2. IPFC Controller Design
The dynamic performance of IPFC suffers from the strong dynamic interaction
between real and reactive power flows due to inherent properties of AC power
transmission. To reduce or eliminate this coupling effect, a number of studies on other
members of FACTS devices are available in literature. It is realized that these studies
rely on decoupling parameters of approximated FACTS device model or converter-
level model of elementary six-pulse VSCs driven by high frequency PWM methods,
which is not realistic for high power applications. Firstly, a d-q current controller is
proposed with no-cross coupling for a grid connected inverter (Schauder, 1991). Later
on a new control scheme originated from this controller in which a decoupled
controller with an internal predictive loop for UPFC is suggested (Papic et al., 1997).
In the proposed control scheme, the parameters of reactance of series coupling
transformer and system bandwidth are required for gain design. Other articles propose
different types of decoupled controllers for non-converter-level model of UPFC where
an equivalent ideal voltage source model of UPFC is considered with no harmonics
(Yu et al., 1996), (Yam et al., 2002), (Papic et al., 2003), (Ye et al., 2006), (Farahani et
al., 2006), (Ande et al., 2007), (Ma, 2007). A decoupling controller is designed, but
110
control performance counts on exact system parameters and UPFC model (Yu et al.,
1996). A dynamic decoupled compensator for UPFC is designed (Yam et al., 2002).
The design relies on classical control design techniques which rely on exact
mathematical model of the system, damping ratio and system bandwidth should also
be exactly known. A decoupling matrix compensator consisting of four controllers is
developed that relies on ABCD parameters of approximated UPFC model (Farahani et
al., 2006). A decoupled UPFC controller for dynamic control of real and reactive
power flows is considered (Ande et al., 2007). UPFC is experimentally validated by
six-pulse VSCs where PWM control is used (Liming et al., 2005). To achieve
decoupling, reactance values of shunt and series coupling transformers should be
exactly known.
In this research, the decoupling effect between real and reactive power flow
control loops is reduced by a new hybrid fuzzy PI (HFPI) control scheme applied to
IPFC. The proposed controller is based on conventional simplex optimized PI
controller operating in conjunction with Mamdani-type fuzzy inference system with
linearly distributed linguistic rules. With this way, a fast response is obtained with
minimal interaction to track the changes in reference values of the real and reactive
power flows. Design phase neither requires exact mathematical description nor system
transfer function. The performance of the proposed HFPI controller is compared with
both conventional PI control and PI control with analytically computed feed-forward
decoupled gains.
5.4.2.1. Decoupled Controller Design
Assuming series resistance and inductance of tr1 in Figure 5.11 are included
into the transmission line parameters RL and XL, respectively. Then, the current on
Line-1, IL can be derived as
VS − V R − V X
IL = (5.3)
R L + jX L
111
where VX is line-to-neutral rms voltage phasor of series injected voltage which is

synthesizes by VSC1, VS and VR are the line-to-neutral rms voltage phasors of the
sending-end and receiving-end sides, respectively. Complex power at the sending-end
side is
S s = Ps + jQs = 3VS I L * (5.4)
where symbol (*) denotes complex conjugate and PS and QS denote sending-end real
and reactive power flows on Line-1, respectively. Assuming VR leads VS by a small
angle δ (cos δ≈1, sinδ≈0), PS and QS can be expressed as
 Ps  RL 2 + X L 2  VQ  RL  Qs − Qs 0   Ps 0 
Q  = A  − V  + X − P + P  + Q  (5.5)
 s X L2  D L  s s0   s0 
where A=3VS/(RL+XL), VD and VQ are the d- and q-axis components of VX in rotating

reference frame, respectively (VX=VD+jVQ). PS0 and QS0 are the uncompensated real
and reactive power flows when there is no compensation (VD=VQ=0). In equation (5.5),
real and reactive power flows are naturally coupled and needs to be decoupled for
efficient dynamic power flow control. Taking first-derivative of equation (5.5) with
respect to time yields equation (5.6). Assuming PS0 and QS0 are at certain values and VS
is regulated at a constant value.
d  PS  RL 2 + X L 2 d  VQ  RL d  QS 
Q  = +
dt − VD  X L dt
A − P  (5.6)
dt  S X L2  S
The derivative terms in equation (5.6) can be approximated using forward

difference operator with a small time Δt as represented below:
1  ∆PS  RL 2 + X L 2  ∆VQ  RL  ∆QS 

∆Q  = A  − ∆V  + ∆tX − ∆P  (5.7)
∆t  S ∆tX L 2  D L  S
112
where df/dt≈Δf/Δt with Δf=f(t+Δt)˗f(t) (Levy et al., 2011). According to equation

(5.7), the required changes in PS and QS are respectively related to VQ and VD. This
result also confirms the GUPFC’s controllers design, given in Section 5.3.2. The last
summing terms are the coupling terms and can vanish if the following feed-forward
gains are added to the conventional PI control scheme once the reference of power
flows (PSref and QSref) are externally defined by the user or supervisory control.
1 R
VQ ref = K p1 ( PS ref − PS ) + ∫ ( PS ref − PS )dt − L Q S
Ti1 XL
1 R
− VD ref = K p 2 (QS ref − QS ) + ∫ (QS ref − QS )dt + L PS (5.8)
Ti 2 XL
where VQref and VDref denote desired d-q components of series converter voltage. PS
and QS are respectively the current values of real and reactive power flows measured at
time t. Kp1 and Kp2 are the proportional gains of real and reactive power flow
controllers, respectively. Ti1 and Ti2 are the integration time constants of real and
reactive power flow controllers, respectively. In this case, the control scheme
mentioned so far is regarded as PI control with decoupled gains (PI+DG). PSCAD
implementation of PI+DG control scheme is shown in Figure 5.12. Noting that if RL/XL
ratios are set to zero or port A of sum blocks are disabled then the control scheme
simply becomes PI control.
Figure 5.12. PSCAD implementation of PI+DG controllers
113
5.4.2.2. Proposed Hybrid Fuzzy PI (HFPI) Controller
In previous section, decoupling gain design is made analytically under some

assumptions in system model. Moreover the dynamic performance of PI+DG relies on
exact knowledge of RL and XL which can change due to environmental factors. In this
regard, a HFPI controller is designed in support to PI controllers as shown in Figure
5.13. The supplementary signals (delVd and delVq) are generated by a Mamdani-type
fuzzy inference system in MATLAB fuzzy logic toolbox and called “fuzzy decoupler”
(FUDE), which is realized using heuristic information based on coupling
characteristics. It is aimed that the power flow controller adapts itself based on
instantaneous system states rather than off-line system parameters which are
substantially liable to changes during real-life operation. FUDE which is designed in
MATLAB communicates online with PSCAD in real-time. For this purpose, a module
is prepared in PSCAD to link the MATLAB as shown in Figure 5.14. PSCAD
communicates with a MATLAB m-file through FORTRAN scripts written in that
module. All related codes are presented in Appendix E. PSCAD and MATLAB
exchange information at every time step in a continuous manner.
Figure 5.13. PSCAD implementation of HFPI controller
114
Figure 5.14. PSCAD-MATLAB interface
FUDE is activated by a set-point change detector (SEPOCHDET) when the

first set-point change either in PS or QS occurs. In this sense, SEPOCHDET shown in
Figure 5.15 keeps the advantage of executing simple PI controllers during start-up.
First derivatives of PSref and QSref are absolute valued and sent to a monostable
multivibrator which is a binary-logic, edge-triggered PSCAD component. A positive
edge on its input results in the output going high and remaining high for the rest of the
simulation after being turned on. Consequently, SEPOCHDET produces the signal
“defico” as logic one to connect the outputs of FUDE to the PI controllers to reduce
interaction between real and reactive power flow controllers during set-point changes.
Figure 5.15. PSCAD implementation of SEPOCHDET
115
5.4.2.3. FUDE Design
The system response is examined for sequences of set-point changes when only
PI controllers with optimum parameters are employed. For example, if QS hugely
deviates from its set-point while PSref is decreased sharply, a large control signal ΔVD
that pulls it toward to its set-point is expected. Similarly, when QSref is suddenly
increased, PS tends to decrease and a large control signal ΔVQ is required. As a first
step, x(k) is defined as the input set of crisp numerical signals of Pe, ΔPe, Qe, ΔQe at
sampling instant k, limited to its universe of discourse. Pe and Qe are the real and
reactive power flow errors, ΔPe and ΔQe are the real and reactive power flow error
rates, respectively. x(k) is then fuzzified according to seven linguistic characteristics,
defined for its each element. Abbreviations in Figure 5.16 for the membership
functions (MFs) that quantify the meaning of linguistic characteristics are the
following: N3: big negative, N2: medium negative, N1: small negative, Z: zero, P1:
small positive, P2: medium positive, and P3: big positive. Intersection point M is
specific for each member in x(k).
Figure 5.16. Universe of Discourse
Output set y(k) also needs fuzzification at the sampling instant k using
membership function (MF) set for ΔVQ and ΔVD depicted in Figure 5.17. Intersection
point N is specific for each member in y(k).
116
Figure 5.17. MFs for FUDE output set
Rule base for output ΔVQ is listed in Table 5.3 which is identical to that of ΔVD,
but designed for ΔQe/Qe. Every entity merges the error rate and the error fuzzy set
values. For instance, first rule is
R1: If ΔPe is P3 and Pe is P3 then ΔVQ is P3
Table 5.3. Rule base for ΔVQ

ΔPe/Pe N3 N2 N1 Z P1 P2 P3
N3 N3 N3 N3 N2 N2 N1 Z
N2 N3 N3 N2 N2 N1 Z P1
N1 N3 N2 N2 N1 Z P1 P2
Z N2 N2 N1 Z P1 P2 P2
P1 N2 N1 Z P1 P2 P2 P3
P2 N1 Z P1 P2 P2 P3 P3
P3 Z P1 P2 P2 P3 P3 P3
In the next step, the min fuzzy operator is applied as the antecedent of the rule,
which has more than one part that should be ANDed with each other. The min fuzzy
operator is also used in the implication step, implemented for each rule. Here, the
output fuzzy set is truncated by a real number given by the antecedent of the rule. The
result of implication is innately fuzzy, so to determine crisp outputs (ΔVQ, ΔVD), the
popular centroid defuzzification scheme is utilized as the last step. Finally, the actual
outputs of FUDE are obtained. For instance, ΔVQ at the sampling instant k can be
written using equation (5.9). µ(i) and bi are the aforementioned MF and the center of
MF of the consequent of rule i, respectively. The control surfaces of the proposed
FUDE are shown in Figure 5.18.
117
∑i49=1 bi ∫ µ (i )
∆VQ (k ) = 49
(5.9)
∑ ∫ µ (i )
i =1
Figure 5.18. Control surfaces of the proposed FUDE
The conceptual controller configurations presented in terms of block diagrams

are illustrated in Figure 5.19 for VSC2 or “the master VSC” of IPFC where real and
reactive power flow control is made simultaneously. The configuration is implemented
in two stages, outer control loop by i) HFPI controller, ii) PI+DG controller, iii) PI
controller and inner control loop to implement 2-angle control method mentioned in
previous chapter. Limiters can be either defined externally or internally as the PI
controller parameters. They are used to limit the values of d-q voltage components by
consideration of the maximum voltage generation capacity of each VSC of IPFC.
VSC1 or “the slave VSC” of IPFC regulates DC link capacitor voltage and
controls real power flow on Line-1, simultaneously. The control scheme based on
parameter optimized PI controllers is shown in Figure 5.20. Error in DC link voltage
drives PI controller to produce d-component of VSC1 output voltage to achieve DC
link voltage control. Similarly, real power flow control on Line-1 is carried out by q-
component of VSC1 output voltage.
118
Figure 5.19. Conceptual control configurations for the master VSC
Figure 5.20. Control scheme for the slave VSC
5.4.3. Finding Optimum Controller Parameters
Simplex method iteratively finds the optimal parameter set p={Kp1, Kp2, Ti1,
Ti2} for PI controllers of the master VSC and the slave VSC by minimizing the cost
functions given in equations (5.10) and (5.11) depending on sum of ISEs of the
controlled variables. PSCAD implementation of simplex method is shown in Figure
5.21.
119
( )
T
F ( p ) = 100 ∫ ( P2 ref − P2 )2 + (Q2 ref − Q2 )2 dt (5.10)
t =0
( )
T
H ( p ) = 100 ∫ ( P1ref − P1 ) 2 + ( E ref − E ) 2 dt (5.11)
t =0
T is total simulation time in equations (5.10) and (5.11). While FUDE is off,
simplex method is executed for a sequence of unit step changes applied to P2ref and
Q2ref for Line-2, which are same as in case 1 in the next section. During optimization
routine, the variations of F(p) and H(p) against iteration number are plotted in Figures
5.22a and 5.22b, respectively.
(a) Cost function for master controller
(b) Cost function for slave controller

Figure 5.22. Cost function minimization in simplex method
120
Optimum parameter set is listed in Table 5.4. First, parameters of the master
control scheme are optimized using equation (5.10) under the condition that slave
controller is employed with pre-defined parameters providing a robust and stable IPFC
performance. Here it is not ensured that these parameters are optimal, but they give
satisfactory dynamic performance. Secondly, parameters of slave control scheme are
optimized using equation (5.11) while the solution of the first case results is applied to
master control scheme. The algorithm is executed for a tolerance of 1.0E-6.
Table 5.4. Simplex optimized controller parameters of IPFC

F(p)
Method Kp1 Ti1 Kp2 Ti2
VSC2
Non-
9.1940 0.8 0.01 0.8 0.01
optimized
Simplex 0.2930 192 0.00086 192 0.00124
H(p)
Method Kp1 Ti1 Kp2 Ti2
VSC1
Non-
266.886 0.01 0.001 0.01 0.001
optimized
Simplex 43.7829 0.05154 0.00529 0.01489 0.00434
To test and evaluate the decoupling performances of different controllers, 4-

Machine 4-Bus System embedded with IPFC shown in Figure 5.11 is simulated in
PSCAD. Solution time-step is set to 100 µs. While IPFC is de-activated when the
switches (sw1, sw2) are closed, real and reactive power flows of Lines 1 and 2 are
measured to design reasonable set-point changes. The following control tasks and the
controllers are considered for the case studies:
• Line-1 sending-end real and reactive power flow control by optimized PI

controllers
• IPFC DC link voltage control by optimized PI controller
121
• Line-2 sending-end real and reactive power flow control by

- Optimized PI controllers (with zero-decoupled gains)
- Optimized PI controllers with decoupled gains (PI+DG)
- Optimized PI controllers with FUDE (HFPI controller)
5.4.4.1. Case 1
In this case study, IPFC is activated by opening the switches (sw1, sw2) and
the dynamic performances of aforementioned controllers are simulated and compared
when the system is subjected to a sequence of unit-step changes in real and reactive
power flow commands of Line-2. Reference for real power flow on Line-1 is set as 2.3
pu and IPFC DC link voltage is regulated at 1.0 kV throughout the case study. As
observed in Figure 5.23, reactive power flow command is altered to force coupling
during the instants when real power flow command is constant. Although PI controller
is parameter optimized, relatively large fluctuations in real power flow are observed at
times, t=1.0 s, 2.0 s, and 3.0 s, respectively (Figures 5.23a-c).
PI controller with decoupled gains (PI+DG) gives better results when the
dynamic performance is compared with that of PI controller only. Although PI
controller or PI+DG gives satisfactory steady-state tracking performance, inherent
coupling between power flow control loops are not avoided and IPFC dynamic
performance is adversely affected. On the other hand, HFPI controller has the superior
decoupling feature as evidence from the response curves since the variations in real
power flow is effectively minimized when reactive power flow command is changed.
Moreover, Figures 5.23d-f gives a comparison between the responses of
different controllers to step-changes in real power flow command. HFPI controller
responses with less oscillations and shows reduced overshoot characteristics. The
dynamic performance of reactive power flow control loop with different control
schemes are also evaluated in this case study.
Figure 5.24 shows the traces of different reactive power flow controllers in
response to unit-step change in real power flow command. As shown in Figures 5.24a-
c, HFPI controller performance is superior to either PI controller or PI+DG on tracking
122
reference signal and HFPI controller effectively minimizes the coupling effect between
two power flow control loops.
Figure 5.23. Dynamic performances of real power flow controllers
As the consequence of unit-step command, reactive power flow fluctuations are

minimized better by HFPI controller with less oscillatory and reduced overshoot
response when compared with the other control schemes. Two commonly used
measures for control system performance, namely ISE and integral absolute error
(IAE) are calculated for (0.9 s ≤ t ≤ 5.0 s) in Table 5.5 to have a quantitative and exact
comparison between different control schemes.
Figure 5.25 shows the dynamic performance of slave VSC real power flow
controller and it is found that among the three controllers, the variations are the
smallest in case of HFPI controller which gives a smoother response when compared
123
with PI+DG. DC link voltage excursions of IPFC for different control schemes are
depicted in Figure 5.26. DC voltage controller is almost robust and gives satisfactory
response for all control modes. But when the comparisons are particularly made at
instants (t=1.0 s, 1.5 s, 2.0 s, 2.5 s, 3.0 s, 3.5 s), relatively smaller spikes are observed
at the simulated waveforms in case of HFPI controller. Figure 5.27 compares the d-q
components of injected current of the master converter in case of three controllers.
Prominent time instants are marked with red rectangles when real power flow
reference is changed in case of iD and when reactive power flow reference is changed
in case of iQ. These spikes in marked regions showing the interactions between the two
power flow controllers are effectively reduced by the proposed HFPI controller.
Although the spikes caused by HFPI controller are practically the same when
compared with the ones caused by PI controller, HFPI controller weakens the spikes
much better than PI+DG. Figure 5.28 shows control signals (VDref and VQref) for inner
control loop and the measured voltages (VD and VQ) of the master converter at the
primary windings of series coupling transformer Tr1. It is ensured that the “2-angle
control” block operates stable and the orthogonal components of the master converter
voltage perfectly trace their pertinent reference values in case of HFPI controller.
Figure 5.29 depicts anode-to-cathode voltage of one selected GTO from Group M of
the master converter in case HFPI controller is activated. As designed for quasi multi-
pulse operation, GTO is triggered only once in one fundamental cycle of 50 Hz.
5.4.4.2. Case 2
In this case study, controller references are kept exactly the same as in case 1
and RL/XL ratio of the Line-2 is increased by three times to investigate and compare
the parameter sensitivity of the three control schemes. Figures 5.30 and 5.31 show
comparative tracking performances of the controllers for real and reactive power
flows of the Line-2, respectively. ISE and IAE performance indices are listed in
Table 5.5 for 0.9 s ≤ t ≤ 5.0 s. As shown in Figures 5.30a-c, real power flow control
loop is interacted adversely with reactive power flow control loop when PI+DG is
employed.
124
Figure 5.24. Dynamic performances of reactive power flow controllers
Figure 5.25. Dynamic performance of real power flow controller for slave VSC
125
Figure 5.26. Dynamic performance of DC voltage controller for slave VSC
Figure 5.27. d-q components of master VSC injected current
Figure 5.28. d-q components of master VSC voltage by HFPI controller
126
Figure 5.29. Anode-to-cathode voltage of a selected GTO in converter M
Figure 5.30. Dynamic performances of real power flow controllers
PI+DG has the maximum overshoot of all controllers and gives relatively the
slowest response when compared with the other control schemes. The same situation
is also observed in Figures 5.31a-c when reactive power flow of Line-2 is controlled
by PI+DG during set-point changes in real power flow. As expected, the
performance of PI+DG for both real and reactive power flow control loops degrades
127
significantly, since decoupled gains are designed offline using transmission line data.
When comparing cases 1 and 2 quantitatively, an increase of 32.68% in ISE and an
increase of 80% in IAE are observed for real power PI+DG controller. Similarly, an
increase of 62.55% in ISE and an increase of 127.18% in IAE are observed for
reactive power PI+DG controller. It is because of the PI gains are optimized for the
operating conditions in case 1 that the dynamic performance of PI controller
slightingly weakens when compared with that of case 1.
Even though the system parameters are changed, HFPI controller successfully
reduces the interactions between real and reactive power flows with the lowest ISE
and IAE indices when compared with either PI controller or PI+DG. Furthermore, it
is observed in Figure 5.30d-f and in Figure 5.31d-f that HFPI controller gives a
smooth response and greatly improves rise time and settling time of the control loops
when responding to set-point changes.
Table 5.6 lists the highest THD values computed using the first 63 harmonics
at four common coupling points between IPFC and the power system. Records for
1.0 s≤t≤5.0 s confirm that IPFC does not cause the violation of the THD upper limit
of 2.5 % for 154 kV transmission level (IEEE, 1993). Consequently, filtering is not
required even GTOs are switched at fundamental frequency.
128
Figure 5.31. Dynamic performances of reactive power flow controllers
Table 5.5. Quantitative performance analysis of different controllers

