Effective December 6, 2006, this report has been made publicly available in accordance with Section 734.

3(b)(3) and published in accordance with Section 734.7 of the U.S. Export Administration Regulations. As a result of this publication, this report is subject to only copyright protection and does not require any license agreement from EPRI. This notice supersedes the export control restrictions and any proprietary licensed material notices embedded in the document prior to publication.

EPRI Power Electronics-Based Transmission Controllers Reference Book (The Gold Book)
2005 Progress Report
1010635

EPRI Power Electronics-Based Transmission Controllers Reference Book (The Gold Book)
2005 Progress Book 1010635 Technical Update, December 2005

EPRI Project Manager A. Edris

ELECTRIC POWER RESEARCH INSTITUTE 3420 Hillview Avenue, Palo Alto, California 94304-1395 PO Box 10412, Palo Alto, California 94303-0813 USA 800.313.3774 650.855.2121 askepri@epri.com www.epri.com

DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES

THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM: (A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE; OR (B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT. ORGANIZATION(S) THAT PREPARED THIS DOCUMENT EPRI

This is an EPRI Technical Update report. A Technical Update report is intended as an informal report of continuing research, a meeting, or a topical study. It is not a final EPRI technical report.

NOTE
For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or e-mail askepri@epri.com. Electric Power Research Institute and EPRI are registered service marks of the Electric Power Research Institute, Inc. Copyright 2005 Electric Power Research Institute, Inc. All rights reserved.

CITATIONS
This document was prepared by EPRI This document describes research sponsored by the Electric Power Research Institute (EPRI). This publication is a corporate document that should be cited in the literature in the following manner: EPRI Power Electronics-Based Transmission Controllers Reference Book (The Gold Book): 2005 Progress Book. EPRI, Palo Alto, CA: 2005. 1010635.

iii

ABSTRACT
EPRI is sponsoring development of a first edition of the EPRI Power Electronics-Based Transmission Controllers Reference Book. The book will provide a broad overview on power electronics-based controllerswith information on historical perspectives, basic design considerations, factory testing, site installations, commissioning, operating performance, operation and maintenance, and future trends. The document will assist users in planning, developing, installing, and utilizing this technology.

CONTENTS
1 BACKGROUND......................................................................................................................1-1 Introduction ..........................................................................................................................1-1 Brief History and Overview on Power Electronics-Based Transmission Controllers............1-1 Objective of Reference Book ...............................................................................................1-4 2 APPROACH ...........................................................................................................................2-1 Introduction ..........................................................................................................................2-1 Selection of Material.............................................................................................................2-1 Proposed Organization and Table of Contents ....................................................................2-2 Composition, Uniformity, and Style Issues of the Reference Book......................................2-4 Implementation Plan ............................................................................................................2-4 3 ACTIVITIES IN 2005...............................................................................................................3-1 Introduction ..........................................................................................................................3-1 Initial Planning......................................................................................................................3-1 Development of Two Chapters.............................................................................................3-1 A DETAILED OUTLINES OF THE CHAPTERS ...................................................................... A-1 B DRAFT CHAPTER 1, ALTERNATING CURRENT (AC) TRANSMISSION SYSTEM.......... B-1 C DRAFT CHAPTER 2, VOLTAGE-SOURCED CONVERTER-BASED DC TRANSMISSION SYSTEMS................................................................................................................................. C-1 D DRAFT AUTHORS GUIDE ................................................................................................... D-1 E SAMPLE ONE-PAGE LAYOUT............................................................................................ E-1

vii

BACKGROUND
Introduction EPRI is sponsoring development of a first edition of the EPRI Power Electronics-Based Transmission Controllers Reference Book. The book will provide a broad overview on power electronics-based controllerswith information on historical perspectives, basic design considerations, factory testing, site installations, commissioning, operating performance, operation and maintenance, and future trends. The document will assist users in planning, developing, installing, and utilizing this technology. The new EPRI Power Electronics-Based Transmission Controllers Reference Book will be developed in the tradition of the landmark series of EPRI transmission reference books, first published in the 1970s and 1980s, and currently being updated. These bookson overhead line transmission, underground transmission, wind-induced conductor motion, and compact line designbrought together leading experts to compile state-of-science information on advanced research in these areas. Most notable of these volumes was the EPRI Transmission Line Reference Book: 345 kV and Above. First published in 1975 with a red cover, the book became commonly known in the industry as the EPRI Red Book. In this tradition, it is proposed that the planned book on power electronics-based controllers be printed with a gold cover and be referred to as the Gold Book. Section 1 of this report provides some brief history and overview on power electronics-based transmission controllers, and describes the books objective. Brief History and Overview on Power Electronics-Based Transmission Controllers FACTS In the late 1970s, dynamic voltage variation and rotational stability problems of AC power transmission, which had necessitated the underutilization of lines and other system assets, provided the incentives in the electric utility industry to introduce power electronics-based control for reactive compensation. A decade later, after years of supporting the development of high power electronics for reactive compensation, EPRI formalized the broad concept of Flexible AC Transmission System (FACTS). The acronym FACTS identifies alternating current transmission systems incorporating power electronics-based controllers to enhance the controllability and increase power transfer capability. The FACTS initiative was originally launched to solve emerging system problems due to restrictions on transmission line construction, and to facilitate the growing power export/import and wheeling transactions among utilities, with two main objectives: 1. To increase the power transfer capability of transmission systems. 2. To keep power flow over designated routes.
1-1