Case 1 Case 2
Control Control
Action Controller ISE IAE Action Controller ISE IAE
for Line-2 for Line-2
PI 5.8784 136.2015 PI 5.8918 140.4160
real power real power
PI+DG 5.3502 115.8114 PI+DG 7.0987 208.7502
flow flow
HFPI 4.5300 88.8933 HFPI 5.5127 112.5168
reactive PI 4.5269 168.0034 reactive PI 4.5761 172.4952
power PI+DG 3.6717 116.8833 Power PI+DG 5.9683 265.5316
flow HFPI 2.6631 73.9114 flow HFPI 2.9794 69.4454
129
Table 5.6. THD values in case of three control schemes

Case 1 Case 2
Controller THD@Bus1 THD@Bus2 Controller THD@Bus1 THD@Bus2
PI 0.48 % 0.75 % PI 0.45 % 0.70 %
PI+DG 0.48 % 0.74 % PI+DG 0.46 % 0.70 %
HFPI 0.46 % 0.71 % HFPI 0.46 % 0.69 %
Controller THD@Bus3 THD@Bus4 Controller THD@Bus3 THD@Bus4
PI 0.91 % 0.98 % PI 0.80 % 0.89 %
PI+DG 0.90 % 0.97 % PI+DG 0.82 % 0.89 %
HFPI 1.3 % 1.51 % HFPI 1.21 % 1.47 %
5.4.5. Discussion
The proposed HFPI controller minimizes the interactions between the control
loops of real and reactive power flows and gives a smoother response when
compared with either PI+DG or PI controller. Even system coefficients change, it is
still able to alleviate these interactions and robust to uncertainty. On the other hand,
the performance of PI+DG strongly relies on the knowledge of system parameters
and only performs better than PI controller under the condition that the model
parameters match with the parameters of decoupled gain design. HFPI controller
does not disturb other IPFC control loops, such as power flow control on Line-1 and
DC link voltage control although it introduces small voltage ripples in the DC
interface. So the interactions between the controllers are obtained minimum for
multi-functioning FACTS device which is highly desired.
5.5. Converter-Level Modeling of BtB-STATCOM
5.5.1. BtB-STATCOM Interacting with Power System
BtB-STATCOM having two quasi multi-pulse VSCs is located between two

neighboring buses, Buses 1 and 3 of 3-Machine 7-Bus System as shown in Figure
5.32. The same system is used for power flow studies of BtB-STATCOM in Chapter
3.
130
Figure 5.32. 3-Machine 7-Bus System embedded with BtB-STATCOM
Two-VSC configuration extends the capabilities of conventional STATCOM

that the bi-directional real power transfer from one bus to another is possible. The
AC terminal of VSC1 is connected at Bus 1 through a magnetic interface which is
conceptually drawn and named as “tr1”. Similarly, the AC terminal of VSC2 is
connected at Bus 3 through “tr2”. A total of three power system parameters are
aimed to control simultaneously in this simulation configuration. These are voltage
magnitudes of Buses 1 and 3 and DC power flow among Buses 1 and 3 in two
directions. The DC terminals of each VSC are joined in a DC link, which is
represented by capacitor C (0.2 F), provides real power exchange among the two
converters. Each VSC of BtB-STATCOM is rated at 100 MVA. Each single-phase
three-winding transformer in twelve-pulse converter unit is rated at 50 Hz, 8.33
MVA, 20/2/1.1548 kV with a leakage reactance of j0.025 pu. Each single-phase
transformer of summing and magnetic interface is rated at 50 Hz, 16.67 MVA,
137.92/38 kV, j0.1 pu.
131
5.5.2. BtB-STATCOM Controller Design
In BtB-STATCOM, two-degrees of freedom (voltage magnitude and phase

angle) for each VSC is based on the decomposition of quasi multi-pulse VSC voltage
into its d- and q-axis components as mentioned in Chapter 4. The inputs of 2-angle
control method, VDref and VQref of each VSC can be calculated from the PI control
blocks as shown in Figure 5.33. VSC1 control, the same as that of shunt VSC of
GUPFC, is the DC link voltage control, E and the voltage magnitude control of Bus
1, V1, which are achieved by Vsh1D (Figure 5.32a and Vsh1Q (Figure 5.32b),
respectively.
Figure 5.33. Control loops of BtB-STATCOM
VSC2 control is the real power transfer among the two VSCs through the DC
link and voltage magnitude control of Bus 3, V3, achieved by Vsh2D (Figure 5.32c)
Vsh2Q (Figure 5.32d), respectively. The chosen PI parameters for VSC1 and those of
VSC2 of BtB-STATCOM are given in Appendix C.
132
The power system embedded with BtB-STATCOM, shown in Figure 5.32,

are simulated in PSCAD to test quasi multi-pulse VSCs and their control schemes for
BtB-STATCOM operation through comprehensive simulation cases. Solution time
step of PSCAD is set as 100 μs. DC link voltage is controlled at 2.0 kV throughout
all cases.
5.5.3.1. Case1: Start-up Transients
BtB-STATCOM is started up to control bus voltages (V1 and V3) at 1.0 pu

and to enable real power transfer between Buses 1 and 3. Noting that uncontrolled
bus voltages are 0.99 pu and 0.97 pu on 154 kV base, respectively. Real power
transfer (Ptransfer2) is set at 0.5 pu on 100 MVA base. Since start-up transients are
highly dependent upon voltage ramp up time of the generators in PSCAD, a ramp-up
time of 0.05 s, divisible into the period of the fundamental frequency, is selected
(PSCAD, 2005). The results are graphically presented in Figures 5.34a-j. No
overshoot is observed in the tracking signals of bus voltages in Figures 5.34a and
5.34b. V1 and V3 come to their desired values with no steady-state error within 0.06 s
and 0.8 s, respectively. This is due to different reactive power requirements of each
bus. The steady-state magnitudes of injected reactive powers into Buses 1 and 3 are
13 MVAR and 65 MVAR, respectively. Figure 5.34c shows the response of the real
power transfer from Bus 1 to Bus 3 to a step change of 0.5 pu at t=0+. After 1.4 s,
real power flow comes to its desired value with no overshoot. The DC link voltage
regulation is a key factor for the successful operation of voltage source based
converters. On account of this, E is tightly regulated at 2.0 kV (Figure 5.34d). Actual
DC link voltage settles on 2.0 kV line within 0.2 s. Figures 5.34e-j show some
related signals taken from the simulation.
133
(a) Bus 1 line-to-line rms voltage control at start-up
(b) Bus 3 line-to-line rms voltage control at start-up
(c) Real power transfer control from Bus1 to Bus3 at start-up
(d) DC link voltage control at start-up
(e) Phase shifts for converters M and N of VSC1 (ΦM and ΦN) during start-up
134
(f) Phase shifts for converters M and N of VSC2 (ΦM and ΦN) during start-up
(g) d-q components of VSC1 voltage during start-up
(h) d-q components of VSC2 voltage during start-up
(i) Injected three-phase currents by VSC1 during start-up
(j) Injected three-phase currents by VSC2 during start-up

135
5.5.3.2. Case 2: Response to Real Power Transfer Step Changes
BtB-STATCOM is commanded to reverse real power transfer between Buses

1 and 3 from 0.4 pu to -0.4 pu at 4.0 s and restore it back to its previous value at 6.0
s. The other reference points of BtB-STATCOM control loops are kept unchanged as
in case 1. Figures 5.35a-h show the traces of this case study. Figure 5.35a shows that
real power transfer reaches to its commanded values in around 1.0 s after the step
change command is applied. The response is robust with no steady-state error. The
DC link voltage stays constant after following a 20 % undershoot and 25 %
overshoot. The response of V1 and V3 is constant on 1.0 pu line and not shown here.
Although practical PI controllers are utilized, all responses regarding this case study
exhibit satisfactory and stable performance. BtB-STATCOM is able to reverse the
direction of real power transfer between neighboring buses without disturbing other
control parameters.
(a) Reversing real power transfer from Bus1 to Bus3
(b) DC link voltage transients
136
(c) Phase shifts for converters M and N of VSC1 (ΦM and ΦN)
(d) Phase shifts for converters M and N of VSC2 (ΦM and ΦN)
(e) d-q components of VSC1 voltage
(f) d-q components of VSC1 voltage
(g) Injected three-phase currents by VSC1
137
(h) Injected three-phase currents by VSC2

The dynamic performance of the BtB-STATCOM is evaluated by applying

phase-A to ground fault at the middle of the Line 3-P at 10.5 s for 100 ms and
cleared without any change in the network configuration. The reference values of all
control loops of BtB-STATCOM are kept constant as in case 1. Figures 5.36a-j show
simulated responses to the fault. The rms voltage magnitudes of compensated buses
without/with BtB-STATCOM are compared and presented in Figures 5.36a and
5.36b. V1 and V3 returns to their controlled values within 1.0 s after the fault is
cleared and an overshoot of less than 0.25 % and 2.5% is observed, respectively.
Voltage sag does not go below more than 0.5 % of nominal value of V1 and 7.5 % of
nominal value of V3, respectively. For both measurements, BtB-STATCOM could
hold voltage levels higher than the uncompensated case during fault. Real power
transfer drops around 17.0 % of its nominal value during fault but it restores within
1.65 s with an overshoot of 2.5 % after the fault is cleared (Figure 5.36c). PI control
could hardly keep DC link voltage at its steady-state value against sustained
oscillation, but DC link voltage returns to its controlled value within around 0.7 s
following after the fault (Figure 5.36d). It is deduced that the magnitude of the real
power transfer is inversely proportional to the DC link voltage controller
performance. Tracking performances of all PI controllers show that the controlled
parameters do not pass to unstable region even though the controllers have simple
structure in nature. Figures 5.36e-j demonstrates some recorded signals of BtB-
STATCOM related with this case study.
138
(a) Bus 1 voltage response to phase-A to ground fault
(b) Bus 3 voltage response to phase-A to ground fault at 10.5 s
(c) Real power transfer response to phase-A to ground fault at 10.5 s
(d) DC link voltage response to phase-A to ground fault at 10.5 s
(e) Phase shifts for converters M and N of VSC1 (ΦM and ΦN)
139
(f) Phase shifts for converters M and N of VSC2 (ΦM and ΦN)
(g) VSC1 controller output signals in response to phase-A to ground fault at 10.5 s
(h) VSC2 controller output signals in response to phase-A to ground fault at 10.5 s
(i) Injected three-phase currents by VSC1
(j) Injected three-phase currents by VSC2

140
A three-phase to ground fault, which is the most severe disturbance is applied

to the middle of Line 3-P at 10.5 s for 100 ms and cleared. The reference values of
all control loops of BtB-STATCOM are kept constant as in case 1. Figures 5.37a-j
show comparative simulated responses to the short circuit. Voltage sag less than 1.5
% and swell no more than 1.0 % are recorded for V1 in Figure 5.37a. BtB-
STATCOM retrieves the voltage magnitude to its controlled value in less than 1.5 s.
In Figure 5.37b, voltage sag for Bus 3 goes 40% of nominal value and recovers in
around 1.5 s after the fault is cleared. Real power transfer is severely disturbed since
Bus 3 is next to the fault location. It drops to a negative value that the direction of
flow is temporary reversed from 0.4 pu to -0.04 pu due to fault. The disturbance lasts
no more than 2.5 s. Combining Figures 5.37c and 5.37d, the excursions occurred in
real power transfer for this case study is more than those of case 3. DC link voltage
recorded in Figure 5.37d is more ludic than the trace observed during phase-A to
ground fault condition (Figure 5.36d). The performances of PI controllers are
relatively poor in case of three-phase fault but without losing stability. Figures 5.37e-
j also present some recorded signals of BtB-STATCOM related with this case study.
(a) Bus 1 voltage response to three-phase fault
(b) Bus 3 voltage response to three-phase fault at 10.5 s
141
(c) Real power transfer response to three-phase fault at 10.5 s
(d) DC link voltage response to three-phase fault at 10.5 s
(e) Phase-shifts for converters M and N of VSC1 (ΦM and ΦN)
(f) Phase-shifts for converters M and N of VSC2 (ΦM and ΦN)
(g) VSC1 controller output signals in response to three-phase fault at 10.5 s
142
(h) VSC2 controller output signals in response to three-phase fault at 10.5 s
(i) Injected three-phase currents by VSC1
(j) Injected three-phase currents by VSC1

Table 5.7 shows THD of voltage waveforms during BtB-STATCOM steady-

state operation in case 1. The voltage harmonics are within acceptable limits of
IEEE-519 standard (IEEE, 1993).
Table 5.7. THD values

Common coupling point Bus 1 Bus 3
THD (%) 0.33 0.30
143
5.5.4. Discussion
Quasi multi-pulse VSC designed in Chapter 4 is adapted for BtB-STATCOM

operation. The power circuit, pulse generating circuits, and BtB-STATCOM control
scheme is verified through different simulation scenarios in PSCAD. Conventional
PI controllers are robust and yet practically applicable to voltage magnitude and real
power transfer control owing to their simplicity. THD levels of the two converters
comply with the regulations even the switching is made at low frequency. This study
can easily be extended to VSC-HVDC transmission with DC transmission line
models. With back-to-back converters, it is also possible to supply controlled real
power transfer in micro grid applications.
5.6. Summary
In this chapter, quasi multi-pulse VSC is adapted for the three-types of multi-
converter FACTS devices. Power circuit, gating pulse generation circuits, and 2-
angle control method for the quasi multi-pulse VSC, mentioned in Chapter 4, are also
verified for different control actions together with different disturbance scenarios in
one sense. If no further performance is required in terms of power quality, eight six-
pulse converters operating together for VSC configuration can be used without the
need for any AC filters, since measured THD levels always lie below IEEE-519
standard. Independent control of voltage magnitude and phase angle of the
converters without high frequency PWM methods makes use of separate control
functions for real and reactive power possible. Although the concerned FACTS
devices have many possible operating modes, it is anticipated that the shunt
converter is operated in automatic voltage-control mode (GUPFC, BtB-STATCOM)
and the series converter (GUPFC, IPFC) is operated in automatic power-flow control
mode. The bottleneck of finding a lot of suitable parameters for many controllers that
should operate together in stable and robust is overcome by simultaneously tuning of
these parameters using simplex method. This solution is practically applied for the
controllers of GUPFC and IPFC when different cost functions are defined. A novel
144
HFPI controller for IPFC is designed to decouple the control loops of real and
reactive power flows that can be generalized to any series converter of the multi-
converter FACTS device. The simulation results prove superior dynamic
performance when compared with the simplex optimized PI controller both
without/with analytically computed feed-forward decoupling gains.
145
146
6. TRANSIENT STABILITY STUDIES A. Mete VURAL
6. TRANSIENT STABILITY STUDIES
6.1. Introduction
Transient stability is “the ability of the power system to maintain

synchronism when subjected to a severe transient disturbance such as faults on
transmission facilities, loss of generation, or loss of a large load” (Kundur, 1994).
Conventionally, electromechanical oscillations, either being local mode (1-2.5 Hz) or
inter-area mode (0.1-0.8 Hz), occur due to the natural dynamic behavior of
synchronous machines when the system is subjected to faults and may lead to total or
partial power interruption if not damped out effectively. On the other hand,
integration of large wind farms with electrical network is inevitable due to growing
electrical demand and environmental reasons which leads to interconnected operation
of synchronous generators and wind turbine driven generators (Chen et al., 2009),
(Kaldellis et al., 2011).
The use of induction machines in wind generation is widely accepted as a
generator of choice due to their simple structure and cost (Junyent-Ferre et al., 2010),
(Li et al., 2011). On one hand, wind farms employing induction generators consume
reactive power which produces low voltage profile and dynamic instability observed
following after faults (Chompoo-inwai et al., 2005), (Sentil-Kumar et al., 2011). On
the other hand, the integration not only requires stable operation of synchronous
generators, but also induction generators should operate stable without disturbing
demand side. This situation comes into prominence since the induction generator’s
behavior during a fault is very different from that of a synchronous generator. Rotor
speed instability of the induction generator is seen when it is subjected to a nearby
fault. In this situation, the rotor may accelerate and reach higher steady-state speed
(Samuelsson et al., 2005).
Self-excited double cage induction generators (SEDCIG) which energize the
wind farm is considered due to following two reasons (Li et al., 2011): (i) Although,
doubly fed induction generator (DFIG) has gained remarkable attention currently,
SEDCIGs are still operated in many grid-connected wind turbines, (ii) the transient
147
behavior of DFIG is similar to SEDCIG when the crowbar system of the DFIG
protects the converter under grid fault by bypassing the rotor circuit over the crowbar
impedance.
Combining two completely different machine stability concerns mentioned
above, stability of the power systems connected with the wind farm is enhanced by
GUPFC, with the following simultaneous control tasks which are proposed for the
first time: i) oscillation damping of wind farm integrated power system by a self-
tuning fuzzy damping controller (STFDC), ii) multi-line real and reactive power flow
control, iii) AC bus voltage control.
STFDC proposed in this chapter is further adapted for IPFC for damping
inter-area mode of oscillations of a power system consisting of several conventional
synchronous generators. Finally, the performances of the PI controllers proposed for
BtB-STATCOM in the last chapter is investigated in terms of oscillation damping
and hence to improve transient stability of the power systems in this chapter.
6.2. Literature Survey on Transient Stability Studies
A literature survey is carried out here for transient stability studies of

GUPFC, IPFC, and BtB-STATCOM. In spite of intensive research work on UPFC
related transient stability studies, very few papers are available about GUPFC. An
average model of GUPFC is integrated into Phillips-Heffron model of a multi-
machine system to investigate the impact of GUPFC on power system oscillation
damping (Lubis et al., 2012). Another GUPFC’s average model is embedded into
user-defined model of PSASP software to investigate the effect of GUPFC on
stability of China’s Sichuan Power Grid (Sun et al., 2003).
IPFC based transient stability studies generally uses IPFC’s average model
included in Phillips-Heffron model of single-machine infinite-bus (SMIB) system
rather than using converter-level model (Kazemi et al., 2006), (Parimi et al., 2008),
(Banaei et al., 2009a), (Veeramalla et al., 2010), (Parimi et al., 2010a), (Belwanshi et
al., 2011). In average model based approaches, some IPFC studies are extended for
multi-machine systems (Parimi et al., 2010b), (Parimi et al., 2011), (Shan et al.,
148
2011). On the other hand, IPFC is fuzzy and/or neural network controlled to improve
stability based on average modeling approach (Mishra et al., 2002), (Parimi et al.,
2010a), (Belwanshi et al., 2011), (Banaei et al., 2011). Different from those, transient
stability of a multi-machine system is improved by IPFC using derived energy
function for IPFC which is developed from an average model (Azbe et al., 2009).
Three-phase model of IPFC is derived using switching functions for a twelve-pulse
VSC based IPFC for SSR studies (Padiyar et al., 2007).
A non-linear control scheme for BtB-STATCOM is developed using an
average modeling approach (Lee et al., 2011). The stability studies are conducted on
Phillips-Heffron model of SMIB system which includes the average model of two
BtB VSCs (Banaei et al., 2009b). A supplementary control for VSC based BtB link
in damping SSR of series capacitive compensated transmission system is studied
using an average model of each VSC (Faried et al., 2009).
According to literature review results, transient stability studies of GUPFC,
IPFC, and BtB-STATCOM generally rely on average models included into Phillips-
Heffron SMIB system or multi-machine systems, rather than using converter-level
models. This chapter is aimed to investigate these FACTS devices on transient
stability enhancement using converter-level modeling approach, together with a
novel damping control scheme.
6.3. Transient Stability Improvement Using GUPFC
6.3.1. Dynamic Equations for Power Generation
6.3.1.1. Wind Model
The wind can be modeled with the following equation that properly includes
spatial effects of the wind behavior such as gusting, ramp changes, and background
noise (Anderson et al., 1983),
VW = VWB + VWG + VWR + VWN (6.1)
149
where VW is the wind speed, VWB is the base or mean wind speed which is always
assumed to be present where the wind generator is required to be in service. VWG is
the gust wind component, VWR is the ramp wind component, and VWN is the noise
wind component. In this chapter, only transient fault simulations are considered
where the simulated events last up to only 12.5 s. Moreover, the wind farm
considered here is aggregations of many single wind turbines in which wind speed
variations can cancel each other (Sorensen et al., 2002), (Jauch et al., 2007), (Erlich
et al., 2007). That’s why natural wind variations (VWG,VWR,VWN) are not taken into
account. VWB is set to 14 m/s allowing all turbines to produce rated power (Jauch et
al., 2007), (Kusiak et al., 2010).
6.3.1.2. Blade Dynamics
The mechanical system mainly consists of blade and shaft which transforms
wind kinetic energy into rotational motion. Shaft dynamics are not presented in this
research which is characterized by blade speed, hub speed, gear box speed, and the
generator mechanical speed (Anderson et al., 1983). The available wind power is
assumed to be captured by horizontal axis wind turbine with three blades. The blade
dynamics are represented by the following functions (Anderson et al., 1983),
VW
γ =
wB
1
Cp = (γ − 0.022 β p 2 − 5.6)e −0.17γ
2
1
PW = ρAC p VW 3 (6.2)
2
where wB is the blade angular velocity, γ is the tip speed ratio, βp is the blade pitch
angle, Cp is the dimensionless power coefficient, ρ is air density, and A is blade
impact area. PW is the resultant mechanical power which is extracted from the wind.
150
6.3.1.3. Self-excited Double Cage Induction Generator
Self-excited double cage induction generator (SEDCIG) can be modeled by

the following equations for its one phase while saturation effects are ignored (Levi,
1997). Underlined variables denote space vectors in the arbitrary rotating reference
frame with a speed of wa.
v s = R s i s + d ϕ s / dt + jwa ϕ s
0 = R r1 i r1 + d ϕ r1 / dt + j ( w a − w)ϕ r1 + Rc (i r1 + i r 2 )
0 = R r 2 i r 2 + d ϕ r 2 / dt + j ( wa − w)ϕ r 2 + Rc (i r1 + i r 2 )
ϕ s = L s i s + Lm (ir1 + ir 2)
ϕ r1 = Lr1 i r1 + Lm i s + L12 i r 2
ϕ r 2 = Lr 2 i r 2 + Lm i s + L12 ir1
TE = (3 / 2) Pϕ s × is
TE − TL = ( J / P)dw / dt (6.3)
where v, i, φ respectively describes voltage, current, and flux linkage. Subscripts r

and s stands for rotor and stator, respectively. Subscripts 1 and 2 represent the rotor
winding numbers, respectively. R describes resistance and RC is common end-ring
resistance between the two cages in the SEDCIG. L represents inductance and Lm is
the mutual leakage inductance between stator and the two rotor windings. L12 is the
mutual leakage inductance between the two rotor windings, w is the rotation speed
and P is the number of pole pairs. TE and TL stand for electrical torque and load
torque, respectively. J is the inertia of the machine.
6.3.1.4. Salient-Pole Synchronous Generator
The unsaturated dq model of the salient-pole synchronous generator (SG) can

be approximated by the following functions (Teng et al., 2010).
151
Stator equations:
Vq = − Ra iq + e ′q′ − X d′′ i d
Vd = − Ra i d + e d′′ − X q′′i q
de q′′
Td′′0 = e ′q − e q′′ − ( X d′ − X d′′ )i d
dt
de ′q X d − X d′
Td′ 0 = V f − e q′ − (e ′q − e ′q′ )
dt X d′ − X d′′
de d′′
Tq′′0 = e ′d − e ′d′ − ( X q′ − X q′′ )iq
dt
de d′ X q − X q′
Tq′0 = −e d′ − (e d′ − e d′′ )
dt X q′ − X q′′
T ′ = e ′d′ i d + e ′q′ i q + ( X q′′ + X d′′ )i d i q (6.4)
Rotor equation:
d 2δ dδ
M +D = TM − TE (6.5)
dt 2 dt
V, i, X respectively describes voltage, current, and reactance. Subscripts d and

q stand for direct and quadrature axis, respectively. Subscript 0 means open circuit
and Ra describes armature resistance. T defines time constant. Vf and e is the
excitation winding voltage and internal generated voltage, respectively. Superscripts
' and '' respectively denote transient and sub-transient modes. M and D describe
inertia constant and damping coefficient, respectively. TE is electrical torque which is
opposing the mechanical torque, TM. δ is the torque angle of the machine.
6.3.2. Power System Configuration
Time domain simulation studies are carried out on wind farm integrated
power system installed with GUPFC, which is shown in Figure 6.1. The system is
kept as simple as possible and grid data are inspired from IEEE first benchmark
model (IEEE, 1997). Series converters VSC2 and VSC3 are inserted into Lines 2 and
1, respectively. Shunt converter (VSC1) is connected to Bus 1.
152
Figure 6.1. Power system configuration embedded with GUPFC
153
The wind farm is rated at 50 MVA and comprised of 20 SEDCIGs operating

coherently, each driven by a horizontal axis three-blade wind turbine. 320 μF
capacitor for each phase is installed at wind farm bus for unity power factor
operation. AC grid power generation side is the aggregated model of 5 parallel
hydro-governed SGs with solid-state exciters, rated at 120 MVA each. 100 MVA,
154 kV, and 60 Hz are chosen as base values and start-up transients of the generators
are not taken into account since the faults are considered soon after the system comes
to steady-state. Wind farm integrated power system, GUPFC with three quasi multi-
pulse VSCs, and control blocks are modeled in PSCAD while fuzzy interfaces are
designed in MATLAB fuzzy logic toolbox. PSCAD is interfaced with MATLAB
through a custom written interface in PSCAD that exchanges data with MATLAB
continuously at every solution time step of 100 µs, shown in Figure 6.2.
Each VSC of GUPFC is rated at 100 MVA. Each single-phase three-winding
and magnetic interface is rated at 60 Hz, 16.67 MVA, 137.92/38 kV, j0.1 pu for
shunt VSC. For series VSC, the rating is modified as 23.42/23.42 kV, j0.1 pu. Each
33.33 MVA, 104.13/40 kV, j0.01 pu.
154
6.3.3. Damping Control Scheme of GUPFC
The self-tuning scheme for PI type fuzzy controllers is originally proposed by

the authors (Mudi et al., 1999) and later applied to TCSC to improve the stability of
multi-machine power systems (Hameed et al., 2008). STFDC is designed with
altered fuzzy inputs for GUPFC. The performance is further improved by optimizing
scaling gains using simplex optimization method. Damping control scheme of the
GUPFC is based on the modification of real power flow control loop shown in
Figures 5.3c or 5.3e. All other control loops of GUPFC in Figure 5.3 are applied
without any alteration in this section. Real power flow controllers shown in Figures
5.3c or 5.3e can be expressed alternatively as in equation (6.6) where s is Laplace
operator, Kp and Ti are proportional gain and integral time constant of the PI
controller, respectively. ΔP=(Pref-P) stands for real power flow error and VseiQ is the
q-axis voltage component of series converter-i of GUPFC in which damping control
function is added.
VseiQ = ( K p + 1 / Ti s )∆P (6.6)
In case of damping mode, error at sample-k is simply modified as in equation (6.7)

by an auxiliary damping signal based on the speed error of SG (Δw=wref - w),
e(k ) = ∆P + K w ∆w (6.7)
where Kw is the damping gain. Since aggregated synchronous machine model is used,
w and wref represents the speed at sample-k, and base speed of all parallel operating
generators, respectively. As opposed to one of the originally proposed fuzzy inputs in
the paper (Mudi et al., 1999), control system is made insensitive to noise in the error
measurement using error-integral instead of error-derivative which lessen control
signal oscillations highly observed in simulation cases. In this case, the error-integral
at sample-k can be computed as
155
Σe(k ) = Σe(k − 1) + e(k ) (6.8)
It is important to note that the series converter of the GUPFC, where STFDC is not
utilized, controls real power flow on Line-1 using equation (6.6) with auxiliary
damping signal, as shown in Figure 6.1. STFDC is the alternative of this control
mode and constructed from a fuzzy damping controller (FDC) and a fuzzified gain
tuner (FGT), as shown in Figure 6.1.
6.3.3.1. Fuzzy Damping Controller (FDC)
In FDC scheme, the signals e and Σe in equations (6.7) and (6.8) are
respectively multiplied by gains (a1, a2), which needs to be optimized. Crisp values
are then mapped to their equivalent fuzzy values by the membership functions of
knowledge base in Figure 6.3. Membership functions for eO and ΣeO are symmetrical
triangles (except the two at both ends) which have equal 50% base overlap, divides
the domain [-1,1] into 7 equal regions. The term sets of eO and ΣeO contain the same
linguistic expressions for the magnitude part of the linguistic values and
characterizes rule matrix-1 in Figure 6.3, which contains 49 rules. The cell defined
by the intersection of the first row and the first column represents a rule such as, {“If
ΣeO is P1 and eO is N2 then ΔVq is N1”}. The antecedents are evaluated by applying
“min” operator and the output fuzzy set is truncated by applying “min” implication
operator. The fuzzy sets are aggregated into a single fuzzy set by “max” operator that
should be later dezuffied to resolve a single real number for each output variable.
Centroid defuzzification method is applied to get incremental change in series
converter voltage as in equation (6.9):
∑i49=1 bi ∫ µ (i)
∆Vq (k ) = 49
(6.9)
∑ ∫ µ (i )
i =1
156
where µ(i) and bi are the output membership function and the center of output
membership function of the consequent of rule i, respectively. Finally, at sample-k,
q-axis component of the series converter voltage for oscillation damping (as well as
for dynamic real power flow control) is calculated in equation (6.10) where β is the
gain factor at sample-k which is decided by FGT.
Vq (k ) = Vq (k − 1) + a 3 β ∆Vq (k ) (6.10)
Figure 6.3. Membership functions and fuzzy rules for STFDC
6.3.3.2. Fuzzified Gain Tuner (FGT)
In STFDC scheme, FDC performance is further enhanced by FGT which

computes the gain factor β by a self-tuning mechanism independent from FDC itself.
So, two fuzzy modules (FDC and FGT) operate concurrently to generate Vq signal
reference for the series converter. The value of β is not fixed and modified at each
sample-k according to the following relation,
β (k ) = f (eo (k ), Σeo (k ) ) (6.11)
157
where f denotes a non-linear mapping function, described by rule matrix-2 with 49

rules in Figure 6.2 and associated by the FGT scheme whose structure is exactly the
same for its fuzzy operators and input membership functions with those of FDC.
Universe of discourse for β lies in the domain [0,1] and is obtained by shifting and
scaling (add 1 and multiply with 0.5) input membership functions of FDC along the
horizontal axis. Rule matrix-2 is designed to improve the damping performance of
GUPFC under large disturbances such as three-phase short circuit on transmission
lines. For instance, after a fault occurs, error may be small-positive (P1) but error-
integral can be sufficiently large (P3). In this case, β should be big enough (VB) to
increase converter voltage. Under such a situation, the rule is {“If ΣeO is P3 and eO is
P1 then β is VB”}. The control surfaces of the proposed STFDC are shown in Figure
6.4.
(a) FDC (b) FGT

Figure 6.4. Control surfaces of the proposed STFDC
6.3.3.3. Tuning of Scaling Factors
The scaling factors (a1, a2,) are used to normalize input variables of the FDC.
{eO=a1e; ΣeO=a2Σe}. Similarly, FDC output variable (ΔVq) is first multiplied by a3
then tuned by FGT adaptively. Commonly, there is no well-defined method for
selection of scaling factors (Mudi et al., 1999). In this research, these parameters are
optimized by simplex method. The cost function is based on integral time absolute
error (ITAE) and given in equation (6.12) where t is the simulation time, t0 is the
fault time. T is the total simulation time for case 1 in the next section. PSCAD
158
configuration for this task is shown in Figure 6.5. The convergence performance of
cost function in simplex method is shown in Figure 6.6. The value of f is minimized
from 0.0720 to 0.0114 in 51 iterations for a tolerance of 1.0E-6 when only FDC is
executed while FGT is deactivated. The optimized parameters are listed in Table 6.1.
( )
T
f (a1 , a 2 , a 3 ) = ∫ t ⋅ wref − w ⋅ dt (6.12)
t =t0
Figure 6.6. Convergence performance of cost function in simplex method
159
Table 6.1. Optimization results of scaling factors

scaling factors a1 a2 a3
initial guess 0.001 0.001 0.001
converged result 0.7369 0.3253 0.8221
The stability concern is first evaluated for three-phase and single-phase faults
without GUPFC in the power network and then with GUPFC. The dynamic
simulations investigate the impact of faults (i) on the stability of
synchronous/induction generators, (ii) on GUPFC performance when controlling real
and reactive power flows as well as AC bus voltage. PSCAD and MATLAB are used
simultaneously for simulating transient behavior of the models. The parameters of PI
controllers, shown in Figure 6.1, are given in Appendix C. The capacitance of DC
link is C=0.2 F. The performance of STFDC is examined for different disturbance
conditions which lead to local mode of oscillations in conjunction with the following
dynamic control tasks of the GUPFC:
• Line-2 real power flow (PL2) using either FDC or STFDC by VSC2
• Line-2 reactive power flow (QL2) by VSC2
• Line-1 real power flow (PL1) by VSC3
• Line-1 reactive power flow (QL1) by VSC3
• Bus 1 voltage (V1) by VSC1
• DC link voltage (E) by VSC1
A three-phase to ground fault of 120 ms duration is applied to Line-3 near

Bus 1 at 8.5s. Pre-disturbance operating conditions are; PL1ref=1.75 pu, PL2ref=0.5 pu,
QL1ref=QL2ref=0.0 pu, Eref=2.0 kV, and V1ref=1.0 pu. To have a quantitative
comparison, the ITAE values between 8.5 s and 12.5 s are calculated for different
160
control schemes. Although, STFDC is only activated for VSC2 and Line-2 reactive
power flow and Line-1 real and reactive power flows are controlled by simple PI
controllers, it is found that STFDC indirectly smoothens the variations of simulated
waveforms against fault and shows better performance than FDC in general. As
evidence by response curves depicted in Figure 6.7a, STFDC performance is superior
to FDC on SEDCIG rotor speed damping, being also better than that of PI controller.
In Figure 6.7b, STFDC responses better than FDC and PI controller in damping SG
oscillations with reduced undershoot/overshoot and less settling time. As the
consequence of the fault, real and reactive power flow variations of Line-1, presented
in Figures 6.7c and 6.7d, and those of Line-2 presented in Figures 6.7e and 6.7f are
minimized better by STFDC with less undershoot/overshoot compared with the FDC.
DC link voltage excursions of GUPFC for different damping control schemes are
depicted in Figure 6.7g and it is found that among the two control schemes, the ITAE
index is smaller for STFDC. In Figure 6.7h, PI controller settles Bus 1 voltage to its
controlled value of 1.0 pu with a smaller ITAE value in case of STFDC. Effect of
employing STFDC with optimized gains improves transient responses of both
SEDCIG speed and IG speed. This situation is illustrated on Figures 6.7i and 6.7j,
respectively. Voltage and current signals of the quasi multi-pulse converters after the
fault are presented in Figure 6.8. In more detail, simulated phase shift angles of
converters M and N and one selected GTO voltage are shown in Figure 6.9.
(a) Transient response of SEDCIG speed under different control modes
161
(b) Transient response of SG speed under different control modes
(c) Variation of Line-1 real power flow under different control modes
(d) Variation of Line-1 reactive power flow under different control modes
162
(e) Variation of Line-2 real power flow under different control modes
(f) Variation of Line-2 reactive power flow under different control modes
(g) DC link voltage excursions of GUPFC under different control modes
163
(h) Variation of Bus 1 voltage under different control modes
(i) Effect of optimized gains on transient response of SEDCIG speed
(j) Effect of optimized gains on transient response of SG speed

Figure 6.7. Simulated STFDC performance against three-phase fault
164
Figure 6.8. Simulated voltage and current waveforms of GUPFC converters
Figure 6.9. Simulated phase shift angles (ɸM and ɸN) and one GTO voltage
6.3.4.2. Case 2: Three-phase Fault with Longer Duration
The same fault is applied as in case 1 to Line-3 with an increased duration of

160 ms. Pre-disturbance operating conditions are changed as; PL1ref=0.8 pu,
PL2ref=1.5 pu, QL1ref=QL2ref=0.0 pu, Eref=2.0 kV, and V1ref=1.0 pu. As shown in Figure
6.10a, longer fault results in a speed increase of SEDCIGs without GUPFC (steady-
state speed is 1.483 pu which is not shown), making wind farm unstable in its
165
operation (Samuelsson et al., 2005). STFDC damps speed fluctuation of SEDCIG

slightly better than FDC or PI controller with less overshoot characteristics. On the
other hand, Figure 6.10b shows that STFDC exhibits good damping response to SG
speed oscillations as compared with the FDC or PI controller with less settling time
and less undershoot characteristics. It is observed that for both case1 and case2,
SEDCIGs and SGs get stabilized and regain their original speed after fault clearance.
The variations of real and reactive power flows of Line-1 shown in Figures 6.10c and
6.10d and the variation of reactive power flow of Line-2 in Figure 6.10f are
indirectly improved by appointing STFDC generally with better control
characteristics than FDC. When a comparison between Figures 6.10c and 6.10e is
made particularly, STFDC holds better real power flow of Line-2 in its reference
value than that of Line-1. Since STFDC commands the series converter which is
inserted into Line-2, while only PI controller is activated for the series converter
inserted into Line-1. DC voltage excursions of GUPFC depicted in Figure 6.10g are
practically the same in case of both FDC and STFDC. Figure 6.10h shows that the
Bus 1 voltage settles down to 0.7 pu and comes to 1.0 pu steadily, practically the
same response for both FDC and STFDC, but with improved ITAE index in case of
STFDC. Real and reactive power fluctuations of the wind farm and AC grid under
two control modes are shown in Figure 6.11. STFDC mitigates these fluctuations
effectively which overcomes the instability of PI controller. In particular, Figure
6.11b shows that without STFDC, the reactive power demand of the wind farm is
very high due to the fault, which reduces substantially once the STFDC is activated
instead of PI controller.
166
(a) Transient response of SEDCIG speed under different control modes
(b) Transient response of SG speed under different control modes
167
168
(g) DC voltage excursions of GUPFC under different control modes
(h) Variation of Bus 1 voltage under different control modes

Figure 6.10. Simulated STFDC performance against longer three-phase fault
(a) Real power output of the wind farm following three-phase fault
169
(b) Reactive power output of the wind farm following three-phase fault
(c) Real power output of the SGs following three-phase fault
(d) Reactive power output of the SGs following three-phase fault

Figure 6.11. Power fluctuations following three-phase fault
170
The system is subjected to phase-A to ground fault on Line-3 near Bus 1 for a
duration of 265 ms at 8.5 s. Pre-disturbance operating conditions are changed as;
PL1ref=1.0 pu PL2ref=0.75 pu, QL1ref=QL2ref=0.0 pu, Eref=2.0 kV, V1ref=1.0 pu. Figures
6.12a and 6.12b shows the transient fluctuations of the SEDCIG speed and SG speed,
respectively and provide a comparison between different control schemes. Besides,
rise time and settling time of FDC and STFDC is practically the same for both
generators, STFDC gives lower undershoot in case of SEDCIG and quantitative
comparison shows better STFDC results for oscillation damping of SG speed. The
waveforms in Figures 6.12c-f indicate that STFDC is again found to be superior to
FDC in general when controlling real and reactive power flows of the lines after the
fault both with reduced overshoot/undershoot characteristics and with smaller ITAE
indices. Although undershoot in case of STFDC exceeds the undershoot in case of
FDC by approximately 4.5% in Figure 6.12e, the steady-state error is more
effectively minimized by STFDC and a minimum ITAE index is reached. DC
voltage regulation of the GUPFC is satisfactory in Figure 6.12g and STFDC reduces
DC voltage fluctuations significantly better than FDC. Figure 6.12h shows Bus 1
voltage variations following single-phase to ground fault. The AC voltage controller
is again satisfactory like in previous case studies and gives practically the same
response in case of FDC and STFDC with a better ITAE index than that of FDC.
(a) Transient response of SEDCIG speed without GUPFC and with GUPFC
171
(b) Transient response of SG speed without GUPFC and with GUPFC
172
(g) DC voltage excursions of GUPFC under different control modes
173
(h) Variation of Bus 1 voltage following single-phase fault

Figure 6.12. Simulated STFDC performance against single-phase to ground fault
Table 6.2 summarizes voltage distortions at Buses 1 and 2 as a measure of

THD. Records of the simulated cases taken at 12.5 s show that THD values are
within acceptable limits when STFDC is activated (IEEE, 1993). Consequently,
filtering is not required even GTOs are switching at fundamental system frequency.
Table 6.2. THD values of power system bus voltages

THD THD THD THD THD THD
Case 1
Case 2
Case 3
V1(L-L) V2(L-L) V1(L-L) V2(L-L) V1(L-L V2(L-L)

1.26 % 1.25 % 1.53 % 1.29 % 1.48 % 1.27 %
6.3.5. Discussion
The newly proposed damping controller is robust to change in fault type and
fault duration as well as robust to changing operating conditions of the power
system. Better damping characteristics for local mode of oscillations of SG are
achieved by GUPFC equipped with STFDC. Furthermore, STFDC can control
SEDCIG speed better than FDC in case of a fault although SEDCIG speed signal is
174
not measured in the proposed damping control scheme. The successful operation of
the shunt and series converters of the GUPFC is proven by maintaining constant DC
link voltage and after faults GUPFC shows stable operation and able to restore real
and reactive power flows of the transmission lines to their regulated values with
significantly less variations in case of STFDC. This situation can claim longer
transient fault duration that the system can withstand. It is also noted that shunt
reactive power support of GUPFC improves voltage profile of the wind farm bus
during transient conditions.
6.4. Transient Stability Improvement Using IPFC
It is demonstrated in the previous section that, STFDC exhibits superior

dynamic performance than classical PI controller with supplementary damping signal
or a fixed fuzzy damping scheme in damping local mode of oscillations of SG and
improves SEDCIG speed stability. In this section, STFDC, originally proposed for
GUPFC, is adapted for IPFC to damp out inter-area mode of oscillations in a multi-
machine power system having many SGs, spread out two remote areas.
Two-Area System shown in Figure 6.13 is used to illustrate an interconnected

system of two remote areas without an infinite bus. Both areas are represented by
aggregate machines which are connected together via double transmission line
intertie. Following a large disturbance, such as short circuit, the system exhibits
inter-area mode of oscillations where the two machines in each area act as single unit
and swing coherently against the network at the other end of the line. Series
converters VSC1 and VSC2 of the IPFC are positioned on Line-1 and Line-2,
respectively by means of series coupling interfaces, tr1 and tr2. Using switches, sw1,
sw2, and sw3, IPFC can also be operated as SSSC, required for the simulation cases.
10, 5, and 8 aggregated SGs, rated 120 MVA each, are operated in parallel to
produce 1200 MVA (G1), 600 MVA (G2), and 960 MVA (G3) output, respectively.
175
Figure 6.13. Two-Area System embedded with IPFC and its control scheme
176
All generators are driven by hydro-governors with solid-state exciters.