The first objective implies that power flow in a given line could be increased up to the thermal limit by forcing the necessary current through the series line impedance if, at the same time, stability of the system is maintained via appropriate real-time control of power flow during and following system faults. This objective, of course, does not mean to imply that the lines would normally be operated at their thermal limit loading (the transmission losses would be unacceptable), but this option would be available, if needed, to handle severe system contingencies. However, by providing the necessary rotational and voltage stability via FACTS controllers, instead of large steady-state margins, the normal power transfer over the transmission lines is estimated to increase significantly (in some cases up to about 50%, according to some studies conducted). The second objective implies that, by being able to control the current in a line (by, for example, changing the effective line impedance), the power flow can be restricted to selected (contracted) transmission corridors (which have been contracted and have the capacity), while parallel and loop-flows can be mitigated. It is also implicit in this objective that the primary power flow path must be rapidly changeable to an available secondary path under contingency conditions to maintain the desired overall power transmission in the system. It is easy to see that the achievement of the two basic objectives would significantly increase the utilization of existing (and new) transmission assets, and could play a major role in facilitating deregulation with minimal requirements for new transmission lines. The implementation of the above two basic objective requires the development of high-power compensators and controllers. The technology needed for these devices is high-power (multihundred Mva) electronics with real-time operating control. Transmission Compensators and Controllers In the concept of power flow, three basic types of compensators and controllers may be defined: Shunt Compensators. Shunt compensators are primarily used to regulate the transmission voltage at critical system locations under varying load and system conditions. Thus, an ideal shunt compensator is functionally a controllable ac current source that can force the necessary current through the effective line impedance to maintain the desired bus voltage by negating the voltage drop caused by the prevailing line (load) current. Series Compensators. Series compensators are primarily used to control transmission line current and thereby the transmitted power. Thus, an ideal series compensator is functionally a controllable AC voltage source that can inject the compensating voltage in-phase, or antiphase, with the prevailing voltage drop across the line impedance to increase or decrease the line current, as if the effective line impedance was changed. Phase Shifters. Phase shifters (angle regulators) and general active and reactive power flow controllers insert the compensating voltage with respect to a selected bus voltage to control the corresponding effective transmission angle. For phase shifting, the injected voltage is adjusted so as to establish the desired transmission angle between the sending- and receivingend voltages, regardless of the phase angle with respect to the line current.

1-2

Power Electronics-based Realization of Basic Transmission Compensators and Controllers There are two approaches to the realization of power electronics-based compensators and controllers: The first approach employs thyristor-switched capacitors and reactors, and tap-changing transformers. This approach employs capacitor and reactor banks with thyristor valves in traditional shunt or series circuit arrangements. The thyristor valves control the on and off periods of the fixed capacitor and reactor banks realizing a variable reactive (shunt or series) impedance. This concept has resulted in the Static Var Compensator (SVC), the ThyristorControlled Series Capacitor (TCSC), and the Thyristor-Controlled Phase Shifter. The other approach employs voltage-sourced converters as synchronous voltage sources. This approach, which uses power semiconductors with gate turn-off capability (GTO, GCT, IGBT, etc.), can emulate a synchronous voltage source that internally generates reactive power for transmission line compensation, without the use of AC capacitors or reactors. The converter, if supported by a DC power supply or energy storage device, can also exchange real power with the AC system. This concept has produced the Static Synchronous Compensator (STATCOM), the Static Synchronous Series Compensator (SSSC), the Unified Power Flow Controller (UPFC), the Interline Power Flow Controller (IPFC), and the Backto-Back Tie (BtB).

Shunt Compensators: SVC and STATCOM Static Var Compensator (SVC). A typical shunt-connected static var compensator (SVC) is composed of thyristor-switched capacitors (TSCs) and thyristor-controlled reactors (TCRs). With proper coordination of the capacitor switching and reactor control, an SVC can vary the var output continuously between the capacitive and inductive ratings of the equipment. Static Synchronous Compensator (STATCOM). The STATCOM is analogous to an ideal, rotating synchronous compensator (condenser), which is a synchronous machine generating a balanced set of (three) sinusoidal voltages at the fundamental frequency, with controllable amplitude and phase angle. This ideal machine has no inertia, its response is practically instantaneous, and it can internally generate reactive (both capacitive and inductive) power by excitation control. Series Compensators: TCSC and SSSC Thyristor-Controlled Series Capacitor (TCSC). There are two basic schemes of thyristorcontrolled series capacitors: one uses thyristor-switched capacitors, and the other employs a fixed capacitor in parallel with a thyristor-controlled reactor. In the thyristor-switched capacitor, the degree of series compensation is controlled by increasing or decreasing the number of capacitor banks in series. To accomplish this, each capacitor bank is inserted or bypassed by a thyristor valve (switch). In the fixed-capacitor, thyristor-controlled reactor, the degree of series compensation in the capacitive operating region is increased (or decreased) by increasing (or decreasing) the thyristor conduction period, and thereby the current in the TCR.

1-3

Static Synchronous Series Compensator (SSSC). In contrast to the series capacitor, the SSSC is able to maintain a constant compensating voltage in face of variable line current, or control the amplitude of the injected compensating voltage independent of the amplitude of the line current. Phase Shifters and Power Flow Controllers: TCPAR and UPFC, IPFC, BtB Thyristor-Controlled Phase Angle Regulator (TCPAR). A Thyristor-Controlled Phase Angle Regulator consists of a shunt-connected excitation transformer with appropriate taps, a series insertion transformer, and a thyristor valve arrangement connecting a selected combination of tap voltages to the secondary of the insertion transformer. The TCPAR could be applied to regulate the transmission angle to maintain balanced power flow in multiple transmission paths, or to control it so as to increase the transient and dynamic stability of the power system. Unified Power Flow Controller. The UPFC, with a shunt- and a series-coupled converter, offers complete simultaneous or individual control of all three basic transmissions parameters voltage, line impedance, and phase angle. Or, alternatively, it can provide independent real and reactive power flow control. With this operating flexibility, it can readily adapt to particular short-term contingencies or future system modifications. Interline Power Flow Controller. The IPFC, with two or more back-to-back connected converters, each coupled in series with a different line, provides the means to transfer real power among lines, and execute independently controllable reactive compensation for each line. This capability makes it possible to equalize power flow among lines or manage power flow according to demand and line capacity. In general, the IPFC provides a highly effective scheme for the optimized use of a multi-line transmission system. Back-to-Back Tie. In a back-to-back tie, two voltage-sourced converters are in back-to-back connection with shunt coupling at each side. This configuration provides a perfect arrangement for tying two power system asynchronously or synchronously, with fully controllable power transfer and terminal voltage regulation at both sides. Objective of Reference Book The objective of the Reference Book is to provide clearly written and factual information, primarily for utility engineers and technical managers, on power electronics-based transmission controllers. The information will include material on the controllers functional purpose, operating principles, implementation options, installation requirements, performance, and availability. The book will also summarize the accumulated practical experience gained in existing installations. For users, the book will constitute a unique reference and educational tool for the selection, specification, and application of power electronics-based controllers for improving the operation and security of transmission systems.

1-4

APPROACH
Introduction The strategy for developing the EPRI Power Electronics-Based Transmission Controllers Reference Book will be to define the scope of the subject to be covered, outline the optimum organization of the material, arrange with qualified experts to draft chapters of the book, coordinate a peer review of the chapters, edit the chapters for consistency and uniformity, and publish the book. Section 2 of this report defines the scope of the book, outlines a proposed table of contents, identifies issues related to the books composition and style, and describes an implementation plan. Selection of Material As the title of the Reference Book indicates, only transmission controllers based on power electronics technology will be considered in the book. With this stipulation, the term transmission controller (or, just controller) will refer to a high-power equipment that employs a power-semiconductor switching circuit, which, in combination with appropriate conventional passive network components (capacitors, reactors, and transformers), is able to provide one or more transmission operating and control functions. These functions include rapidly controllable shunt or series reactive compensation, transmission voltage, current or angle regulation, and through AC to DC and DC to AC conversionsophisticated power flow controls, asynchronous system interties, and DC transmission links. Power electronics-based controllers can be grouped into several categories. Two broad categories can be established according to the type of power semiconductor used: 1. Line-Commutated. Conventional thyristors provide the solid-state switches (valves) for circuit arrangements in which the switches are turned on by a control signal, but are turned off by the forcing effect of the alternating system voltage. This type of equipment is referred to as line-commutated or naturally-commutated. Devices using some type of line-commutated thyristor circuits include well-established AC network compensators, such as the Static Var Compensator (SVC) and the Thyristor Controlled Series Capacitor (TCSC), some voltage and transmission angle controllers, such as the Thyristor Controlled Voltage and Angle Regulators, as well as high-power HVDC transmission system and asynchronous interties. The line-commutated technology for high-power application was largely developed from the late 1960s through the 1970s and 1980s. 2. Self-Commutated. The other broad category of power electronics-based controllers use power semiconductors that can be both turned on and off by a control signal. This type of semiconductor switches provide much greater operating flexibility than conventional thyristors, which makes possible the realization of various self-commutated controller

2-1

circuits, with far superior operating and performance characteristics than those attainable with their line-commutated counterparts. The development of self-commutated technology for high-power utility applications started in the mid 1980s and still continues with considerable effort. The self-commutated circuit developed, and exclusively used today with different implementations for transmission applications, is the voltage-sourced converter, which provides the basic building block for a synergetic family of AC transmission controllers, as well as for a growing number of DC transmission links and system interties. Considering the history and status of line-commutated controllers, the following observations can be made: 1. The major line-commutated AC controllers, SVC and TCSC, have been broadly accepted and are still being applied. In many ways, they provide the benchmark for application performance and cost. 2. Line-commutated voltage regulators with fast response may provide the most economical solution at modest power levels, and the application prospects of line-commutated phaseshifters are still considered good by respected industry experts. 3. Line-commutated HVDC systems for bulk power transmission and asynchronous interties are firmly established, and the technology is superbly documented with many years of application and performance data. On the basis of these observations, it seems logical and appropriate to include the linecommutated compensators, SVC and TCSC, as well as the line-commutated voltage and angle regulators in the Reference Book material, but to exclude the line-commutated HVDC technology, except for possible reference comparison and data. As regards self-commutated technology, recent developments and prototype installations have conclusively demonstrated that the relatively new, self-commutated Voltage-Sourced Converterbased technology can provide practically all of the traditional AC transmission controller functions. It is also instrumental to the realization of new controllers with multifunctional characteristics unattainable by line-commutated technology. In addition, this technology fuels the expansion of DC transmission links and system interties into a whole range of new applications requiring functional and operating characteristics that only self-commutated technology can costeffectively provide. Indeed, the Voltage-Sourced Converter-based, coordinated AC and DC transmission controller technology may be the key for the effective control and optimized operation of future power systems. In spite of this technological trend, application-oriented and systematically-organized literature and data, which would sufficiently help the utilities to evaluate and operate these more complex and sophisticated controllers, are not yet readily available. For these reasons, it is logical and necessary to include in the Reference Book the whole family of Voltage-Sourced Converter-based AC controller technology, together with the akin self-commutated DC transmission system and related technology. Proposed Organization and Table of Contents Based on the scope of the book, as defined above, the following organization is proposed. First, it will be necessary to provide a common background on transmission system operation laws and
2-2