Transmission line data of the system are given in Appendix B. 20 μF capacitor for
each phase is installed at Bus 1 to boost bus voltage in steady-state operation. 100
MVA and 154 kV are chosen as base values and the start-up transients of the
generators are not taken into account since the faults are considered soon after the
system comes to steady-state. Two-Area System and IPFC having two quasi multi-
pulse VSCs, and control blocks are modeled in PSCAD while fuzzy interfaces are
designed in MATLAB fuzzy logic toolbox. PSCAD is interfaced with MATLAB
through the custom written interface in PSCAD that exchanges data with MATLAB
continuously at every solution time step of 100 µs, shown in Figure 6.14. On the
contrary of GUPFC based transient stability studies in previous section, error
derivative is adapted, instead of error integral as one of the inputs for STFDC since
this configuration gives better results.
Each VSC of IPFC is rated at 100 MVA. Each single-phase three-winding
and magnetic interface is rated at 60 Hz, 16.67 MVA, 23.42/23.42 kV, j0.1 pu. Each
33.33 MVA, 23.42/9.0 kV, j0.01 pu.
177
6.4.2. Tuning of Scaling Factors
The scaling factors (a1, a2, and a3) in STFDC configuration for GUPFC
should be re-optimized using simplex method to normalize input and output variables
of the STFDC, since a different type of multi-converter FACTS device is embedded
into a different power system. The cost function is based on ITAEs of different
measured generator speeds given in equation (6.13) where w1, w2, and w3 are the
speeds of G1, G2, and G3 in Figure 6.13, respectively. For IPFC, the value of f is
minimized from 0.2078 to 0.0399 in 97 iterations for a tolerance of 1.0E-6 as shown
in Figure 6.15a. Similarly, for SSSC, the value of f is minimized from 0.2452 to
0.2221 in 60 iterations for a tolerance of 1.0E-6 as shown in Figure 6.15b. Simplex
method is run when only FDC is executed while FGT is deactivated for both FACTS
devices. The optimized parameters are listed in Table 6.3.
f (a1 , a2 , a3 ) = ∫ (t ⋅ w1 − w2 + t ⋅ w1 − w3 )⋅ dt
T
(6.13)
t =t0
The stability of the Two-Area System is investigated without and with IPFC
having STFDC by applying different types of faults with different durations.
Moreover the damping feature of IPFC is compared with that of SSSC for all cases
under the same control scheme. The impact of faults is also investigated on the
performances of control loops of IPFC which is shown in Figure 6.13. PSCAD
having a solution time step of 100 μs and MATLAB are communicated on-line for
simulating transient behavior of the models. The chosen parameters of the PI
controllers for the IPFC are given in Appendix C. The capacitance of DC link is
C=0.2 F. Steady-state uncontrolled real power flows of the intertie are 0.975 pu for
each transmission line. IPFC is activated for both Lines 1-2 when switch sw1 and
sw3 are opened and sw2 is closed. SSSC is activated on Line-2 when switch sw2 and
sw3 are opened and sw1 is closed. The performance of STFDC for both IPFC and
178
SSSC is examined individually for the same disturbance conditions applied to Two-
Area System which lead to inter-area mode of oscillations in conjunction with the
following dynamic control tasks of the IPFC and SSSC:
• Line-1 real power flow by VSC1 of IPFC

• Line-2 real power flow by VSC2 of IPFC
• DC link voltage by VSC2 of IPFC
• Line-2 real power flow by VSC2 (or SSSC)
• DC link voltage by VSC2 (or SSSC)
(a) for IPFC
(b) for SSSC

Figure 6.15. Cost function minimization for both FACTS devices
Table 6.3. Optimization results of scaling factors

scaling factors a1 a2 a3 a1 a2 a3
IPFC
SSSC
initial guess 0.1 0.1 0.1 0.1 0.1 0.1

converged result 0.6 0.6 3.67 0.75 0.45 4.60
179
Before applying disturbance, the reference values of tie-line flows, PLine-1 and
PLine-2 are respectively set to 1.1 pu and 1.2 pu at the real power flow controllers of
IPFC while the DC link voltage is regulated at 1.4 kV. The same reference value of
PLine-2 is set for SSSC’s real power flow controller. Then a three-phase to ground
fault near Bus 1 on Line-1 with 140 ms duration is applied at t=2.0 s. As shown in
Figures 6.16a and 6.16b, the angle oscillations of generators G2 and G3 with respect
to generator G1 are cumulative and lead to unstable operation when no FACTS
device is activated. SSSC having only VSC2 exhibits weakly damped inter-area
modes at approximately 0.50 Hz for both G2 and G3 with respect to G1. On the other
hand, IPFC, having both VSC1 and VSC2, effectively damps out the oscillations
caused by this severe disturbance in relatively short duration. Comparing the
responses of IPFC to the SSSC compensation scheme in Figures 6.16c and 6.16d, the
positive contribution of the proposed STFDC adapted for IPFC is clear when
controlling intertie real power flows caused by inter-area oscillations. Figure 6.16e
shows that the time responses of the DC link voltage of both SSSC and IPFC are
practically the same which is highly required for proper VSC operation. Figure 6.16f
shows reactive power flow fluctuations on Line-1 caused by three-phase disturbance
when reactive power flow control function of IPFC is disabled to make a fair
comparison to SSSC. Figures 6.16g and 6.16h show that STFDC equipped IPFC
better improves bus voltage profiles of the intertie with smoother responses
following three-phase fault when compared with STFDC equipped SSSC. Figures
6.17 and 6.18 shows some selected time domain signals of the two VSCs of IPFC
which reveal stable converter operation.
6.4.3.2. Case 2: Two-phase to Ground Fault
The system is disturbed by a two-phase (phases B and C) to ground fault near

Bus 1 on Line-1 for 160 ms duration at t=2.0 s, while keeping the same pre-
disturbance steady-state operating conditions as in case 1. The system is unstable
180
when there is no compensation is applied. Figures 6.19a and 6.19b show the
responses of the generators G2 and G3 with respect to generator G1 when SSSC with
STFDC are applied and when IPFC with STFDC are applied. The comparative time-
domain results show that the stabilizing function of IPFC for inter-area oscillations is
superior to those of SSSC even STFDC is adapted individually to both FACTS
devices by optimizing its scaling factors. IPFC with STFDC easily stops the real
power oscillations both on Lines 1 and 2 and forces them to their steady-state
controlled values as shown in Figures 6.19c and 6.19d. When a particular
comparison between Figures 6.16c and 6.19c is made, SSSC weakly suppresses
power oscillation in case of two-phase to ground fault due to longer duration of fault.
DC link voltage controllers of both SSSC and that of IPFC gives practically the same
response to the short circuit as shown in Figure 6.19e. Figure 6.19f shows reactive
power flow fluctuations on Line-1 when IPFC and SSSC are operated separately
when reactive power flow control function of IPFC is disabled. Accordingly, as in
case 1 the fluctuations are less as in case of IPFC when compared with SSSC.
Figures 6.19g and 6.19h show that STFDC equipped IPFC better improves bus
voltage profiles of the intertie with smoother responses following two-phase fault
when compared with STFDC equipped SSSC.
(a) Generator G2 rotor angle measured with respect to generator G1 rotor angle
181
(b) Generator G3 rotor angle measured with respect to generator G1 rotor angle
(c) Variation of Line-2 real power flow following three-phase fault
(d) Variation of Line-1 real power flow following three-phase fault
182
(e) DC link voltage excursions of two FACTS devices following three-phase fault
(f) Variation of Line-1 reactive power flow following three-phase fault
(g) Variation of Bus 1 voltage following three-phase fault
183
(h) Variation of Bus 2 voltage following three-phase fault

Figure 6.16. Simulated STFDC performance following three-phase fault
Figure 6.17. Simulated voltage and current waveforms of IPFC converters
184
(c) Variation of Line-2 real power flow following two-phase fault
185
(d) Variation of Line-1 real power flow following two-phase fault
(e) DC link voltage excursions of two FACTS devices following two-phase fault
(f) Variation of Line-1 reactive power flow following two-phase fault
186
(g) Variation of Bus 1 voltage following two-phase fault
(h) Variation of Bus 2 voltage following two-phase fault

Figure 6.19. Simulated STFDC performance against two-phase fault
The system is disturbed by a single-phase (phase C) to ground fault near Bus

1 on Line-1 for 200 ms duration at t=2.0 s, while keeping the same pre-disturbance
steady-state operating conditions as in case 1. This relatively longer fault makes the
multi-machine system operation unstable as large cumulative oscillations are
observed both in time responses of generators’ relative angles and real power flows
of intertie without any compensation scheme. In detail, Figures 6.20a and 6.20b
187
shows that IPFC with STFDC robustly stabilizes the inter-area mode of oscillations
while SSSC with STFDC shows a poor suppressing function. Figures 6.20c and
6.20d show that IPFC endowed with the proposed STFDC eliminates the oscillations
of the real power transmission of Line-2, between the two areas, and resumes the real
power transmission to its controlled level before the fault. Figure 6.20e indicates that
the DC link voltage controllers of both SSSC and that of IPFC gives practically the
same response to the short circuit as in previous fault cases. Figure 6.20f shows
reactive power flow fluctuations on Line-1 when IPFC and SSSC are operated
separately when reactive power flow control function of IPFC is disabled as in
previous fault scenarios. It is shown that the reactive power fluctuations are
practically the same for two FACTS devices. Figures 6.20g and 6.20h show that
STFDC equipped IPFC better improves bus voltage profiles of the intertie with
smoother responses following single-phase fault when compared with STFDC
equipped SSSC.
188
(c) Variation of Line-2 real power flow following single-phase fault
(d) Variation of Line-1 real power flow following single-phase fault
189
(e) DC link voltage excursions of two FACTS devices following single-phase fault
(f) Variation of Line-1 reactive power flow following single-phase fault
(g) Variation of Bus 1 voltage following single-phase fault
190
(h) Variation of Bus 2 voltage following single-phase fault

Figure 6.20. Simulated STFDC performance against single-phase fault
Table 6.4 summarizes voltage distortions of the intertie buses, namely Buses
1 and 2, as a measure of THD. Records of the simulated cases taken at 12.5 s show
that THD values are within acceptable limits when STFDC is activated in both
control loops of IPFC and SSSC (IEEE, 1993). Consequently, filtering is not
required for the two FACTS devices even GTOs are switched at fundamental system
frequency.

THD for THD for THD for THD for
Case 1 V1(L-L) V2(L-L) V1(L-L) V2(L-L)
0.29 % 0.18 % 0.20 % 0.12 %
IPFC
SSSC

0.25 % 0.15 % 0.20 % 0.12 %
0.12 % 0.10 % 0.12 % 0.08 %
191
6.4.4. Discussion
The originally proposed STFDC for one of the series converters of the
GUPFC is adapted for one of the IPFC’s converters and SSSC by changing one of
the inputs of the fuzzy interface and re-optimizing the scaling factors of the STFDC.
STFDC is robust to change in fault-type and fault duration while damping inter-area
mode of oscillations in Two-Area System. IPFC equipped with STFDC mitigates
better angle oscillations of the generators than SSSC equipped with STFDC.
Moreover, IPFC can control line real power flows of the intertie better than SSSC in
case of faults. These results show that IPFC shows superior control characteristics,
owing to the fact that IPFC has more control degrees of freedom than SSSC.
Although there is no voltage control function is included either to IPFC or SSSC
operations, both are able to make voltages of the intertie buses less oscillatory in case
of severe faults. Successful operations of the IPFC and SSSC are proven by
maintaining constant DC link voltage under fault scenarios.
6.5. Transient Stability Improvement using BtB-STATCOM
To investigate BtB-STATCOM behavior under large disturbances, the SMIB

system shown in Figure 6.21 is simulated for the study purpose when fault scenarios
in different case studies are separately applied. Shunt converters VSC1 and VSC2 of
the BtB-STATCOM are positioned at Buses 1 and 2, respectively by means of shunt
coupling interfaces, tr1 and tr2. It is aimed to provide controlled real power transfer
from 120 MVA rated SG to the infinite bus, while regulating neighboring bus
voltages by means of BtB-STATCOM. The control schemes of the two converters
are also depicted in Figure 6.21. SG is driven by hydro-governor with solid-state
exciter. Simulated system has two 50 km transmission lines having data identical to
the line positioned between Buses 1 and 4 in Section 6.4 with base values of 100
MVA, 154 kV, and 60 Hz.
192
Figure 6.21. Power system configuration embedded with BtB-STATCOM
193
Power system and BtB-STATCOM having two quasi multi-pulse VSCs, and
control blocks are simulated in PSCAD with a solution time step of 100 µs. Each
VSC of BtB-STATCOM is rated at 100 MVA. Each single-phase three-winding
and magnetic interface is rated at 60 Hz, 16.67 MVA, 137.92/38 kV, j0.1 pu.
The stability of the SMIB system is investigated without and with BtB-
STATCOM having the PI control schemes shown in Figure 6.21 by applying
different types of faults with different durations. The damping ability of BtB-
STATCOM is evaluated for all cases with this respect. The impact of faults is also
investigated on the performances of control loops of BtB-STATCOM. PSCAD
having a solution time step of 100 μs is used for simulating transient behavior of the
SMIB system embedded with BtB STATCOM. The chosen parameters of the PI
controllers for BtB-STATCOM are given in Appendix C. The capacitance of DC link
is C=0.2 F. Using switches sw1 and sw2, BtB-STATCOM can be bypassed with a
line, required for the simulation cases. For instance, when sw1 is closed while sw2 is
opened, BtB-STATCOM is bypassed by a short line. When sw1 is opened while sw2
is closed, BtB-STATCOM is in operation alternatively. The following dynamic
control tasks of the BtB-STATCOM are examined for different disturbance
conditions:
• Bus-1 voltage control by VSC1 of BtB-STATCOM

• Bus-2 voltage control by VSC2 of BtB-STATCOM
• DC link voltage by VSC2 of BtB-STATCOM
• Real power transfer control by VSC1 of BtB-STATCOM
194
6.5.2.1. Case 1: Three-phase to Ground Fault at Generator Bus
Three-phase to ground fault at 30.0 s for a duration of 120 ms is applied near

Bus 1 in Figure 6.21. Simulated waveforms of the SMIB system embedded with
BtB-STATCOM following this severe disturbance are presented in Figure 6.22.
Figure 6.22a shows that SG speed under the fault decreasingly oscillates for about 15
s when BtB-STATCOM is deactivated, but returns to its steady-state value with a
slight drop when BtB-STATCOM is in operation. The 0.0015 pu steady-state speed
difference is due to different demanded real power from SG in two different cases.
The nominal value of the real power exchange from Bus 1 to Bus 2 is around 0.45 pu
(45 MW) when a short line connects these two buses. Following fault, real power
exchange oscillates for a duration of around 8 s as shown in Figure 6.22b. With the
inclusion of BtB-STATCOM which ties neighboring buses, the oscillation is almost
damped out with a steady-state increase of real power exchange to 0.8 pu (80 MW).
Figure 6.22c shows that the DC link voltage of the BtB-STATCOM decreases for a
short time when the fault occurs at Bus 1 and restored to its controlled value
immediately, not affecting the operation of BtB-STATCOM. Reactive power
demand of SG in Figure 6.22d is temporarily increased under the fault, but it restores
immediately without losing stability when the fault is cleared. Figures 6.22e and
6.22f depicts that the dynamic voltage support within the study system is effectively
provided by BtB-STATCOM at two neighboring buses under the three-phase fault.
Figure 6.23 shows the traces of simulated voltage and current waveforms of BtB-
STATCOM converters following three-phase fault. Figure 6.24 shows the traces of
simulated phase shift angles (ɸM and ɸN) and selected GTO’s anode-to-cathode
voltages of BtB-STATCOM converters following three-phase fault.
195
(a) Transient response of SG speed following three-phase fault
(b) Transient response of SG real power output following three-phase fault
(c) DC link voltage excursions of BtB-STATCOM following three-phase fault
196
(d) Variation of SG reactive power output following three-phase fault
(e) Variation of Bus 1 voltage following three-phase fault
(f) Variation of Bus 2 voltage following three-phase fault

Figure 6.22. Simulated BtB-STATCOM performance in case 1
197
Figure 6.23. Simulated voltage and current waveforms of the converters
6.5.2.2. Case 2: Three-phase to Ground Fault at Infinite Bus
In this case, a relatively longer three-phase to ground fault is applied near Bus
2 (infinite bus) in Figure 6.21. The fault is lasted for 160 ms. Simulated waveforms
of the power system configuration embedded with BtB-STATCOM following the
disturbance are presented in Figure 6.25. The speed oscillations of the SG like the
one in previous case study are observed in Figure 6.25a when BtB-STATCOM is
deactivated. The oscillation duration is relatively shorter than that of previous case
198
since there is short line between SG and the fault location. When BtB-STATCOM is
activated, SG speed oscillation is better damped out when compared with the
previous case study. Moreover the slight drop in speed is not observed following
three-phase fault. This is due to the fact that the fault is occurred near a stiff bus and
BtB-STATCOM isolates the disturbance from SG with its converters. The same
steady-state speed difference of the SG (0.0015 pu) is observed like the one in
previous case study, due to the same loading condition of the SG as expected. In
Figure 6.25b, the maximum overshoot of the controlled SG real power output (0.8
pu) is significantly reduced in this case with a better response of the BtB-STATCOM
to the fault. The DC link voltage fall in Figure 6.25c is unavoidable. However, as
soon as the fault is cleared, DC link voltage restores to its reference without affecting
BtB-STATCOM operation. Reactive power demand of the SG in Figure 6.25d
restores immediately to its pre-fault value as soon as the fault is cleared. A less
undershoot shows up than that of previous case due to the fault isolation feature of
the BtB-STATCOM. Figures 6.25e and 6.25f depicts that the dynamic voltage
support within the study system is effectively provided by BtB-STATCOM at two
neighboring buses under the three-phase fault. In detail, the drop in Bus 2 voltage is
not avoided. However the voltage is controlled with less overshoot when BtB-
STATCOM is activated.
(a) Transient response of SG speed following three-phase fault
199
(b) Transient response of SG real power output following three-phase fault
(c) DC link voltage excursions of BtB-STATCOM following three-phase fault
(d) Variation of SG reactive power output following three-phase fault
200
(e) Variation of Bus 1 voltage following three-phase fault
(f) Variation of Bus 2 voltage following three-phase fault

Figure 6.25. Simulated BtB-STATCOM performance in case 2
Voltage distortions at Buses 1 and 2 as a measure of THD are listed in Table

6.5. Records of the simulated cases show that THD values are within acceptable
limits when BtB-STATCOM is in operation (IEEE, 1993). Even GTOs are switched
at fundamental system frequency, filtering is not required due to low THD content.
201