limitations. This background will set the stage for understanding the original development reasons for various controllers, and the motivation for creating new controllers with improved or different functional characteristics and added operating flexibility. To this end, Chapter 1 will be structured to provide a systematic review of the basic theory of AC transmission systems with particular focus on the operation requirements and limitations and the corresponding structural, compensation, and control needs, as well as on the conceptual solutions developed and evolved to meet those needs. Similarly, in order to explain the operation and limitations of various controllers, a general background is needed in power semiconductors, and their use in high-voltage and high-power valves, which ultimately constitutes the power electronic circuit. The DC-to-AC converter is an especially important power circuit for the realization of the modern transmission controllers. Thus, an understanding of the various practical converter implementations, their operation and performance, and the corresponding economic and cost tradeoffs is vital to understanding the operating characteristics, functional capabilities and limitations, hardware structure and installation requirements of the various Voltage-Sourced Converter-based controllers. For these reasons, Chapter 2 includes the necessary information on power semiconductors and different valves. Chapter 3 reviews the voltage-sourced converter with various implementations. Chapter 4 provides generally used terminology and approved definitions that evolved in the relevant transmission controller literature, and that will be used in the Reference Book Chapters 5 through 12 provide overviews on eight types of compensators and controllers. Chapters 13 and 14 discuss important related topicsharmonics, and considerations and guidance regarding studies, equipment selection, and specification. Chapter 15 provides a unique archive reference of existing installations in all categories of power electronics-based transmission controllers. The information will include application background and objectives, equipment description and site details with relevant technical data, operating modes and features, control functions, and achieved performance. Given the rationale just described, the proposed Table of Contents is as follows: Introduction Chapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter 5 Chapter 6 Chapter 7 Chapter 8 Chapter 9 Chapter 10 Chapter 11 Chapter 12 Chapter 13 Chapter 14 Alternating Current (AC) Transmission System Power Semiconductors and Valves AC-DC Converters Guide for and Definitions of FACTS Controllers Shunt Compensators (SVC and STATCOM) Series Compensators (TCSC and SSSC) Voltage Regulators and Phase Shifters Unified Power Flow Controllers (UPFC) The Interline Power Flow Controllers (IPFC) Convertible and Expandable Static Compensators Voltage-Sourced Converter-Based Back-to-Back (BtB) System Interties Voltage-Sourced Converter-Based DC Transmission Systems Harmonics Studies for Transmission Controller Selection and Specification Preparation
2-3

Chapter 15

Archive Reference for Selected Installations

Composition, Uniformity, and Style Issues of the Reference Book The writing of the Reference Book will adhere to several general principles: Clear Writing. The book will be written in a concise manner, with clearly composed sentences. The emphasis will be on easily understandable explanations, without the burden of unnecessary analytical proofs or lengthy derivations of formulae. Tailored to Users. The information will be tailored to potential users to understand the essential principles and issues involved with credible comparisons, sufficient practical data, and accumulated installation experience to help them in the evaluation, potential acquisition, and use of the right transmission controller for their specific applications. Self-Sufficient Chapters. Each chapter is expected to be largely self-sufficient, containing all the necessary information needed for understanding the subject matter. However, Chapters 5 through 12 will, of necessity, rely on and use the basic ideas, conceptual network, and hardware solutions described in the support material in Chapters 1 through 4 on basic principles, circuit topologies, and technical terminology relevant to transmission systems and power electronics. To maintain uniformity and consistency among chapters written by different authors, it may be necessary to set some simple ground rules regarding standard units, usage of terms, and references. A simple Authors Guide may be developed to provide common agreement in these areas (see Appendix D). Even with these stipulations, and with the best intentions of cooperation on part of the authors, it may also be useful to form an Editorial Committee, composed of two or three technical experts and a professional editor. This Committee would seek to achieve the desirable uniformity, balance, material correlation, and readability, as well as to avoid unnecessary repetitions and undesirable gaps of continuity in the text. The Committee would be responsible for the integrity, balance, and structural uniformity of the material included in the Reference Book. Implementation Plan If funding permits, development of the Reference Book will proceed in 2006. To date, two chapters have been drafted (see Appendices B and C). The remaining chapters will be written over two years2006 and 2007. Writers will be selected based on their expertise with specific areas of the book. Colleagues will also be identified to provide a peer technical review of the chapters. In addition, work on the Reference Book will be guided by the members of the EPRI Power Electronics-Based Controllers Task Force. The developed material will be prepared in draft form for review and comment by the project advisors. The principal authors may meet periodically over the two years and/or participate in regular conference calls to discuss the books direction and progress.

2-4

Once the chapters have been completed and peer-reviewed, they will be edited for spelling, punctuation, grammar, and adherence to agreed-upon guidelines for uniformity. Following that, the book will be laid out in a simple two-column style modeled after the recent publication of the EPRI Red Book (see Appendix E). EPRI will publish the book in electronic format for distribution.

2-5

ACTIVITIES IN 2005
Introduction Initial work commenced in 2005 on development of the EPRI Power Electronics-Based Transmission Controllers Reference Book.. This included developing a scope of work, outlining a table of contents, and writing two chapters. Section 3 of this report describes the initial steps taken in 2005. Initial Planning The first step in writing the Reference Book was to determine the books parameters. As noted in Section 2, it was decided to include the line-commutated compensators (SVC and TCSC), as well as the line-commutated voltage and angle regulators, but to exclude the line-commutated HVDC technology. Further, the book will include voltage-sourced converter-based AC controller technology, together with the akin self-commutated DC transmission system and related technology. With the books boundaries established, the sequence of presentation was also determined, and a draft table of contents outlined (see Section 2). The Contents includes 15 chapters covering foundational information on the technology, descriptions of eight types of controllers or compensators, related information on harmonics, studies for selections and specification, and an archive of existing installations. Development of Two Chapters Two chapters were drafted in 2005: Chapter 1, Alternating Current (AC) Transmission System. This chapter provides a systematic review of the basic theory of AC transmission systemswith historical background, the fundamental relationships of AC power transmission, the role of power electronics and FACTS, the basic types of compensators and controllers, and issues related to the future of transmission control. Chapter 12, Voltage-Sourced Converter-Based DC TransmissionVSC Transmission. This chapter offers a comprehensive overview on VSC technologyincluding differences from conventional HVDC; application areas; basic operating principles; practical implementation; basic operating limits; converter rating; practical operating characteristics; control principles and structures; installation considerations; modularity; reliability, availability, and maintainability; and application examples.

3-1

DETAILED OUTLINES OF THE CHAPTERS

INTRODUCTION CHAPTER 1 ALTERNATING CURRENT (AC) TRANSMISSION SYSTEM 1.1 Historical Perspective 1.2 Fundamental Relationships of AC Power Transmission 1.2.1 Analytical Characterization 1.2.1.1 Ideal (lossless) line 1.2.1.2 Lossless inductive line 1.2.2 Transmission Via Multiple Paths in Intertied Power Systems 1.2.3 Inherent Steady-State Limits of AC Power Transmission 1.2.4 Extension of Power Transmission Limits by Line Compensation and Power Flow Control 1.2.4.1 Shunt Compensation 1.2.4.2 Series Compensation 1.2.4.3 Transmission Angle Control 1.2.4.4 Independent Control of Active and Reactive Power Flow 1.2.5 Dynamic Limits of AC Power Systems 1.2.6 Dynamic Compensation for Stability Enhancement 1.2.6.1 Transient Stability Improvement8 1.2.6.2 Power Oscillation Damping 1.2.6.3 Increase of Voltage Stability Limit
1.3 Transmission Problems and NeedsThe Role of Power Electronics

1.3.1 Problems and needs 1.3.2 The role of power electronics 1.3.3 The Objectives of FACTS

A-1

1.4 Transmission Compensators and Controllers 1.4.1 Theoretical concepts 1.4.2 Power electronics-based realization of basic transmission compensator and controllers 1.4.2.1 Shunt Compensators: SVC and STATCOM 1.4.2.2 Series Compensators: TCSC and SSSC 1.4.2.3 Phase Shifters and Power Flow Controllers: TCPAR and UPFC, IPFC, BtB 1.5 Issues and Options of Future Transmission Control 1.6 References

A-2

CHAPTER 2 POWER SEMICONDUCTORS AND VALVES 2.1 Basic Power Semiconductor Switch Types and Their Characteristics 2.1.1 No Turn-on or Turn-off control: Diodes 2.1.2 Turn-on Control only: Conventional Thyristors 2.1.3 Turn-on and Turn-off control: Gate Turn-off (GTO) Thyristors and Various Transistors 2.2 Comparison of Basic Power Semiconductors 2.2.1 Voltage and Current Ratings 2.2.2 Turn-on and Turn-off Characteristics 2.2.3 Conduction and Switching Losses 2.2.4 Surge Capabilities 2.3 Trends in Power Semiconductor Characteristics 2.4 Functional Switch Requirements in High Power Transmission Controllers 2.5 The Need for, and Issues of Semiconductor Valves Using SeriesConnected Devices 2.5.1 Voltage vs. Current Rating for Optimal Design 2.5.2 Redundancy and Protection Issues 2.6 Design Considerations of Thyristor Valves for Reactor Control 2.6.1 Turn-on of Series-Connected Thyristors 2.6.2 Turn-off of Series-Connected Thyristors 2.6.3 Gating Requirements 2.6.4 Protection of the Valve 2.7 Design Considerations of Thyristor Valves for Capacitor Switching 2.7.1 Turn-on of Series-Connected Thyristors 2.7.2 Turn-off of Series-Connected Thyristors 2.7.3 Gating Requirements 2.7.4 Protection of the Valve 2.8 Design Considerations of GTO Thyristor Valves for Voltage-Sourced

A-3

Converters 2.8.1 Turn-on of Series-Connected GTO Thyristors 2.8.2 Turn-off of Series-Connected GTO Thyristors 2.8.3 Gating Requirements 2.8.4 Protection of the GTO Valve 2.9 Design Considerations of Transistor Based Valves for Voltage-Sourced Converters 2.9.1 Turn-on of Series-Connected Transistors 2.9.2 Turn-off of Series-Connected transistors 2.9.3 Gating Requirements 2.9.4 Protection of the Transistor Valve 2.10 Selected References