THD THD THD THD
Case 1
Case 2
V1(L-L) V2(L-L) V1(L-L) V2(L-L)
1.38 % 1.36 % 1.37 % 1.36 %
6.5.3. Discussion
It is shown that BtB-STATCOM, although not utilized primarily for

enhancing power system stability, can also be used to damp out generator speed
oscillations effectively even with simple PI controllers without any speed signal
measurement. DC power transmission among the two VSCs enables both controlled
real power exchange from one bus to another and improving power system stability
with the feature of isolating the fault from the rest of the power system in case of a
fault. As it can be seen the oscillations in the DC link are hardly noticeable. The
internal simulated signals of BtB-STATCOM show that the stable operation can be
achieved both in steady- and transient states in case of a severe disturbance while
providing reactive power to the neighboring buses for voltage control. Near-full rate
of the converters are achieved by setting the reference value of the real power
exchange among the two VSCs to relatively a large value of 0.8 pu.
6.6. Summary
In this chapter, strong control capability of the GUPFC with regulating multi-
line flows and bus voltage is extended with an optimized self-tuned fuzzy control
scheme for oscillation damping in a wind farm integrated power system. It is shown
both graphically and quantitatively that the proposed damping scheme is robust in its
performance over a range of disturbance conditions and does not only improves
transient stability of induction/synchronous generators but also assists indirectly to
other GUPFC control functions which are tightly interacted with each other. The
proposed control scheme is model independent since the design is based on
instantaneous system states rather than system parameters. With the inclusion of
quasi-multi pulse converters switching at 60 Hz into the grid, harmonic content
202
complies with the regulations. Hence, no filter is required for harmonic reduction at
the line side of the GUPFC converters.
Multi-line power flow control function of IPFC is extended with the
optimized self-tuned fuzzy control scheme, originally proposed for GUPFC, to
robustly mitigate inter-area mode of oscillations of a multi-machine power system
having two remote areas which are tied by double transmission circuit. The
performance of the damping scheme is verified graphically using time domain
instantaneous responses of the system states to various faults. As also shown in
GUPFC based transient stability studies, the proposed damping scheme assists
indirectly to the other real power flow control loop of the IPFC where damping
scheme is not utilized. The robustness of the proposed fuzzy damping scheme is
further verified by adapting it to the real power flow control loop of SSSC, which
yield particular performance comparison between IPFC and SSSC. The quasi-multi
pulse converters of the FACTS devices do not disturb power quality in terms of
harmonic content, which complies with the regulations. Hence, no filter is required at
the line side of the converters.
Multi-control function of BtB-STATCOM is examined without any damping
control scheme for improving power system stability considering various faults with
different locations and durations. The obtained results confirm that the real power
transfer controller of PI type of BtB-STATCOM can provide adequate damping of
generator speed oscillations owing to the segmentation of the power system with the
DC link of BtB-STATCOM. At the same time, it is ensured that all BtB-STATCOM
control loops are working truly without losing stability under different fault
scenarios. Voltage profiles of the neighboring buses are also improved with fast and
independent reactive power support of BtB-STATCOM converters in both steady-
and transient states.
203
204
7. CONCLUSIONS AND FUTURE WORK A. Mete VURAL
7. CONCLUSIONS AND FUTURE WORK
Now, more than ever, electrical power systems should be engineered in a

flexible and controllable fashion in response to ongoing expansion and growth of the
electrical power demand as well as increasing competitiveness and strict regulations
of transactions between electric power companies. To achieve both operational
flexibility and reliability, the existing transmission network should be utilized more
efficient under the fact that there are many obstacles to build new ones. Flexible
Alternating Current Transmission Systems (FACTS) are the power electronics based
high power equipment with advanced control techniques that can provide solutions to
new operating challenges of modern power systems. As the price of power
semiconductor devices are getting low with increasing voltage/current ratings,
FACTS devices are seen to be principal for the reliable and secure operation of
evolving electrical power systems and the number of FACTS device installations will
increase in the world.
Although, the practical application of the multi-converter FACTS devices is
still in its infancy, multi-converter FACTS devices are promising as multiple and fast
response compensators for the modern power systems in near future.
For the multi-converter FACTS devices to be included in transmission system
plans, there must be appropriate models for all the analyses that are normally
performed. Simulation has long been recognized as an important and necessary
procedure for the development, design, and testing phases of FACTS devices. Recent
advances in both computing hardware, and sophisticated power system component
modeling techniques have significantly increased the applications of digital
simulation of the power system industry embedded with advanced compensators,
such as FACTS devices. In this thesis, both 50 Hz and 60 Hz systems are addressed
for the investigation of potential FACTS device applications. Moreover, PSCAD is
interfaced to MATLAB through a custom written interface so that PSCAD is able to
execute one or more fuzzy inference systems which is/are not available in PSCAD
master library.
205
In the work reported in this thesis, a review of the state of the art is introduced
based on the simplified power circuit configuration and operating characteristics of
the first and the second generation of FACTS devices ranging from single- to multi-
converter topologies.
Power flow (load flow) analysis is an important tool for power system
planning in which the transmission constraints can also be determined in a given
network. Power flow solution gives information about the magnitude and phase
angle of the voltage at each bus as well as real and reactive power flows in each line
for given generation, load, and transmission network data of a power system. In this
context, the steady-state models of the GUPFC, IPFC, and BtB-STATCOM are
proposed and designed in PSCAD environment even PSCAD is primarily aimed to
simulate transient responses of the power system components. Developed models are
verified in various multi-bus power systems to demonstrate the capability of steady-
state controls of the real and reactive power flows on transmission lines as well as to
regulate system bus voltage. Steady-state models of STATCOM, SSSC, and UPFC
are also developed. Particular performance comparison is made between the
aforementioned FACTS devices. The advantage of this approach is fast, modular and
requiring no programming effort to include power injections and their derivatives
with respect to the state variables power system, such as bus voltages and their
respective phase angles, at the suitable locations of the Jacobian matrix and
mismatch vector. It is concluded that as long as the operational and control
constraints are satisfied, theoretically there is no limit in the number of VSCs which
are employed for building up the FACTS device. The method can suffer from long
computation time and may require high CPU computing power with large memory if
many multi-converter FACTS devices are embedded into relatively large power
systems having many buses.
In this thesis, eight two-level force-commutated converters are joined
together using magnetic interfaces to realize quasi multi-pulse converter operation
for multi-converter FACTS device applications. The quasi multi-pulse converter is
the building block of converter-level modeling studies of the multi-converter FACTS
devices. Appropriate adjustment of individual phase-shifted angles of the two groups
206
of series converters makes it possible to independently control the amplitude and the
phase of the AC voltage of the quasi multi-pulse converter. The quasi multi-pulse
converter based GUPFC, IPFC, and BtB-STACOM are successfully modeled in
PSCAD, including a detailed representation of the simulation design parameters. The
presented time-domain simulation results verify adequate operations of the GUPFC,
IPFC, and BtB-STATCOM separately, as well as demonstrating the successful
operations of the designed controllers. The proposed converter-level models of the
multi-converter FACTS devices can be directly implemented in any software
package that has a graphical interface. The models are independent of the type of the
control schemes applied for any multi-converter FACTS device.
The simplex optimization method and fuzzy logic techniques are used in
designing controllers of the multi-converter FACTS devices for various control
purposes. In the simplex optimization method, the cost function is minimized to
optimize single- and multi-controllers of the concerned multi-converter FACTS
device. Fuzzy logic theory is used to design two novel controllers for IPFC and
GUPFC, which are examined through time domain simulations of various case
studies applied in a variety of power systems. The first novel controller is based on
the combination of a conventional PI controller and a Mamdani-type fuzzy inference
system for the quasi multi-pulse IPFC, designed for high performance decoupling
action between controlled real and reactive power flows of a transmission line. In
general this control scheme can be employed in any series converter of the multi-
converter FACTS device, for instance UPFC or GUPFC to relieve inherent real and
reactive power flow coaction. On the contrary of analytically decoupled gain design,
the proposed control scheme is robust and does not rely on system mathematical
model. Consequently it adapts itself to parameter variations in the power system and
performs better. There is also an option to activate fuzzy component only when a
change in either real or reactive power flow command occurs. Such coordination can
yield improved rise time and settling time for start-up transients in simulation
environment.
Low frequency generator oscillations are commonly experienced due to
severe disturbances in the form of either local mode or inter-area mode of
207
oscillations in some parts or between parts of the interconnected power systems.

Local modes of oscillations take place as the synchronous generator swing with
respect to its reference speed or against the other generators in a certain area. On the
other hand, inter-area modes of oscillations occur as the synchronous generator
swing against each other which are remotely located with each other. The frequency
range of these oscillations is ranging from 0.1 to 2.5 Hz and may lead to total or
partial power interruption if not damped out effectively.
The second novel controller is based on self-tuned fuzzy damping scheme and
successfully implemented in GUPFC and IPFC configurations to suppress oscillation
damping and hence improve the transient stability of the power system, even
interconnected with wind farm. The damping control strategy is based on the
fuzzified manipulation of the q-component of the series converter voltage vector in
response to the change in both error and its integral/derivative. The self-tuning
mechanism enables the damping controller to cope with different operating points
since the output of the controller is tuned on-line. Extensive digital simulations in
PSCAD are carried out to examine the damping action of the proposed controller
under various system conditions. These include changes in power flow levels, change
of fault type together with its location and duration. The speed and rotor angle
deviations of the synchronous generators are damped out quickly than a system
without either GUPFC or IPFC. In this sense, the controller performance is proven in
terms of robustness and the presented time-domain simulations validate the proposed
control design. A brief comparison of the IPFC and SSSC is given in terms of
providing stability to a disturbed power system. It is concluded that for parallel
transmission lines, installing IPFC with two VSCs seems to be a better solution
option to damp inter-area mode of oscillations, as an alternative to SSSC.
It is shown that for BtB-STATCOM, conventional PI controllers are not only
adequate for steady-state power system parameter control but also suitable for
oscillation damping to enhance power system stability. Back-to-back operation of
high power converters enable load transfer between two interconnected grids,
without having to disconnect and then reconnecting it to the other system using
mechanical switches.
208
The controllers designed in this work are general and can be applied to other
FACTS devices easily. The results and discussions presented in this thesis will
provide valuable information to electric power utilities/companies in the near future
that are engaged in the planning and operation of the FACTS devices in Turkey
where mostly few STATCOM installations are reported.
This thesis reveals the potential usage and benefits of GUPFC, IPFC, and
BtB-STATCOM applications for intelligent control of future grids having distributed
energy resources, with emphasis on high power applications of converters with low
THD. Further research can be carried out in the following paragraphs:
• This thesis is focused on the performance evaluation of the multi-converter