A-4

CHAPTER 3 AC-DC CONVERTERS 3.1 Operating Principles of Switching Converters 3.1.1 Concept of Switch Matrix 3.1.2 Input and Output Complementary Termination Requirement 3.1.3 Voltage- and Current-Sourced Converters 3.1.4 Line- and Self-Commutation of Switches 3.2 Role of AC-DC Converters in Transmission Systems 3.2.1 DC transmission 3.2.2 Asynchronous System Intertie 3.2.3 FACTS Controllers 3.2.4 Comparison of Converters for Transmission Applications 3.2.4.1 Voltage-Sourced vs. Current-Sourced 3.2.4.2 Self- vs. Line-Commutated 3.3 Voltage-Sourced AC-DC Converters 3.3.1 Basic Operating Principles 3.3.2 Methods of Output Voltage Synthesis 3.3.2.1 Pulse-Width Modulation (PWM) A. Sinusoidal Modulation B. Selective Harmonic elimination (Waveform notching) 3.3.2.2 Harmonic Neutralization (Multi-Pulse Structures) A. 12-Pulse Converter B. 24-Pulse Converter C. 48-Pulse Converter 3.3.2.3 Multi-Level Approach with Diode Clamp Pole A. Three-Level B. Five-Level 3.3.2.4 Multi-Level Approach with Flying Capacitor Pole Structure A. Three-Level Structure

A-5

B. Five-Level 3.3.2.5 Multi-Level Approach with Chain Circuit Structure A. Three-Level B. Five-Level 3.3.2.6 Hybrid Approaches and Structures (Harmonic Neutralization and/or Notching and/or Multi-Level and/or PWM) 3.3.3 Methods of Output Voltage Control 3.3.3.1 DC Voltage Control 3.3.3.2 Pulse-Width Variation 3.3.3.3 Pulse-Width Modulation 3.3.4 High Power Converter Structure Options and Trade-offs for High MVA Output 3.3.4.1 Single Large 2- or 3-Level Converter with Large Number of Series Connected Semiconductors 3.3.4.2 Several Single Device Module Converters in Parallel Connection 3.3.4.3 Several Single Device Module Converters in Series Connection 3.3.4.4 Hybrid Approaches: Few Module Converters with Modest Number of Series Connected Semiconductors 3.3.4.5 Practical Methods of Summing the Outputs of Module Converters 3.3.4.6 Comparison of Options 3.3.5 Converter Voltage and Current Rating Considerations and Issues in Transmission Applications 3.3.6 Voltage-Sourced Converter Protection Consideration and Issues in Transmission Applications 3.3.6.1 Output Current Limitation

A-6

3.3.6.2 Coupling Transformer Saturation 3.3.6.3 AC System Overvoltage 3.3.6.4 DC Capacitor Overvoltage 3.3.6.5 Shoot-through 3.3.7 Reliability, Availability and Redundancy Issues 3.4 Selected References

A-7

CHAPTER 4 GUIDE FOR AND DEFINITIONS OF FACTS CONTROLLERS 4.1 Basic Categories Determined by the Type of Semiconductor Valves Employed 4.1.1 Controllers Using Line-Commutated Conventional Thyristor Valves 4.1.1.1 Inherent Advantages 4.1.1.2 Inherent Disadvantages 4.1.2 Controllers Using Self-Commutated Semiconductor Valves 4.1.2.1 Inherent Advantages 4.1.2.2 Comparative Disadvantages 4.2 Categories by Functional Connection 4.2.1 Shunt-Connected Controllers 4.2.1.1 Thyristor-Controlled Type A. Thyristor-Controlled Reactor B. Thyristor-Switched Capacitor C. Static Var Compensator (SVC) D. Thyristor-Controlled Resistor E. Thyristor-Controlled Voltage Regulator (Tap Changer) 4.2.1.2 Self-Commutated (Converter-Based) Types A. Static Synchronous Compensator (STATCOM) B. Static Synchronous Compensator with Energy Storage 4.2.2 Series-Connected Controllers 4.2.2.1 Thyristor-Controlled Type A. Thyristor-Controlled Series Capacitor (TCSC) B. Thyristor-Controlled Series Reactor (TCSR) 4.2.2.2 Self-Commutated (Converter-Based) Types A. Static Synchronous Series Compensator (SSSC) B. Static Synchronous Series Compensator with Energy Storage 4.2.3 Combined Shunt- and Series-Connected Controllers 4.2.3.1 Thyristor-Controlled Type

A-8

A. Thyristor-Controlled Voltage Regulator (TCVR) B. Thyristor-Controlled Phase Angle Regulator (TCPAR) 4.2.3.2 Self-Commutated (Converter-Based) Types A. Unified Power Flow Controller (UPFC) 4.2.4 Combined Shunt- and Shunt-Connected Controllers 4.2.4.1 Thyristor-Controlled Type: Line Commutated Back-toBackConverter Asynchronous Tie 4.2.4.2 Self-Commutated Type: Self-Commutated Back-to-Back Converter Asynchronous or Synchronous Tie (BtB) 4.2.5 Combined Series- and Series-Connected Controllers 4.2.5.1 Thyristor-Controlled Type: None 4.2.5.2 Self-Commutated Type: Interline Power Flow Controller (IPFC) 4.2.6 Generic Characteristics of Thyristor-Controlled Type Controllers A. Can Realize Controllable Shunt Admittance or Series Impedance B. Reactive and Active Power Absorbed or Generated by Passive Components; Thyristor Valves Only Provide Control. C. Controlled Active Power Exchange with the System is Inherently Associated with Reactive Power Absorption 4.2.7 Generic Characteristics of Self-Commutated Types A. Can Realize Controllable Reactive Power Generator with Inherent Four-Quadrant Operation. B. Variable Reactive Power Generation/Absorption is Accomplished by the Self-Commutated Controller without Passive AC Capacitors or Reactors. C. Active Power Exchange with the System is Independent of the Reactive Power Generation or Absorption. 4.3 Categories by Main Application Function 4.3.1 Shunt Compensators: SVC and STATCOM and Related Combinations

A-9

4.3.2 Series Compensators: TCSC and SSSC and Related Combinations 4.3.3 Voltage and Angle Regulators: Thyristor Controlled Various TapChangers 4.3.4 Multi-Function Power Flow Controllers: UPFC and IFFC 4.3.5 Back-to-Back System Interties 4.4 Selected References

A-10

CHAPTER 5 SHUNT COMPENSATORS 5.1 Objectives of Shunt Compensation 5.1.1 Voltage Regulation and Prevention of Voltage Collapse 5.1.2 Stability Improvement 5.1.3 Var Flow Management 5.2 Concepts for Controllable Var Generation 5.2.1 Thyristor-Controlled and Switched Capacitors and Reactors 5.2.2 AD-DC Converters 5.3 Thyristor Controlled/Switched Schemes for Static Var Compensators (SVCs) 5.3.1 Thyristor-Controlled Reactor (TCR) 5.3.1.1 Delay Angle Control A. Basic Concept B. Delay Angle vs. Fundamental and Harmonic Relationships C. Methods of Harmonic Reduction 5.3.1.2 Basic Control Scheme for Output Control 5.3.1.3 Valve Rating and Protection 5.3.2 Thyristor-Switched Capacitor (TSC) 5.3.2.1 Issues of Capacitor Switching 5.3.2.2 Conditions for Repeated (Transient-Free) Switching 5.3.2.3 Basic Control Scheme for Output Control 5.3.3.4 Valve Rating and Protection 5.3.3 Fixed-Capacitor, Thyristor-Controlled Reactor (FC-TCR) Scheme 5.3.3.1 Basic Concept 5.3.3.2 Practical Circuit Arrangements with Harmonic Filter 5.3.3.3 System Voltage vs. Output Current Characteristic 5.3.3.4 Loss vs. Var Output Characteristic 5.3.4 Thyristor-Switched Capacitor, Thyristor-Controlled Reactor (TSC-TCR) Scheme

A-11

5.3.4.1 Basic Concept of and Reasons for the TSC-TCR Arrangement 5.3.4.2 Basic Control Scheme for Output Control 5.3.4.3 Practical Circuit Arrangements 5.3.4.4 System Voltage vs. Output Current Characteristic 5.3.4.5 Loss vs. Var Output Characteristic 5.3.5 Hybrid Circuit Arrangements with Mechanically Switched Capacitors 5.3.5.1 Basic Concept 5.3.5.2 Features and Limitations 5.4 Voltage-Sourced AC-DC Converter Schemes for Static Synchronous Compensators (STATCOMs) 5.4.1 Operational Analogy to the Rotating Synchronous Generator 5.4.2 Functional Control Scheme 5.4.3 Methods of Output Control 5.4.3.1 Internal Control of DC Voltage 5.4.3.2 Pulse-Width Variation or Modulation 5.4.4 Converter and Filter Circuit Issues and Trade-offs 5.4.5 Converter Protection 5.4.6 System Voltage vs. Output Current (V-I) Characteristic 5.4.7 Loss vs. Var Output Characteristic 5.4.8 Output V-I Characteristic Tailoring for Application Requirements 5.4.9 Combined Converter and Thyristor Controlled Schemes 5.5 Compensator System Control Scheme for Network Shunt Compensation 5.5.1 Application Requirements for Reactive Shunt Compensation 5.5.2 Regulation Loop with Adjustable Slope 5.5.3 Transfer Function and Dynamic Performance 5.5.4 Special Control Functions 5.5.4.1 Var Reserve Control 5.5.4.2 Transient Stability Enhancement

A-12

5.5.4.3 Power Oscillation Damping 5.5.5 Practical Control System Structure 5.5.5.1 Interface with the Valves 5.5.5.2 Internal Operational and External System Functional Controls 5.5.5.3 Operation Supervisor, Diagnostic Control, and Status Monitor 5.5.5.4 SCADA and Local Display 5.6 Comparison Between SVC and STATCOM Schemes 5.6.1 V-I Characteristics 5.6.2 Compensation Performance Under Small System Disturbances 5.6.3 Compensation Performance Under Large System Disturbances 5.6.4 Operation Losses 5.6.5 Real Power Compensation and Exchange 5.6.6 Installation Size and requirements 5.7 Application Examples 5.7.1 SVC Installations and Experience 5.7.2 STATCOM Installations and Experience 5.8 STATCOM for Arc Furnace Compensation 5.8.1 Electrical Characteristics of Arc Furnace Loads and Corresponding Power System Disturbances 5.8.1.1 Fluctuating and Large MW and MVar Demand 5.8.1.2 Large and Changing Load Unbalance 5.8.1.3 Generation of Large Harmonic and Other Components with Non-Synchronous Frequencies 5.8.1.4 Voltage Fluctuation and Distortion Causing Lamp Flicker, Television Disturbance and Telephone Interference 5.8.2 Compensation Requirements 5.8.2.1 Mvar Rating 5.8.2.2 DC Energy Storage Capability

A-13

5.8.2.3 Speed of Response (Frequency Bandwidth) 5.8.2.4 Harmonic Attenuation/Filtering 5.8.3 STATCOM Design Features to Meet Requirements 5.8.3.1 Power Circuit Structure 5.8.3.2 Control Approach 5.8.4 Installation Examples and Performance Obtained 5.9 Selected References