FACTS devices operating individually. As a future work, the interaction
and the coordination of different types of single- and/or multi-converter
FACTS devices, spread out in large power systems, can be studied in
steady- and/or transient states.
• The benefits of the integration of energy storage systems to the multi-
converter FACTS devices can provide extra dynamic real power
capabilities to enhance stability and reliability of the transmission and
distribution systems. In this regard, potential application characteristics
can be investigated.
• Distribution static synchronous compensator (D-STATCOM) is well
addressed as voltage controller or power factor controller in distribution
networks. Similarly, the concept of GUPFC and IPFC can be shifted from
transmission level to distribution level for real-time control of multi-
system parameters simultaneously and independently in micro grid
applications including renewable energy sources such as solar and/or wind
energy.
• Based on the experience gained from this work and the simulation results
accomplished, practical applications of GUPFC, IPFC, or BtB-STATCOM
operations can be implemented and investigated by designing their scaled
laboratory prototypes operating in low or medium voltage levels.
209
210
REFERENCES
AALI S.R., NAZARPOUR D., 2010. 48-pulse GTO interline unified power flow
controller, 1st Power Electronic & Drive Systems & Technologies
Conference, pp. 52-56.
ABDUL H.M. KISHORE P.V., 2012. Multi converter unified power-quality
conditioning for multi feeder system, International Journal of Engineering
Research and Applications vol. 2, issue 1, pp. 212-221.
AJAMI A., KAMI A.-R., 2009. Modeling and controlling of IPFC based current-
source converter, 2nd International Conference on Computer and Electrical
Engineering, vol. 1, pp. 353-357.
ANDE S., KOTHARI M.L., 2007. Optimization of unified power flow controllers
(UPFC) using GEA, International Power Engineering Conference, pp. 53-58.
ANDERSON P.M., BOSE A., 1983. Stability simulation of wind turbine systems,
IEEE Transactions on Power Apparatus and Systems, vol. 102, no. 12, pp.
3791-3795.
AZBE V., MIHALIC R., 2009. Energy function for an interline power-flow
controller, Electric Power Systems Research, vol. 79, issue 6, pp. 945-952.
BAHRMAN M.P., JOHANSSON J.G., NILSSON B.A., 2003. Voltage source
converter transmission technologies: the right fit for the application, IEEE
Power Engineering Society General Meeting, vol. 3, pp. 1840-1847.
BANAEI M.R., KAMI A., 2009a. Interline power flow controller based damping
controllers for damping low frequency oscillations, 31st International
Telecommunications Energy Conference, pp.1-6.
BANAEI M.R., KAMI A., 2011. Interline power flow controller (IPFC) based
damping recurrent neural network controllers for enhancing stability, Energy
Conversion and Management, vol. 52, issue 7, pp. 2629-2636.
BANAEI M.R., TAHERI N., 2009b. HVDC based damping controllers for power
system stability, 31st International Telecommunications Energy Conference,
pp.1-6.
211
BELWANSHI S.M., CHANDRAKAR V.K., DHURVEY S.N., 2011. Performance
evaluation of IPFC by using fuzzy logic based controller for damping of
power system oscillations, 4th International Conference on Emerging Trends
in Engineering and Technology, pp.168-173.
BHARATHI R., RAJAN C.C.A., 2011. An advanced FACTS controller for power
flow management in transmission system using IPFC, International
Conference on Process Automation, Control and Computing, pp.1-6.
BHOWMICK S., DAS B. KUMAR N., 2009. An advanced IPFC model to reuse
newton power flow codes, IEEE Transactions on Power Systems, vol. 24, no.
2, pp. 525-532.
BORDALO U.A., RODRIGUES A.B., DA SILVA M.G., 2006. A new methodology
for probabilistic short-circuit evaluation with applications in power quality
analysis, IEEE Transactions on Power Systems, vol. 21, no. 2, pp. 474- 479.
CAVALIERE C.A.C, WATANABE E.H., AREDES M., 2002. Multi-pulse
STATCOM operation under unbalanced voltages, IEEE Power Engineering
Society Winter Meeting, vol. 1, pp. 567- 572.
CHAKRABORTY ARINDAM, 2011. [pe2], Advancements in power electronics
and drives in interface with growing renewable energy resources, Renewable
and Sustainable Energy Reviews, vol. 15, issue 4, pp. 1816-1827.
CHEN Z., BLAABJERG F., 2009. Wind farm-A power source in future power
systems, Renewable and Sustainable Energy Reviews, vol. 13, issues 6-7, pp.
1288-1300.
CHENGSHENG W., CHONGJIAN L., YAOHUA L., CHUNYI Z., ZHIMING L.,
2009. [pe3] Investigation on the switching property of IGCT used in large
power voltage source inverter, IEEE 6th International Conference on Power
Electronics and Motion Control Conference, pp. 291-294.
CHOMPOO-INWAI C., LEE W.J., FUANGFOO P., WILLIAMS M., LIAO J.R.,
2005. System impact study for the interconnection of wind generation and
utility system, IEEE Transactions on Industry Applications, vol. 41, no. 1. pp.
163- 168.
212
DAVALOS R.M., RAMIREZ J.M., TAPIA R.O., 2005. Three-phase multi-pulse
converter statcom analysis, International Journal of Electrical Power &
Energy Systems, vol. 27, issue 1, pp. 39-51.
DIVYA K.C., OSTERGAARD J., 2009. Battery energy storage technology for
power systems-An overview, Electric Power Systems Research, vol. 79, issue
4, pp. 511-520.
ERLICH I., KRETSCHMANN J., FORTMANN J., MUELLER-ENGELHARDT S.,
WREDE H., 2007. Modeling of wind turbines based on doubly-fed induction
generators for power system stability studies, IEEE Transactions on Power
Systems, vol. 22, no. 3, pp. 909-919.
FARAHANI E.M., AFSHARNIA S., 2006. DM for UPFC's active & reactive power
decoupled control, IEEE International Symposium on Industrial Electronics,
pp.1916-1921.
FARDANESH B., SHPERLING B., UZUNOVIC E., ZELINGHER S., 2000. Multi-
converter FACTS devices: the generalized unified power flow controller
(GUPFC), IEEE Power Engineering Society Summer Meeting, vol. 2, pp.
1020-1025.
FARIED S.O., TANG G., EDRIS A., 2009. Supplemental control of voltage sourced
converter-based back-to-back for damping subsynchronous resonance, IEEE
Power & Energy Society General Meeting, pp. 1-6.
FUJII T., TEMMA K., MORISHIMA N., AKEDANI T., SHIMONOSONO T.,
HARADA H., 2010. [fapp9], 450MVA GCT- STATCOM for stability
improvement and over-voltage suppression, International Power Electronics
Conference, pp.1766-1772.
GEETHALAKSHMI B., DANANJAYAN P., 2007. An analytic approach to
harmonic analysis of 48-pulse voltage source inverter, 7th International
Conference on Power Electronics and Drive Systems, pp. 417-422.
GOLE A.M., FILIZADEH S., MENZIES R.W., WILSON P.L., 2005. Optimization-
enabled electromagnetic transient simulation, IEEE Transactions on Power
Delivery, vol. 20, no. 1, pp. 512- 518.
213
GOODWIN G.C., GRAEBE S.F., SALGADO M.E., 2000. Control System Design,
Prentice Hall, 2000.
GULTEKIN B., GERCEK C.O., ATALIK T., DENIZ M., BICER N., ERMIS M.,
KOSE K. .N., ERMIS C., KOC E., CADIRCI I., ACIK A., AKKAYA Y.,
TOYGAR H., BIDECI S., 2012. Design and implementation of a 154-kV
±50-mvar transmission STATCOM based on 21-level cascaded multilevel
converter, IEEE Transactions on Industry Applications, vol. 48, no. 3, pp.
1030-1045.
GYUGYI L., 1992. Unified power-flow control concept for flexible AC transmission
systems, IEE Proceedings C, Generation, Transmission and Distribution, vol.
139, no. 4, pp. 323-331.
GYUGYI L., 2000. Application characteristics of converter-based FACTS
controllers, International Conference on Power System Technology, vol. 1,
pp. 391-396.
GYUGYI L., SCHAUDER C.D., SEN K.K., 1997. Static synchronous series
compensator: a solid-state approach to the series compensation of
transmission lines, IEEE Transactions on Power Delivery, vol. 12, no. 1, pp.
406-417.
GYUGYI L., SEN K.K., SCHAUDER C.D., 1999. The interline power flow
controller concept: a new approach to power flow management in
transmission systems, IEEE Transactions on Power Delivery, vol. 14, no. 3,
pp. 1115-1123.
GYUGYI L., SHAUDER C.D., WILLIAMS S.L., RIETMAN T.R., TORGERSON
D.R., EDRIS A., 1995. The unified power flow controller: a new approach to
power transmission control. IEEE Transactions on Power Delivery, vol. 10,
no. 2, pp. 1085-1093.
HAGIWARA M., AKAGI H., 2005. An approach to regulating the DC-link voltage
of a voltage-source BTB system during power line faults, IEEE Transactions
on Industry Applications, vol. 41, no. 5, pp. 1263- 1271.
214
HAGIWARA M., FUJITA H., AKAGI H., 2003. Performance of a self-commutated
BTB HVDC link system under a single-line-to-ground fault condition,
IEEE Transactions on Power Electronics, vol. 18, no. 1, pp. 278-285.
HAGIWARA M., PHUONG V.P., AKAGI H., 2008. Calculation of DC magnetic
flux deviation in the converter-transformer of a self-commutated BTB system
during single-line-to-ground faults, IEEE Transactions on Power Electronics,
vol. 23, no. 2, pp. 698-706.
HAMEED S., DAS B., PANT V., 2008. A self-tuning fuzzy PI controller for TCSC
to improve power system stability, Electric Power Systems Research, vol. 78,
issue 10, pp. 1726-1735.
HEINE P., LEHTONEN M., 2003. Voltage sag distributions caused by power
system faults, IEEE Transactions on Power Systems, vol. 18, no. 4, pp. 1367-
1373.
HINGORANI N.G., 2000. [fa2] FACTS technology - state of the art, current
challenges and the future prospects, IEEE Power Engineering Society
General Meeting, pp. 1-4.
HINGORANI N.G., GYUGYI L., 2000. Understanding FACTS: Concepts and
Technology of Flexible Ac Transmission Systems., IEEE press, New Jersey,
432p.
HINGORANI, N.G., 1988. Power electronics in electric utilities: role of power
electronics in future power systems, IEEE Proceedings, vol. 74 no. 4, pp.
481-482.
HOSSEINI S.H., KHEZRLU A.V., 2011. Interline power flow controller based on
multi output sparse matrix converter, 7th International Conference on
Electrical and Electronics Engineering, pp. 307-311.
IEEE, 1993. IEEE recommended practices and requirements for harmonic control in
electrical power systems, Institute of Electrical and Electronics Engineering,
April 1993.
IEEE, 1994. IEEE PES and CIGRE FACTS Working Group, E. Larsen and D.
Torgerson, CoChairs, FACTS Overview, IEEE PES Special Publication 95-
TP-108, 1995.
215
IEEE, 1997. [22] IEEE Committee Rep, First benchmark model for computer
simulation of sub-synchronous resonance, IEEE Transactions on Power
Apparatus and Systems, vol. 96, no. 5, pp. 1565–1572.
IEEE, 1997. Proposed terms and definitions for flexible AC transmission system
(FACTS), IEEE Transactions on Power Delivery, vol. 12, no. 4, pp. 1848-
1853.
IM S.Y., YOON J.S., CHANG B.H., BAEK D.H., 2005. [fapp5] The operation
experience of KEPCO UPFC, 8th International Conference on Electrical
Machines and Systems, vol. 3, pp.2502-2505.
JAUCH C., SORENSEN P., NORHEIM I., RASMUSSEN C., 2007. Simulation of
the impact of wind power on the transient fault behavior of the Nordic power
system, Electric Power Systems Research, vol. 77, issue 2, pp. 135-144.
JOVCIC D., LAMONT L., ABBOTT K., 2007. Control system design for VSC
transmission, Electric Power Systems Research, vol. 77, issue 7, pp. 721-729.
JUNYENT-FERRE A., GOMIS-BELLMUNT O., SUMPER A., SALA M., MATA
M., 2010. Modeling and control of the doubly fed induction generator wind
turbine, Simulation Modelling Practice and Theory, vol. 18, issue 9, pp.
1365-1381.
KALDELLIS J.K., ZAFIRAKIS D., 2011. The wind energy (r)evolution: A short
review of a long history, Renewable Energy, vol. 36, issue 7, pp. 1887-1901.
KARTHIK B., 2007. Modeling studies of inter line power flow controller for multi-
line transmission system, IET-UK International Conference on Information
and Communication Technology in Electrical Sciences, pp. 475-479.
KAZEMI A., KARIMI E., 2006. The effect of interline power flow controller (IPFC)
on damping inter-area oscillations in the interconnected power systems, IEEE
International Symposium on Industrial Electronics, vol. 3, pp. 1911-1915.
KAZERANI M., YANG Y., 2002. Comparative evaluation of three-phase PWM
voltage- and current source converter topologies in FACTS applications,
IEEE Power Engineering Society Summer Meeting, vol. 1, pp. 473-479.
KUNDUR, P., 1994. Power System Stability and Control. McGraw-Hill, New York,
1176p.
216
KUSIAK A., LI W., 2010. Estimation of wind speed: A data-driven approach,
Journal of Wind Engineering and Industrial Aerodynamics, voli 98, issues
10–11, pp. 559-567.
LARSSON T., PETERSSON A., EDRIS A., KIDD D., ABOYTES F., 2001. Eagle
pass back-to-back tie: a dual purpose application of voltage source converter
technology, IEEE Power Engineering Society Summer Meeting, vol. 3, pp.
1686-1691.
LEE C.K., LEUNG J.S.K., HUI S.Y.R., CHUNG H.S.-H., 2003. Circuit-level
comparison of STATCOM technologies, IEEE Transactions on Power
Electronics, vol. 18, no. 4, pp. 1084- 1092.
LEE Y.O, KANG H.J., 2011, HAN Y., CHUNG C.C., A nonlinear control for a
BTB STATCOM system with asymmetrically structured converters, IEEE
Trondheim PowerTech, pp.1-8.
LEVI E., 1997. General method of magnetising flux saturation modelling in d-q axis
models of double-cage induction machines, IEE Proceedings - Electric Power
Applications, vol. 144, no. 2, pp.101-109.
LEVY H., LESSMAN F., 2011. Finite Difference Equations, Dover Publications,
New York, 278p.
LI H., ZHAO B., YANG C., CHEN H.W., CHEN Z., 2011. Analysis and estimation
of transient stability for a grid-connected wind turbine with induction
generator, Renewable Energy, vol. 36, issue 5, pp. 1469-1476.
LIMING L., PENGCHENG Z., KANG Y., CHEN J., 2007. Power-flow control
performance analysis of a unified power-flow controller in a novel control
scheme, IEEE Transactions on Power Delivery, vol. 22, no. 3, pp. 1613-1619.
LIMING L., YONGGAO Z., PENGCHENG Z., YONG K., KANG J.C., 2005.
Control scheme and implement of a unified power flow controller,
International Conference on Electrical Machines and Systems, pp. 1170-
1175.
LIU J. TANG Y., SUN H., XUE M., LIU B., 2010. Control strategy for power grids
synchronism parallel based on back-to-back VSC, International Conference
on Power System Technology, pp. 1-8.
217
LUBIS R.S., 2011a. Modeling and simulation of generalized unified power flow
controller (GUPFC), 2nd International Conference on Instrumentation,
Communications, Information Technology, and Biomedical Engineering, pp.
207-213.
LUBIS R.S., 2011b. Digital simulation of the FACTS system with 60-pulse GTO-
based voltage source converter, 2nd International Conference on
Instrumentation, Communications, Information Technology, and Biomedical
Engineering, pp.201-206.
LUBIS R.S., HADI S.P., TUMIRAN, 2012. Dynamic simulation of the generalized
unified power flow controller in multi-machine power systems, International
Journal of Electrical & Computer Sciences, vol. 12, no. 3, pp. 75-84.
MA T.T., 2007. P-Q decoupled control schemes using fuzzy neural networks for the
unified power flow controller, International Journal of Electrical Power &
MADHAN MOHAN D., SINGH B., PANIGRAHI B.K., 2009. Harmonic optimised
24-pulse voltage source converter for high voltage DC systems, IET Power
Electronics, vol. 2, no. 5, pp. 563-573.
MENNITI D., PINNARELLI A., SORRENTINO N., 2002. A fuzzy logic controller
for interline power flow controller model implemented by ATP-EMTP,
International Conference on Power System Technology, vol. 3, pp. 1898-
1903.
MISHRA S., 2006. Neural-network-based adaptive UPFC for improving transient
stability performance of power system, IEEE Transactions on Neural
Networks, vol. 17, no. 2, pp. 461-470.
MISHRA S., DASH P.K., HOTA P.K., TRIPATHY M., 2002. Genetically optimized
neuro-fuzzy IPFC for damping modal oscillations of power system, IEEE
Transactions on Power Systems, vol. 17, no. 4, pp. 1140- 1147.
MOGHADAM M.F., ABYANEH H.A., FATHI S.H., KHEDERZADEH M., 2011.
Voltage compensation with interline power flow controller (IPFC) using all
degrees of freedom, 6th IEEE Conference on Industrial Electronics and
Applications, pp. 2179-2184.
218
MOGHADAM M.F., GHAREHPETIAN G.B., ABYANEH H.A., 2010. Optimized
regulation of DC voltage in interline power flow controller (IPFC) using
genetic algorithm, 4th International Power Engineering and Optimization
Conference, pp. 117-121.
MORI S., MATSUNO K., HASEGAWA T. OHNISHI, S., TAKEDA, M., SETO,
M., MURAKAMI, S., ISHIGURO, F., 1993. Development of a large static
var generator using self-commutated inverters for improving power system
stability, IEEE Transactions on Power Systems, vol. 8, no. 1, pp. 371-377.
MUDI R.K., PAL N.R., 1999. A robust self-tuning scheme for PI- and PD-type
fuzzy controllers, IEEE Transactions on Fuzzy Systems, vol. 7, no. 1, pp. 2-
16.
MURUGANANDHAM J., GNANADASS R., 2012. Performance analysis of
interline power flow controller for practical power system, IEEE Students'
Conference on Electrical, Electronics and Computer Science, pp. 1-8.
NATALIA M.R., SANTOS O.P., DIAS V., FERNAO P., 2012. Use of an interline
power flow controller model for power flow analysis, Energy Procedia, vol.
14, pp. 2096-2101.
NEIDER J.A., MEAD R., 1965. A simplex method for function minimization, The
Computer Journal, vol. 7, no. 4, pp. 308-313.
PADHY N.P., MOAMEN M.A.A., 2005. Power flow control and solutions with
multiple and multi-type FACTS devices, Electric Power Systems Research,
vol. 74, issue 3, pp. 341-351.
PADIYAR K.R., 2007. FACTS controllers in power transmission and distribution,
New Age International Publishers, New Delphi, 532p.
PADIYAR K.R., PRAPHU N., 2007. Analysis of SSR with three-level twelve-pulse
VSC-based interline power-flow controller, IEEE Transactions on Power
Delivery, vol. 22, no. 3, pp. 1688-1695.
PAPIC I., ZUNKO P., 2003. UPFC converter-level control system using internally
calculated system quantities for decoupling, International Journal of Electrical
Power & Energy Systems, vol. 25, issue 8, pp. 667-675.
219
PAPIC I., ZUNKO P., POVH D., WEINHOLD M., 1997. Basic control of unified
power flow controller, IEEE Transactions on Power Systems, vol. 12, no. 4,
pp. 1734-1739.
PARIMI A.M., ELAMVAZUTHI I., SAAD N., 2008. Interline power flow
controller (IPFC) based damping controllers for damping low frequency
oscillations in a power system, IEEE International Conference on Sustainable
Energy Technologies, pp. 334-339.
PARIMI A.M., ELAMVAZUTHI I., SAAD N., 2010a. Fuzzy logic control for IPFC
for damping low frequency oscillations, International Conference on
Intelligent and Advanced Systems, pp. 1-5.
PARIMI A.M., SAHOO N.C., ELAMVAZUTHI I., SAAD N., 2010b. Dynamic
modeling of interline power flow controller for small signal stability, IEEE
International Conference on Power and Energy, pp. 583-588.
PARIMI A.M., SAHOO N.C., ELAMVAZUTHI I., SAAD N., 2011. Transient
stability enhancement and power flow control in a multi-machine power
system using interline power flow controller, International Conference on
Energy, Automation, and Signal, pp. 1-6.
PARKHIDEH B., BHATTACHARYA S., 2009. Resilient operation of voltage-
sourced BTB HVDC systems under power system disturbances, IEEE Power
& Energy Society General Meeting, pp. 1-7.
PIZANO-MARTINEZ A., FUERTE-ESQUIVEL C.R., AMBRIZ-PEREZ H.,
ACHA E., 2007. Modeling of VSC-based HVDC systems for a Newton-
Raphson OPF algorithm, IEEE Transactions on Power Systems, vol. 22, no.
4, pp. 1794-1803.
PRAKASH T.R.D., NAIR N.K., 2007. Voltage sag mitigation in multi-line
transmission system using generalized unified power flow controller, Journal
of Intelligent Electronic Systems, vol. 1, pp.72-78.
PSCAD, 2010. PSCAD User’s Guide, Version 4.2.1., 2005, http://www.pscad.com.
REED G., PAPE R., TAKEDA M., 2003. Advantages of voltage sourced converter
(VSC) based design concepts for FACTS and HVDC-link applications, IEEE
Power Engineering Society General Meeting, vol. 3, pp. 1816-1821.
220
REED G., PASERBA J., CROASDAILE T., TAKEDA M., MORISHIMA N.,
HAMASAKI Y., THOMAS L., ALLARD W., 2001. STATCOM application
at VELCO Essex substation, IEEE/PES Transmission and Distribution
Conference and Exposition, vol. 2, pp. 1133-1138.
REED G., PASERBA J., CROASDAILE T., WESTOVER R., JOCHI S.,
MORISHIMA N, TAKEDA M., SUGIYAMA T., HAMASAKI Y., SNOW
T., ABED A., 2002. [fapp10] SDG&E Talega STATCOM project-system
analysis, design, and configuration, IEEE/PES Transmission and Distribution
Conference and Exhibition 2002: Asia Pacific, vol. 2, pp. 1393- 1398.
RENZ B.A., KERI A., MEHRABAN A.S., SCHAUDER C., STACEY E.,
KOVALSKY L., GYUGYI L., EDRIS A., 1999. [fapp2] AEP unified power
flow controller performance, IEEE Transactions on Power Delivery, vol. 14,
no. 4, pp. 1374-1381.
RUIHUA S., CHAO Z., RUOMEI L., XIAOXIN Z., 2005. VSCs based HVDC and
its control strategy, IEEE/PES Transmission and Distribution Conference and
Exhibition: Asia and Pacific, pp.1-6.
SAMUELSSON O., LINDAHL S., 2005. On speed stability, IEEE Transactions on
Power Systems, vol. 20, no. 2, pp. 1179- 1180.
SAUER P.W., PAI M.A., 1998. Power System Dynamics and Stability, Prentice
Hall, New Jersey, 357p.
SCHAUDER C., 1991. Vector analysis and control of advanced static VAr
compensators, International Conference on AC and DC Power Transmission,
pp. 266-272.
SCHAUDER C., GERNHARDT M., STACEY E., LEMAK T., GYUGYI L.,
CEASE T.W., EDRIS A., 1997. Operation of ±100 MVAr TVA STATCON,
IEEE Transactions on Power Delivery, vol. 12, no. 4, pp. 1805-1811.
SCHAUDER C., GERNHARDT M., STACEY E., LEMAK T., GYUGYI L.,
CEASE T.W., EDRIS A., 1995. Development of a ±100 MVAr static
condenser for voltage control of transmission systems, IEEE Transactions on
Power Delivery, vol. 10, no. 3, pp. 1486-1496.
221
SEN K.K., 1998. SSSC-static synchronous series compensator: theory, modeling,
and application, IEEE Transactions on Power Delivery, vol. 13, no. 1, pp.
241-246.
SENTHIL-KUMAR N., GOKULAKRISHNAN J., 2011. Impact of FACTS
controllers on the stability of power systems connected with doubly fed
induction generators, International Journal of Electrical Power & Energy
Systems, vol. 33, issue 5, pp. 1172-1184.
SHAN J. GOLE A.M., ANNAKKAGE U.D., JACOBSON D.A., 2011. Damping
performance analysis of IPFC and UPFC controllers using validated small-
signal models, IEEE Transactions on Power Delivery, vol. 26, no. 1, pp. 446-
454.
SINGH B., SAHA R., CHANDRA A., AL-HADDAD K., 2009. Static synchronous
compensators (STATCOM): a review, IET Power Electronics, vol. 2, no. 4,
pp. 297-324.
SOOD V.K., 2004. HVDC and FACTS controllers applications of static converters
in power systems, Kluwer Academic Publishers, Boston, 295p.
SORENSEN P., HANSEN A.D:, ROAS P.A.C., 2002. Wind models for simulation
of power fluctuations from wind farms, Journal of Wind Engineering and
Industrial Aerodynamics, vol. 90, issues 12-15, pp. 1381–1402.
SOTO D., GREEN T.C., 2002. A comparison of high-power converter topologies for
the implementation of FACTS controllers, IEEE Transactions on Industrial
Electronics, vol. 49, no. 5, pp. 1072-1080.
STRZELECKI R., SMERECZYNSKI P., BENYSEK G., 2005. Interline power flow
controller-properties and control strategy in dynamic states, IEEE
Compatibility in Power Electronics, pp. 11- 17.
SUJIN P.R., PRAKASH T.R.D., PADMA S., 2012. Two neuron model for voltage
flicker mitigation using generalized unified power flow controller,
International Journal of Current Engineering and Technology, vol. 2, no. 2,
pp 244-250.
SUN J., HOPKINS L., SHPERLING B., FARDANESH B., GRAHAM M., PARISI
M., MACDONALD S., BHATTACHARYA S., BERKOWITZ S., EDRIS
222
A., 2004. Operating Characteristics of the Convertible Static Compensator on
the 345 kV Network, IEEE PES Power Systems Conference and Exposition,
vol. 2, pp. 732-738.
SUN L., MEI S., LU Q., MA J., 2003. Application of GUPFC in China's Sichuan
power grid - modeling, control strategy and case study, IEEE Power
Engineering Society General Meeting, vol. 1, pp. 175- 181.
TAN J.-S., ARNOLD C.P., 2006. Fundamental voltage modulation for a high power
voltage source converter, 41st International Universities Power Engineering
Conference, vol. 1, pp. 313-317.
TENG H., LIU C., HAN M., MA S., GUO X., 2010. Synchronous generator model
transformation between BPA and PSCAD, Asia-Pasicif Power and Energy
Engineering Conference, pp. 1-4.
TÜMAY M., VURAL A.M., LO K.L., 2004. The effect of Unified Power Flow
Controller (UPFC) location in power systems, International Journal of
Electrical Power & Energy Systems, vol. 26, issue 8, pp. 561-569.
TYAGI A., PADIYAR K.R., 2006. Dynamic analysis and simulation of a VSC based
back-to-back HVDC link, India International Conference on Power
Electronics, pp. 232-238.
VASQUEZ-ARNEZ R.L., ZANETTA L.C., 2008. A novel approach for modeling
the steady-state VSC-based multiline FACTS controllers and their operational
constraints, IEEE Transactions on Power Delivery, vol. 23, no. 1, pp. 457-
464.
VEERAMALLA J., SREERAMA-KUMAR R., 2010. Application of interline power
flow controller (IPFC) for damping low frequency oscillations in power
systems, International Symposium on Modern Electric Power Systems, pp. 1-
6.
VINKOVIC A., MIHALIC R., 2011. Universal method for the modeling of the 2nd
generation FACTS devices in Newton–Raphson power flow, International
Journal of Electrical Power & Energy Systems, vol. 33, issue 10, pp. 1631-
1637.
223
VOBECKY J., 2011. [pe5], Design and technology of high-power silicon devices,
18th International Conference on Mixed Design of Integrated Circuits and
Systems, pp. 17-22.
VURAL A.M., TÜMAY, 2007. Mathematical modeling and analysis of a unified
power flow controller: A comparison of two approaches in power flow
studies and effects of UPFC location, International Journal of Electrical
Power & Energy Systems, vol. 29, issue 8, pp. 617-629.
WANG H.F., JAZAERI M., JOHNS A.T., 2000. Investigation into the dynamic
interactions of multiple multifunctional unified power flow controllers, IEEE
Power Engineering Review, vol. 20, no. 7, pp. 45-48.
WASHINGTON, 2012. Power Systems Test Case Archive,
http://www.ee.washington.edu/ research/pstca/pf14/pg_tca14bus.htm.
XIA J., CHOW J.H., EDRIS A.-A., FARDANESH B., UZUNOVIC E., 2010.
Transfer path stability enhancement by voltage-sourced converter-based
FACTS controllers, IEEE Transactions on Power Delivery, vol. 25, no. 2, pp.
1019-1025.
XIA J., XINGHAO F., CHOW J.H., EDRIS A.-A., UZUNOVIC E., PARISI, M.,
HOPKINS L., 2008. A novel approach for modeling voltage-sourced
converter-based FACTS controllers, IEEE Transactions on Power Delivery,
vol. 23, no. 4, pp. 2591-2598.
XINGHAO F., CHOW J.H., 2009. BTB DC link modeling, control, and application
in the segmentation of AC interconnections, IEEE Power & Energy Society
General Meeting, pp. 1-7.
XUAN W., CHOW J.H., FARDANESH B., EDRIS A.-A., 2004. A common
modeling framework of voltage-sourced converters for load flow, sensitivity,
and dispatch analysis, IEEE Transactions on Power Systems, vol. 19, no. 2,
pp. 934- 941.
YAM C.M., HAQUE M.H., 2002. Dynamic decoupled compensator for UPFC
control, International Conference on Power System Technology, pp. 1482-
1487.
224
YANKUI Z. YAN Z. CHEN C., 2006. A novel power injection model of IPFC for
power flow analysis inclusive of practical constraints, IEEE Transactions on
Power Systems, vol. 21, no. 4, pp. 1550-1556.
YE Y., KAZERANI M., 2006. Power flow control schemes for series-connected
FACTS controllers, Electric Power Systems Research, vol. 76, issues 9–10,
pp. 824-831.
YONGSUG S., STEIMER P.K., 2009. [pe4] Application of IGCT in high-power
rectifiers, IEEE Transactions on Industry Applications, vol. 45, no. 5, pp.
1628-1636.
YU Q., ROUND S.D., NORUM L.E., UNDELAND T.M., 1996. Dynamic control of
a unified power flow controller, 27th Annual IEEE Power Electronics
Specialists Conference, pp. 508-514.
ZELINGHER S., FARDANESH B., SHPERLING B., DAVE S., KOVALSKY L.,
SCHAUDER C., EDRIS A., 2000. Convertible static compensator project-
hardware overview, IEEE Power Engineering Society Winter Meeting, vol. 4,
pp. 2511-2517.
ZHANG X.-P. HANDSCHIN E., YAO M., 2001. Modeling of the generalized
unified power flow controller (GUPFC) in a nonlinear interior point OPF,
IEEE Transactions on Power Systems, vol. 16, no. 3, pp. 367-373.
ZHANG X.-P., 2004. Multiterminal voltage-sourced converter-based HVDC models
for power flow analysis, IEEE Transactions on Power Systems, vol. 19, no. 4,
pp. 1877-1884.
ZHANG X.P., REHTANZ C., PAL B., 2006. Flexible AC transmission systems:
Modelling and control. Springer, Berlin, 383p.
ZHANG, 2003. Modelling of the interline power flow controller and the generalised
unified power flow controller in Newton power flow, IEE Proceedings,
Generation, Transmission and Distribution, vol. 150, no. 3, pp. 268- 274.
225
226
CIRRICULUM VITAE
Education:
PhD Electrical and Electronics Çukurova University 2009-…..
Engineering
Msc Electrical and Electronics University of Gaziantep and 1999-2001
Engineering University of Strathclyde
(as visitor scholar in 2000)
Bsc Electrical and Electronics University of Gaziantep 1995-1999
Engineering
Work Experience:
Hasan Kalyoncu University, Electrical and Full-time July-2011 / present
Electronics Engineering Department instructor
Atılım University, Electrical and Full-time Sept-2008 / June-2011
Electronics Engineering Department instructor
Wuppertal University, Automation and Research Nov-2004 / Mar-2007
Control Engineering Department,Germany assistant
Gaziantep University, Electrical and Research Aug-1999 / Nov-2004
Electronics Engineering Department assistant
Research Interests:
• Modeling and control of FACTS devices
• Computational intelligence applications to FACTS device control
• Power system simulation
Professional Activities:
• Refereeing
International Journal of Electrical Power and Energy Systems
IET Generation, Transmission & Distribution
Turkish Journal of Electrical Engineering & Computer Sciences
Journal of Electrical and Electronics Engineering Research
World Journal of Modeling and Simulation
Journals of Zhejiang University-Science-C (Computers & Electronics)
Ain Shams Engineering Journal
• IEEE student member since 1999
• EMO member since 1999
227
Given Courses:
EE 451 - Power System Analysis
EE 452 - High Voltage Techniques
EE 450 - Electrical Machinery and Drives
EEE 101 - Introduction to Electrical and Electronics Engineering
EEE 120 - Introduction to MATLAB
EEE 201/202 - Circuit Analysis I-II
EE 203 - Digital Circuits and Systems
EE 403 - Communication Networks
EE 491/492 - Design Project I-II
MATH 151 - Calculus I
Publications:
• Journals (SCI-E)
VURAL A.M., BAYINDIR K.Ç., 2012. Transient stability enhancement of the power
system interconnected with wind farm using generalized unified power flow
controller with simplex optimized self-tuning fuzzy damping scheme,
International Review of Electrical Engineering, vol. 7, no. 4, pp. 5091-5107.
VURAL A.M., BAYINDIR K.Ç., 2012. A hybrid fuzzy-PI control scheme for a quasi
multi-pulse interline power flow controller including PQ decoupling feature,
Journal of Power Electronics, vol. 12, no. 5, pp. 787-799.
VURAL A.M., BAYINDIR K.Ç., 2011. Two-level quasi multi-pulse voltage source
converter based generalized unified power flow controller, International
Review of Electrical Engineering, vol. 6, no. 5, pp. 2622-2637.
VURAL A.M., TÜMAY M., 2007. Mathematical modeling and analysis of a unified
power flow controller: A comparison of two approaches in power flow studies
and effects of UPFC location, International Journal of Electrical Power &
TÜMAY M., VURAL A.M., LO K.L., 2004. The effect of unified power flow
controller (UPFC) location in power systems, International Journal of
Electrical Power & Energy Systems, vol. 26, issue 8, pp. 561-569.
VURAL A.M., TÜMAY M., 2004. Analysis and modeling of unified power flow
controller: Modification of Newton-Raphson algorithm and user-defined
228
modeling approach for power flow studies, Arabian Journal for Science and
Engineering, vol. 29, no: 2B, pp. 135-153.
TÜMAY M., VURAL A.M., LO K.L., 2005. Simulation of unified power flow
controller by using modified power injection model, Iranian Journal of
Science and Technology, vol. 29, pp. 49-64.
• Journals
VURAL A.M., BAYINDIR K.Ç., 2011. Quasi multi-pulse back-to-back static

synchronous compensator employing line frequency switching 2-level GTO
inverters, World Academy of Science, Engineering and Technology, issue 60,
pp. 1863-1874.
VURAL A.M., BAYINDIR K.Ç., 2011. Simplex optimized twelve-pulse STATCOM
control system and LC filter, European Journal of Scientific Research vol. 64,
issue 3, 2011.
TÜMAY M., EKER İ., AKSOY H.F., VURAL A.M., ÜNVER M.U., 2002. Dynamic
performances of adjustable Speed AC drives part I: Dynamic modelling and
implementation of PWM-fed synchronous and asynchronous machines,
Information Technology Journal, vol.1, no. 2, pp. 98-105.
TÜMAY M., EKER İ., AKSOY H.F., VURAL A.M., ÜNVER M.U., 2002. Dynamic
performances of adjustable speed AC drives part II: Control and simulation of
AC machines, Information Technology Journal, vol. 1, no. 2, pp. 106-117.
• International Conferences
VURAL A.M., BAYINDIR K.Ç., 2012. Converter level modeling and control of
quasi multi-pulse static synchronous series compensator, IEEE Symposium
on Electrical and Electronics Engineering, EEESYM’2012, pp. 698-702.
VURAL A.M., BAYINDIR K.Ç., 2012. Understanding the steady-state modeling and
analysis of power systems embedded with VSC-based FACTS devices, IEEE
EnergyTech2012.
229
VURAL A.M., BAYINDIR K.Ç., 2010. Optimization of parameter set for
STATCOM control system, IEEE PES, Transmission and Distribution
Conference and Exposition, pp. 1-6.
VURAL A.M., TÜMAY M., 2003. Steady state analysis of unified power flow
controller: Mathematical modeling and simulation studies, IEEE Powertech
Conference, vol. 4.
EKER İ., VURAL A.M., 2003. Experimental on-line identification of a three-mass
mechanical system, IEEE Conference on Control Applications, vol. 1, pp. 60-
65.
VURAL A.M., TÜMAY M., 2003. Power flow analysis of power system embedded
with UPFC using Psasp program, International Conference on Electrical and
Electronics Engineering, ELECO’2003, pp.22-26.
EKER İ., VURAL A.M., SÜSLÜOĞLU B., 2003. Experimental identification of an
electromechanical system running in open-loop conditions, International
Conference on Electrical and Electronics Engineering, ELECO’2003, pp.
284-288.
VURAL A.M., TÜMAY M., 2001. UPFC for controlling power flow in power
systems, International Conference on Electrical and Electronics Engineering,
ELECO’2001, pp. 1-4.
VURAL A.M., EKER İ., 2004. Parameter identification of a permanent magnet DC
motor: An experimental approach, International Conference on Electrical
Machines, ICEM’2004, paper no. 104.
• National Conferences
VURAL A.M., BAYINDIR K.Ç., TÜMAY M., 2009. 12 darbeli bir

STATCOM için denetleyici ve filtre parametrelerinin simplex yöntemi ile
optimizasyonu, 13. Elektrik, Elektronik, Bilgisayar, Biyomedikal Mühendisliği
Ulusal Kongresi, ODTÜ, Ankara.
VURAL A.M., TÜMAY M., 2003. Gelişmiş güç akış denetleyicileri ile
donatılmış güç sistemlerinin Newton-Raphson metodu ile analizi, 10. Ulusal
Elektrik-Elektronik-Bilgisayar Mühendisliği Kongresi, İstanbul, pp. 63-66.
230
VURAL A.M., EKER İ., SÜSLÜOĞLU B., 2003. Doğal mıknatıslı bir DC
motorun deneysel olarak tanılaması, 10. Ulusal Elektrik-Elektronik-Bilgisayar
Mühendisliği Kongresi, İstanbul, pp. 122-125.
VURAL A.M., EKER İ., 2003. Least squares on-line identification of a dc
motor, Mühendislik Bilimleri Genç Araştırmacılar 1. Kongresi, İstanbul, pp. 183-
190.
VURAL A.M., TÜMAY M., MA T.T., 2001. Güç sistemlerindeki güç
akışının UPFC ile kontrolü, 9. Ulusal Elektrik-Elektronik-Bilgisayar Mühendisliği
Kongresi, Bursa, pp. 196-199.
231
232
APPENDIX
233
APPENDIX A: Converter Design Data for Power Flow Studies
A1. WSCC 3-Machine 9-Bus System and IEEE 14-Bus System

• Shunt/Series Converters
Component name : Three-phase voltage source model 2
Base frequency : 60 Hz
Maximum voltage : 30 kV (line-to-line rms)
Voltage ramp up time : 0.05 s
• Shunt Coupling Magnetic Interface
Component name : Three-phase two-winding transformer (x1)
Base power : 100 MVA
Winding voltage : 330 kV / 30 kV (Y/Δ) ()
Leakage reactance : 0.01 pu
• Series Coupling Magnetic Interface
Component name : Single-phase two-winding transformer (x3)
Base power : 33.3333 MVA
Winding voltage : 15.4 kV / 90 kV (line-to-line rms)
A2. 3-Machine 7-Bus System

• Shunt Converter
Component name : Three-phase voltage source model 2
Maximum voltage : 30 kV (line-to-line rms)
Voltage ramp up time : 0.05 s
• Shunt Coupling Magnetic Interface
Component name : Three-phase two-winding transformer (x1)
Base power : 100 MVA
Winding voltage : 170 kV / 30 kV (Y/Δ) (line-to-line rms)
234
APPENDIX B: Test Systems Data
B1. WSCC 3-Machine 9-Bus System (230 kV, 60 Hz, 100 MVA base)
Generation Data:
Generator No: Location Voltage: Real MW
1 Bus 1 16.5 kV Swing Bus
2 Bus 2 18.0 kV 163
3 Bus 3 13.8 kV 85
Transformer Data:
Transformer No: Location Tap: X (pu)
1 Bus 1-Bus 4 16.5 / 230 kV 0.0576
2 Bus 2-Bus 7 18.0 / 230 kV 0.0625
3 Bus 3-Bus 9 13.8 / 230 kV 0.0586
Load Data:
Load No: Location Real MW Reactive MVAR
1 Bus 5 125 50
2 Bus 6 90 30
3 Bus 8 100 35
Line Data:
Line No: Location: R (pu) X (pu) B/2 (pu)
1 Bus 4-Bus 5 0.01000 0.08500 0.04400
2 Bus 4-Bus 6 0.01700 0.09200 0.03950
3 Bus 5-Bus 8 0.03200 0.16100 0.07650
4 Bus 6-Bus 9 0.03900 0.17000 0.08950
5 Bus 7-Bus 8 0.00850 0.07200 0.03725
6 Bus 8-Bus 9 0.01190 0.10080 0.05225
235
B2. IEEE 14-Bus System (230 kV, 60 Hz, 100 MVA base)
Generation and Load Data:
Bus Generation: Bus Load:
Bus No:
Real Reactive Real Reactive
MW MVAR MW MVAR
1 232.4 -16.9 0.0 0.0
2 40.0 42.4 21.7 12.7
3 0.0 23.4 94.2 19.0
4 0.0 0.0 47.8 -3.9
5 0.0 0.0 7.6 1.6
6 0.0 12.2 11.2 7.5
7 0.0 0.0 0.0 0.0
8 0.0 17.4 0.0 0.0
9 0.0 0.0 29.5 16.6
10 0.0 0.0 9.0 5.8
11 0.0 0.0 3.5 1.8
12 0.0 0.0 6.1 1.6
13 0.0 0.0 13.5 5.8
14 0.0 0.0 14.9 5.0
Line Data:
1 Bus 1-Bus 2 0.01938 0.05917 0.02640
2 Bus 2-Bus 3 0.04699 0.19797 0.02190
3 Bus 2-Bus 4 0.05811 0.17632 0.01870
4 Bus 1-Bus 5 0.05403 0.22304 0.02460
5 Bus 2-Bus 5 0.05695 0.17388 0.01700
6 Bus 3-Bus 4 0.06701 0.17103 0.01730
7 Bus 4-Bus 5 0.01335 0.04211 0.00640
8 Bus 7-Bus 8 0.00000 0.17615 0.00000
9 Bus 7-Bus 9 0.00000 0.11001 0.00000
10 Bus 9-Bus 10 0.03181 0.08450 0.00000
11 Bus 6-Bus 11 0.09498 0.19890 0.00000
12 Bus 6-Bus 12 0.12291 0.25581 0.00000
13 Bus 6-Bus 13 0.06615 0.13027 0.00000
14 Bus 9-Bus 14 0.12711 0.27038 0.00000
15 Bus 10-Bus 11 0.08205 0.19207 0.00000
16 Bus 12-Bus 13 0.22092 0.19988 0.00000
17 Bus 13-Bus 14 0.17093 0.34802 0.00000
Transformer Data:
Transformer No: Location: Tap: X (pu)
1 Bus 4-Bus 7 0.978 0.20912
2 Bus 4-Bus 9 0.969 0.55618
3 Bus 5-Bus 6 0.932 0.25202
• Condenser is connected at Bus 8 to regulate bus voltage at 1.09 pu
• Shunt capacitance of 2.6465 µF is connected at Bus 9
236
B3. 3-Machine 7-Bus System (154 kV, 50 Hz, 100 MVA base)
Generation and Load Data:
Bus Generation: Bus Load:
Bus No:
Real Reactive Real Reactive
MW MVAR MW MVAR
1 317.8 369.4 0.0 0.0
2 50.0 0.0 200 150
3 0.0 0.0 0.0 0.0
4 50.0 0.0 200 150
5 0.0 0.0 0.0 0.0
6 0.0 0.0 0.0 0.0
7 0.0 0.0 0.0 0.0
Line Data:
1 Bus 1-Bus 2 0.0075 0.1324 0.04340
2 Bus 1-Bus 5 0.00000 0.00099 0.00000
3 Bus 1-Bus 4 0.00476 0.08390 0.02750
4 Bus 2-Bus 3 0.00527 0.09276 0.03030
5 Bus 2-Bus 3 0.00527 0.09276 0.03030
6 Bus 3-Bus 4 0.00527 0.09276 0.03030
7 Bus 3-Bus 4 0.00527 0.09276 0.03030
8 Bus 2-Bus 6 0.00000 0.00020 0.00000
9 Bus 4-Bus 7 0.00000 0.00020 0.00000
B4. 4-Machine 4-Bus System (154 kV, 50 Hz, 100 MVA base)
G1 and G3 terminal voltage is 1.0 pu with a phase shift of 0.0º. G2 and G4 terminal
voltage is 0.974 pu with a phase shift of 10.0º. Series inductive reactances of all
generators are 0.0265 pu. Each transmission line is identical having
resistance=0.01938 pu, inductive reactance=0.05917 pu, susceptance =0.0528 pu.
B5. Wind Farm Integrated Power System (154 kV, 60 Hz, 100 MVA)
Wind turbine parameters: 2.5 MVA, rated angular mechanical speed= 20 Hz, ρ =
1.225 kg/m3, A= 5026 m2 with a rotor radius of 40 m, gear box efficiency = 97 %,
gear ratio (machine/turbine) = 55.
SEDCIG parameters: 2.5 MVA, 0.86 pf lagging (without fixed capacitors),
VLL=13.8 kV, base angular frequency=60 Hz, Rs = 0.066 pu, Rr1= 0.298 pu,
Rr2=0.018 pu, Ls = 0.046 pu, Lm = 3.86 pu, Lr1=0.122 pu, Lr2 = 0.105 pu, J=2H=3.40
237
s, mechanical damping = 0.01 pu. 20 SEDCIGs are operated in parallel to give 50
MVA output.
SG parameters: 120 MVA, VLL=13.8 kV, base angular frequency=60 Hz, pf=
0.9957, H = 3.117 s, mechanical windage and friction loss = 0.04 pu, iron loss = 300
pu, Ra = 0.0025 pu (armature time constant, Ta = 0.278 s), Xd = 1.014 pu, X'd = 0.314
pu, T'd0 = 6.55 s, X''d = 0.280 pu, T''d0 = 0.039 s, Xq = 0.770 pu, X''q = 0.375 pu T''q0
= 0.071 s, potier reactance Xp = 0.163 pu, air gap factor = 1.0, number of Q-axis
damper windings = 1.
IEEE type 2 hydro governor and turbine parameters: for controller: real pole
gain = 0.88, proportional gain = 3.7, integral gain = 0.44, real pole time constant =
0.02 s, Turbine lead time constant = 0.01 s, turbine lag time constant =0.01 s,
governor time constant =0.05 s, inverse gate velocity limit =4.8 s/pu, gate velocity
time constant =0.1 s, permanent droop gain =0.08, gate position control rate limit =
0.22 pu/s, temporary droop gain = 0.0, temporary droop time constant = 0.01 s,
conversion constant = 0.895, time constant for smoothing = 0.02 s.
IEEE type SCRX solid state exciter parameters: VLN = 7967 V, line current=5020
A, rectifier smoothing time constant = 0.02 s, controller lead/lag time constant =
1.5/1.0 s, exciter time constant = 0.02 s, exciter gain = 100 pu, min/max field voltage
= -+5 pu, reverse resistance = 15 KΩ.
B6. Two-Area System (154 kV, 60 Hz, 100 MVA)

Line Data
Line No: Location: R (Ω/m) X (Ω/m) B (MΩ/m) Length (m)
1 Bus 1-Bus 4 0.178159E-4 0.31388E-3 273.5448 50E3
2 Bus 1-Bus 6 0.178159E-4 0.31388E-3 273.5448 100E3
3 Bus 1-Bus 2 0.178159E-4 0.31388E-3 273.5448 350E3
4 Bus 1-Bus 2 0.178159E-4 0.31388E-3 273.5448 350E3
238
APPENDIX C: PI Controller Parameters
• Power Flow Studies (Chapter 3)
WSCC 3-Machine 9-Bus System

Proportional gain-Kp, integral time constant -τi
STATCOM (Figure 3.4a)
Voltage regulator : 0.001, 0.008
Real power balance regulator : 0.1, 0.0004
SSSC (Figure 3.4b)
UPFC (Figure 3.4c)
Shunt VSC
Series VSC
Series reactive power regulator : 0.1, 0.008
IPFC (Figure 3.4d)
Series VSC-1
Series VSC-2
Series reactive power regulator : 0.1, 0.008
GUPFC (Figure 3.4f)
Shunt VSC
Series VSCs
Series reactive power regulator-1 and 2 : 0.1, 0.008
IEEE 14-Bus System

UPFC (for both UPFCs) (Figure 3.4c)
Shunt VSC
Series VSC
239
Real power flow regulator : 0.06, 0.002
Reactive power flow regulator : 0.001, 0.008
IPFC (Figure 3.4d)
Series VSC-1
Reactive power flow regulator : 0.001, 0.004
Series VSC-2
Reactive power regulator : 0.0001, 0.002
GUPFC (Figure 3.4f)
Shunt VSC
Series VSC-1
Series VSC-2
3-Machine 7-Bus System

STATCOM (for both STATCOMs) (Figure 3.4a)
BtB-STATCOM (Figure 3.4e)
Shunt VSC-1
Shunt VSC-2
Real power transfer regulator : 0.01, 0.08
• Converter-Level Modeling Studies (Chapter 5)
3-Machine 7-Bus System

BtB-STATCOM (Figure 5.33)
Shunt VSC-1
DC link voltage controller (Figure 5.33a) : 0.8, 0.001
AC voltage controller (Figure 5.33b) : 0.1, 0.01
240
Shunt VSC-2
Real power transfer controller (Figure 5.33c): 0.8, 0.01
AC voltage controller (Figure 5.33d) : 0.8, 0.001
• Transient Stability Studies (Chapter 6)
Wind Farm Integrated Power System

GUPFC (Figure 6.1)
Shunt VSC
DC link voltage controller (Figure 5.33a) : 0.8, 0.01
AC voltage controller (Figure 5.33b) : 8, 0.02
Lower Series VSC
Real power flow controller : 1.0, 0.001
Reactive power controller : 1.0, 0.01
Damping gain (Kw) : 500
Upper Series VSC
Reactive power controller : 1.0, 0.01
Two-Area System
IPFC (Figure 6.13)
Lower Series VSC
DC link voltage controller : 0.1, 0.001
Real power controller : 0.2, 0.001
Upper Series VSC
SSSC Controllers (Figure 6.13)
Series VSC
DC link voltage controller : 0.1 0.001
Real power controller : 0.2, 0.001
SMIB System
BtB-STATCOM (Figure 6.21)
Series VSC1
AC voltage controller : 0.8, 0.01
241
Real power transfer controller : 0.8, 0.01
Series VSC2
AC voltage controller : 0.8, 0.01
DC voltage controller : 0.1, 0.1
APPENDIX D: Derivation of Maximum Power Injections for BtB-STATCOM
(Chapter 3)
The following derivations are made for ensuring maximum real power
transfer and maximum reactive power compensation at the same time for BtB-
STATCOM.
For VSC1 (loss meeting function is assigned);
( Pinj1 max 2 + Qinj1 max 2 ) ≤ 1.0 (D.1)
When real power transfer from VSC1 to VSC2

Pinj1max = - 0.003 + 0.7071 = 0.7041 pu
Qinj1 max = + 0.7101 pu (71.01 % capacitive compensation)
When real power transfer from VSC2 to VSC1

Pinj1max = - 0.003 - 0.7071 = - 0.7101 pu
Qinj1 max = + 0.7041 pu (70.41 % capacitive compensation)
For VSC4;
( Pinj 2 max 2 + Qinj 2 max 2 ) ≤ 1.0 (D.2)
Pinj 2 max = Qinj 2 max = 1 / 2 = 0.7071 pu
Pinj 2 max = - 0.7071 pu (when real power transfer from VSC1 to VSC2)
Pinj 2 max = + 0.7071 pu (when real power transfer from VSC2 to VSC1)
Qinj 2 max = + 0.7071 pu (70.71 % capacitive compensation)
242
APPENDIX E: Programming Scripts
• Matlab m-file to Plot Simulation Results Obtained From PSCAD
% This m-file reads PSCAD "*.out" file and puts it in matrix form for Matlab
% written by A. Mete VURAL
% begin
% reading main "*.out" file
sepet = importdata('file_name_no.out'); % specify the output file name with number
extension obtained by saving the channels to disk in PSCAD
% reading input data
t=sepet.data(:,1); % 1st column: PSCAD simulation time
data_namesimcase=sepet.data(:,n); n is the column number of data in *.out file
% end
• Chapter 5
Fortran Script of PSCAD-MATLAB Interface

#STORAGE REAL:5
IF($Enabl.GT.0.9) THEN
STORF(NSTORF) = $sig_inp(1)
STORF(NSTORF+1) = $sig_inp(2)
CALL MLAB_INT("$Path", "$Name", "R R R R", "R R")
$sig_out(1) = STORF(NSTORF+4)
END IF
NSTORF = NSTORF + 6
MATLAB m-file Calling Fuzzy Decoupler (FUDE)

function [E F] = skynet2(A,B,C,D)
fuzdec_3b = readfis('fuzdec_3b');
[E F] = evalfis([A B C D], fuzdec_3b);
Fuzzy Inference System in Matlab for FUDE

[System]
243
Name='fuzdec_3b'
Type='mamdani'
Version=2.0
NumInputs=4
NumOutputs=2
NumRules=98
AndMethod='min'
OrMethod='max'
ImpMethod='min'
AggMethod='max'
DefuzzMethod='centroid'
[Input1]
Name='perrdot'
Range=[-33 33]
NumMFs=7
MF1='n3':'trimf',[-44 -33 -22]
MF2='n2':'trimf',[-33 -22 -11]
MF3='n1':'trimf',[-22 -11 0]
MF4='z':'trimf',[-11 0 11]
MF5='p1':'trimf',[0 11 22]
MF6='p2':'trimf',[11 22 33]
MF7='p3':'trimf',[22 33 44]
[Input2]
Name='perr'
Range=[-21 21]
NumMFs=7
MF1='n3':'trimf',[-28 -21 -14]
MF2='n2':'trimf',[-21 -14 -7]
MF3='n1':'trimf',[-14 -7 0]
MF4='z':'trimf',[-7 0 7]
MF5='p1':'trimf',[0 7 14]
MF6='p2':'trimf',[7 14 21]
MF7='p3':'trimf',[14 21 28]
[Input3]
Name='qerrdot'
Range=[-40 40]
NumMFs=7
MF1='n3':'trimf',[-53.33 -40 -26.67]
MF2='n2':'trimf',[-40 -26.67 -13.33]
MF3='n1':'trimf',[-26.67 -13.33 1.776e-015]
MF4='z':'trimf',[-13.33 -4.441e-016 13.33]
MF5='p1':'trimf',[1.776e-015 13.33 26.67]
MF6='p2':'trimf',[13.33 26.67 40]
MF7='p3':'trimf',[26.67 40 53.33]
[Input4]
Name='qerr'
Range=[-35 35]
NumMFs=7
MF1='n3':'trimf',[-46.67 -35 -23.33]
MF2='n2':'trimf',[-35 -23.33 -11.67]
MF3='n1':'trimf',[-23.33 -11.67 -1.776e-015]
MF4='z':'trimf',[-11.67 1.11e-016 11.67]
244
MF5='p1':'trimf',[-1.776e-015 11.67 23.33]
MF6='p2':'trimf',[11.67 23.33 35]
MF7='p3':'trimf',[23.33 35 46.67]
[Output1]
Name='delvq'
Range=[-80 80]
NumMFs=7
MF1='n3':'trimf',[-106.7 -80 -53.32]
MF2='n2':'trimf',[-80 -53.32 -26.64]
MF3='n1':'trimf',[-53.32 -26.64 0]
MF4='z':'trimf',[-26.64 0 26.68]
MF5='p1':'trimf',[0 26.68 53.32]
MF6='p2':'trimf',[26.68 53.32 80]
MF7='p3':'trimf',[53.32 80 106.7]
[Output2]
Name='delvd'
Range=[-400 400]
NumMFs=7
MF1='n3':'trimf',[-533 -400 -266.8]
MF2='n2':'trimf',[-400 -266.8 -133.3]
MF3='n1':'trimf',[-266.8 -133.3 0]
MF4='z':'trimf',[-133.3 0 133.3]
MF5='p1':'trimf',[0 133.3 266.8]
MF6='p2':'trimf',[133.3 266.8 400]
MF7='p3':'trimf',[266.8 400 533]
[Rules]
7 7 0 0, 7 0 (1) : 1 0 0 7 7, 0 7 (1) : 1
7 6 0 0, 7 0 (1) : 1 0 0 7 6, 0 7 (1) : 1
7 5 0 0, 7 0 (1) : 1 0 0 7 5, 0 7 (1) : 1
7 4 0 0, 6 0 (1) : 1 0 0 7 4, 0 6 (1) : 1
7 3 0 0, 6 0 (1) : 1 0 0 7 3, 0 6 (1) : 1
7 2 0 0, 5 0 (1) : 1 0 0 7 2, 0 5 (1) : 1
7 1 0 0, 4 0 (1) : 1 0 0 7 1, 0 4 (1) : 1
6 7 0 0, 7 0 (1) : 1 0 0 6 7, 0 7 (1) : 1
6 6 0 0, 7 0 (1) : 1 0 0 6 6, 0 7 (1) : 1
6 5 0 0, 6 0 (1) : 1 0 0 6 5, 0 6 (1) : 1
6 4 0 0, 6 0 (1) : 1 0 0 6 4, 0 6 (1) : 1
6 3 0 0, 5 0 (1) : 1 0 0 6 3, 0 5 (1) : 1
6 2 0 0, 4 0 (1) : 1 0 0 6 2, 0 4 (1) : 1
6 1 0 0, 3 0 (1) : 1 0 0 6 1, 0 3 (1) : 1
5 7 0 0, 7 0 (1) : 1 0 0 5 7, 0 7 (1) : 1
5 6 0 0, 6 0 (1) : 1 0 0 5 6, 0 6 (1) : 1
5 5 0 0, 6 0 (1) : 1 0 0 5 5, 0 6 (1) : 1
5 4 0 0, 5 0 (1) : 1 0 0 5 4, 0 5 (1) : 1
5 3 0 0, 4 0 (1) : 1 0 0 5 3, 0 4 (1) : 1
5 2 0 0, 3 0 (1) : 1 0 0 5 2, 0 3 (1) : 1
5 1 0 0, 2 0 (1) : 1 0 0 5 1, 0 2 (1) : 1
4 7 0 0, 6 0 (1) : 1 0 0 4 7, 0 6 (1) : 1
4 6 0 0, 6 0 (1) : 1 0 0 4 6, 0 6 (1) : 1
4 5 0 0, 5 0 (1) : 1 0 0 4 5, 0 5 (1) : 1
4 4 0 0, 4 0 (1) : 1 0 0 4 4, 0 4 (1) : 1
4 3 0 0, 3 0 (1) : 1 0 0 4 3, 0 3 (1) : 1
4 2 0 0, 2 0 (1) : 1 0 0 4 2, 0 2 (1) : 1
245
4 1 0 0, 2 0 (1) : 1 0 0 4 1, 0 2 (1) : 1
3 7 0 0, 6 0 (1) : 1 0 0 3 7, 0 6 (1) : 1
3 6 0 0, 5 0 (1) : 1 0 0 3 6, 0 5 (1) : 1
3 5 0 0, 4 0 (1) : 1 0 0 3 5, 0 4 (1) : 1
3 4 0 0, 3 0 (1) : 1 0 0 3 4, 0 3 (1) : 1
3 3 0 0, 2 0 (1) : 1 0 0 3 3, 0 2 (1) : 1
3 2 0 0, 2 0 (1) : 1 0 0 3 2, 0 2 (1) : 1
3 1 0 0, 1 0 (1) : 1 0 0 3 1, 0 1 (1) : 1
2 7 0 0, 5 0 (1) : 1 0 0 2 7, 0 5 (1) : 1
2 6 0 0, 4 0 (1) : 1 0 0 2 6, 0 4 (1) : 1
2 5 0 0, 3 0 (1) : 1 0 0 2 5, 0 3 (1) : 1
2 4 0 0, 2 0 (1) : 1 0 0 2 4, 0 2 (1) : 1
2 3 0 0, 2 0 (1) : 1 0 0 2 3, 0 2 (1) : 1
2 2 0 0, 1 0 (1) : 1 0 0 2 2, 0 1 (1) : 1
2 1 0 0, 1 0 (1) : 1 0 0 2 1, 0 1 (1) : 1
1 7 0 0, 4 0 (1) : 1 0 0 1 7, 0 4 (1) : 1
1 6 0 0, 3 0 (1) : 1 0 0 1 6, 0 3 (1) : 1
1 5 0 0, 2 0 (1) : 1 0 0 1 5, 0 2 (1) : 1
1 4 0 0, 2 0 (1) : 1 0 0 1 4, 0 2 (1) : 1
1 3 0 0, 1 0 (1) : 1 0 0 1 3, 0 1 (1) : 1
1 2 0 0, 1 0 (1) : 1 0 0 1 2, 0 1 (1) : 1
1 1 0 0, 1 0 (1) : 1 0 0 1 1, 0 1 (1) : 1
• Chapter 6
Fortran Script of PSCAD-MATLAB Interface (for GUPFC and IPFC)

#STORAGE REAL:3
IF($Enabl.GT.0.9) THEN
STORF(NSTORF) = $sig_inp(1)
CALL MLAB_INT("$Path", "$Name", "R R", "R R")
END IF
NSTORF = NSTORF + 4
MATLAB m-file Calling Self-Tuning Fuzzy Damping Controller (STFDC)

(for GUPFC and IPFC)
function [C,D] = osc_damp1(A,B)
strb1 = readfis('strb1');
strb2 = readfis('strb2');
C = evalfis([A B], strb1);
D = evalfis([A B], strb2);
Fuzzy Inference System in Matlab for Fuzzy Damping Controller (FDC)

[System]
246
Name='strb1'
Type='mamdani'
Version=2.0
NumInputs=2
NumOutputs=1
NumRules=49
AndMethod='min'
OrMethod='max'
ImpMethod='min'
AggMethod='max'
[Input1]
Name='perrdot'
Range=[-1 1]
NumMFs=7
MF1='n3':'trimf',[-10 -1 -0.6665]
MF2='n2':'trimf',[-1 -0.6665 -0.3334]
MF3='n1':'trimf',[-0.6665 -0.3334 0]
MF4='z':'trimf',[-0.3334 0 0.3334]
MF5='p1':'trimf',[0 0.3334 0.6665]
MF6='p2':'trimf',[0.3334 0.6665 1]
MF7='p3':'trimf',[0.6665 1 10]
[Input2]
Name='perr'
Range=[-1 1]
NumMFs=7
MF1='n3':'trimf',[-15 -1 -0.6666]
MF2='n2':'trimf',[-1 -0.6666 -0.3332]
MF3='n1':'trimf',[-0.6666 -0.3332 0]
MF4='z':'trimf',[-0.3332 0 0.3332]
MF5='p1':'trimf',[0 0.3332 0.6667]
MF6='p2':'trimf',[0.3332 0.6667 1]
MF7='p3':'trimf',[0.6667 1 15]
[Output1]
Name='delvq'
Range=[-0.4 0.4]
NumMFs=7
MF1='n3':'trimf',[-0.5336 -0.4 -0.2666]
MF2='n2':'trimf',[-0.4 -0.2666 -0.1332]
MF3='n1':'trimf',[-0.2666 -0.1332 0]
MF4='z':'trimf',[-0.1332 0 0.1334]
MF5='p1':'trimf',[0 0.1334 0.2666]
MF6='p2':'trimf',[0.1334 0.2666 0.4]
MF7='p3':'trimf',[0.2666 0.4 0.5336]
[Rules]
7 7, 7 (1) : 1 4 3, 3 (1) : 1
7 6, 7 (1) : 1 4 2, 2 (1) : 1
7 5, 7 (1) : 1 4 1, 1 (1) : 1
7 4, 6 (1) : 1 3 7, 6 (1) : 1
7 3, 5 (1) : 1 3 6, 5 (1) : 1
7 2, 5 (1) : 1 3 5, 4 (1) : 1
7 1, 4 (1) : 1 3 4, 3 (1) : 1
247
6 7, 7 (1) : 1 3 3, 3 (1) : 1
6 6, 6 (1) : 1 3 2, 2 (1) : 1
6 5, 6 (1) : 1 3 1, 1 (1) : 1
6 4, 6 (1) : 1 2 7, 5 (1) : 1
6 3, 5 (1) : 1 2 6, 4 (1) : 1
6 2, 4 (1) : 1 2 5, 3 (1) : 1
6 1, 3 (1) : 1 2 4, 2 (1) : 1
5 7, 7 (1) : 1 2 3, 2 (1) : 1
5 6, 6 (1) : 1 2 2, 2 (1) : 1
5 5, 5 (1) : 1 2 1, 1 (1) : 1
5 4, 5 (1) : 1 1 7, 4 (1) : 1
5 3, 4 (1) : 1 1 6, 3 (1) : 1
5 2, 3 (1) : 1 1 5, 3 (1) : 1
5 1, 2 (1) : 1 1 4, 2 (1) : 1
4 7, 7 (1) : 1 1 3, 1 (1) : 1
4 6, 6 (1) : 1 1 2, 1 (1) : 1
4 5, 5 (1) : 1 1 1, 1 (1) : 1
4 4, 4 (1) : 1
Fuzzy Inference System in Matlab for Fuzzified Gain Tuner (FGT)

[System]
Name='strb2'
Type='mamdani'
Version=2.0
NumInputs=2
NumOutputs=1
NumRules=49
AndMethod='min'
OrMethod='max'
ImpMethod='min'
AggMethod='max'
[Input1]
Name='errdot'
Range=[-1 1]
NumMFs=7
MF1='n3':'trimf',[-10 -1 -0.6665]
MF2='n2':'trimf',[-1 -0.6665 -0.3334]
MF3='n1':'trimf',[-0.6665 -0.3334 0]
MF4='z':'trimf',[-0.3334 0 0.3334]
MF5='p1':'trimf',[0 0.3334 0.6665]
MF6='p2':'trimf',[0.3334 0.6665 1]
MF7='p3':'trimf',[0.6665 1 10]
[Input2]
Name='err'
Range=[-1 1]
NumMFs=7
MF1='n3':'trimf',[-15 -1 -0.6666]
MF2='n2':'trimf',[-1 -0.6666 -0.3332]
248
MF3='n1':'trimf',[-0.6666 -0.3332 0]
MF4='z':'trimf',[-0.3332 0 0.3332]
MF5='p1':'trimf',[0 0.3332 0.6667]
MF6='p2':'trimf',[0.3332 0.6667 1]
MF7='p3':'trimf',[0.6667 1 15]
[Output1]
Name='alfa'
Range=[0 1]
NumMFs=7
MF1='z':'trimf',[-0.167 0 0.1668]
MF2='vs':'trimf',[0 0.1668 0.3335]
MF3='s':'trimf',[0.1668 0.3335 0.5]
MF4='sb':'trimf',[0.3335 0.5 0.6667]
MF5='mb':'trimf',[0.5 0.6667 0.8333]
MF6='b':'trimf',[0.6667 0.8333 1]
MF7='vb':'trimf',[0.835945502645503 1.0026455026455 1.1696455026455]
[Rules]
7 7, 7 (1) : 1 4 3, 5 (1) : 1
7 6, 7 (1) : 1 4 2, 4 (1) : 1
7 5, 7 (1) : 1 4 1, 3 (1) : 1
7 4, 6 (1) : 1 3 7, 2 (1) : 1
7 3, 4 (1) : 1 3 6, 3 (1) : 1
7 2, 3 (1) : 1 3 5, 2 (1) : 1
7 1, 1 (1) : 1 3 4, 7 (1) : 1
6 7, 7 (1) : 1 3 3, 6 (1) : 1
6 6, 7 (1) : 1 3 2, 5 (1) : 1
6 5, 6 (1) : 1 3 1, 7 (1) : 1
6 4, 6 (1) : 1 2 7, 2 (1) : 1
6 3, 5 (1) : 1 2 6, 3 (1) : 1
6 2, 3 (1) : 1 2 5, 5 (1) : 1
6 1, 2 (1) : 1 2 4, 6 (1) : 1
5 7, 7 (1) : 1 2 3, 6 (1) : 1
5 6, 5 (1) : 1 2 2, 7 (1) : 1
5 5, 6 (1) : 1 2 1, 7 (1) : 1
5 4, 7 (1) : 1 1 7, 1 (1) : 1
5 3, 2 (1) : 1 1 6, 3 (1) : 1
5 2, 3 (1) : 1 1 5, 4 (1) : 1
5 1, 2 (1) : 1 1 4, 6 (1) : 1
4 7, 3 (1) : 1 1 3, 7 (1) : 1
4 6, 4 (1) : 1 1 2, 7 (1) : 1
4 5, 5 (1) : 1 1 1, 7 (1) : 1
4 4, 1 (1) : 1
249

8830 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

8830 PDF

Uploaded by

Copyright:

Available Formats

ÇUKUROVA UNIVERSITY

INSTITUTE OF NATURAL AND APPLIED SCIENCES

Ahmet Mete VURAL

MODELING OF MULTI-CONVERTER FACTS (FLEXIBLE

DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING

MODELING OF MULTI-CONVERTER FACTS (FLEXIBLE ALTERNATING

Ahmet Mete VURAL

DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING

……………….................................. …………….……........... ……………………….…...

Prof. Dr. Selahattin SERİN

MODELING OF MULTI-CONVERTER FACTS (FLEXIBLE

Ahmet Mete VURAL

Supervisor: Assoc. Prof. Dr. K. Çağatay BAYINDIR

Multi-converter FACTS devices can increase control flexibility of power

Key Words: FACTS Devices, Modeling, Simulation.

ÇOK KONVERTÖRLÜ ESNEK ALTERNATİF AKIM İLETİM

Ahmet Mete VURAL

1. Danışman: Doç. Dr. K. Çağatay BAYINDIR

Çok konvertörlü FACTS cihazları, güç sistemlerinin denetim esnekliğini,

Anahtar Kelimeler: FACTS Cihazları, Modelleme, Simülasyon.

I would like to express my sincere gratitude to my co-supervisor Prof. Dr.

Figure 2.1. Real power transfer between two buses ................................................... 7

With fast technological advances and rapid population growth, electrical

1.1. Motivation for the Thesis

1.2. Objectives of the Thesis

The objectives of the thesis are as follows:

• To provide appropriate models for both single- and multi-converter

1.3. Contributions of the Thesis

The contributions of this work are as follows:

1.4. General Outline

This thesis is structured in seven chapters. After this introductory chapter,

steady-state performance comparison is also made between the mentioned FACTS

2. OVERVIEW OF FACTS DEVICES

2.1. Background on Alternating Current Power Transmission

2.1.1. Thermal Limit

Thermal limit of a transmission line is defined in terms of the maximum

2.1.2. Maximum Power Transfer

Electrical power transfer from generation side to consumer side is preferred in

Figure 2.1. Real power transfer between two buses

Figure 2.2. Power-angle curve of a transmission line

2.1.3. Angle Stability

2.1.4. Voltage Stability

Voltage stability is defined as “the ability of a power system to maintain

Figure 2.3. Voltage-power characteristics of Figure 2.1 (Kundur, 1994)

Figure 2.4. Reactive power-voltage curves of Figure 2.1 (Kundur, 1994)

2.1.5. Transmission Line Loadability Characteristics

Reliable and efficient power transfer capability of a transmission line is

Figure 2.5. Loadability characteristics of transmission lines (Zhang et al., 2006)

2.2. Classification of FACTS Devices

Inherent limitations on AC power transmission mentioned in the previous

• Shunt FACTS devices

Depending on the switching properties of the power semiconductor devices,

• First generation or conventional FACTS devices with response times of

Second generation FACTS devices usually employ voltage source converter

• They employ self-commutated inverters (converters) as synchronous

Second generation FACTS devices can be further classified according to the

• Single-converter FACTS devices

Combining the above mentioned classification configurations, major FACTS

2.2.1. Static VAR Compensator (SVC)

SVC is a static VAR generator or absorber by injecting capacitive or

Table 2.1. Overview of major FACTS devices with their attributes

oscillation damping, SSR

Real power transfer, voltage