A-14

CHAPTER 6 SERIES COMPENSATORS 6.1 Objectives of Series Compensation 6.1.1 Voltage Regulation and Prevention of Voltage Collapse 6.1.2 Stability Improvement (Transient Stability and Oscillation Damping) 6.1.3 Power Flow Control 6.2 Concepts for Controllable Series Compensation 6.2.1 Thyristor-Controlled and Switched Capacitors 6.2.2 AD-DC Converters 6.3 Thyristor Controlled/Switched Schemes 6.3.1 Thyristor-Switched Series Capacitor (TSSC) 6.3.1.1 Principles of Operation A. Switching Strategy and Restrictions B. Compensating Voltage and Impedance vs. Line Current Characteristics C. Losses vs. Line Current Characteristic 6.3.1.2 Basic Control Scheme 6.3.1.3 Valve Rating and Protection 6.3.2 Thyristor-Controlled Series Capacitor (TCSC) 6.3.2.1 Principles of Operation A. Blocking Mode B. Bypass Mode C. Control in the Capacitive and Inductive Operating Regions D. Impedance Characteristics in the Subsynchronous Region E. Compensating Voltage and Impedance vs. Line Current Characteristics F. Loss vs. Line Current Characteristic G. Harmonics 6.3.2.2 Basic Operating Control Schemes

A-15

6.3.2.3 Valve Rating and Protection 6.3.3 Installation Considerations for the TSSC and TCSC 6.4 Voltage-Sourced AC-DC Converter Scheme 6.4.1 Basic Concept of the Static Synchronous Series Compensator (SSSC) 6.4.1.1 Compensating Voltage and Impedance vs. Line Current Characteristics 6.4.1.2 Loss vs. Line Current Characteristic 6.4.2 Functional Scheme for Compensation Voltage Control 6.4.2.1 Internal Control of DC Voltage 6.4.2.2 Pulse-Width Variation or Modulation 6.4.2.3 Subsynchronous Characteristics 6.4.3 Harmonics 6.4.4 System Control Scheme for Series Line Compensation Requirements 6.4.5 Converter Circuit Issues and Trade-offs 6.4.5.1 Transformer Coupling 6.4.5.2 Transformerless Implementation 6.4.6 Converter Protection 6.4.7 Tailoring the Symmetrical Capacitive and Inductive Operating Range to Application Requirements 6.4.8 Installation Considerations 6.5 Comparison Between the TCSC and SSSC 6.5.1 Compensation V-I and X-I Characteristics 6.5.2 Controllability 6.5.3 Operation Losses 6.5.4 Real Power Compensation and Exchange 6.5.5 Installation Size and Requirements 6.6 Special Application and Control Considerations for Damping Electromechanical Oscillations

A-16

6.7 Application Examples 6.6.1 TCSC Installations and Experience 6.6.2 SSSC Installations and Experience 6.8 Selected References

A-17

CHAPTER 7 VOLTAGE REGULATORS AND PHASE SHIFTERS 7.1 Objectives of Voltage Regulators and Phase Shifters 7.1.1 Voltage Regulation and Prevention of Voltage Collapse 7.1.2 Power Flow Control 7.1.3 Real and Reactive Loop Power Flow Mitigation 7.1.4 Stability Improvement (Transient Stability and Oscillation Damping) 7.2 Concepts for Voltage Regulation and Phase Shifting 7.2.1 Thyristor-Controlled/Switched Voltage Regulator (TCVR/TSVR) and Phase Angle Regulator (TCVR/TSPAR) 7.2.2 AD-DC Converter-Based Schemes 7.3 Thyristor-Controlled Voltage Regulator 7.3.1 Basic Circuit Configurations 7.3.2 Delay Angle Control A. Resistive Load B. Inductive Load C. Capacitive Load D. Delay Angle vs. Fundamental and Harmonic Relationships 7.3.3 Basic Control Scheme 7.3.4 Limitations for High Power Applications 7.4 Thyristor-Switched Circuit Schemes for High Power Voltage Regulators and Phase Shifters 7.4.1 Tap-Changer Schemes with Equal Winding Sections 7.4.2 Tap-Changer Schemes with Ternary Proportioned Winding Sections 7.4.3 Basic Control Scheme 7.4.4 Valve Rating and Protection 7.5 Voltage-Sourced AC-DC Converter Scheme 7.5.1 Application of Back-to-Back Four Quadrant Converter Structure (Note: See circuit, control, and application details at the UPFC Chapter)

A-18

7.5.1.1 In-Phase Voltage Injection (Voltage Regulator) 7.5.1.2 Quadrature Voltage Injection (Phase Shifter) 7.5.1.3 Voltage Injection at an Arbitrary Angle (UPFC) 7.6 Comparison Between the Thyristor Controlled/Switched and Converter-Based Voltage Regulator and Phase Shifter Schemes 7.6.1 Transformer and Power Circuit Rating 7.6.2 Reactive Power Demand 7.6.3 Protection Requirements 7.6.4 Operation Losses 7.6.5 Installation Requirements 7.7 Selected References

A-19

CHAPTER 8 UNIFIED POWER FLOW CONTROLLER (UPFC) 8.1 Motivation for Unified Transmission Control 8.1.1 Capability of Series Active and Reactive Power Compensation 8.1.2 Inherent Operational Flexibility 8.1.3 Optimized Hardware Utilization 8.2 Basic Concept of the Unified Power Flow Controller (UPFC) 8.2.1 Conceptual Representation 8.2.3 Conventional Model: Two Mechanically-Coupled Synchronous Machines 8.2.4 Electronic Implementation 8.3 Functional Capabilities as a Conventional Transmission Controller 8.3.1 Transmission Voltage Control and Regulation 8.3.2 Capacitive and Inductive Series Line Compensation 8.3.3 Transmission Angle Control and Regulation 8.3.4 Combined Control and Compensation Capability 8.4 Generalized Power Flow Controller: Active and Reactive Line Power Control with Simultaneous Bus voltage Regulation 8.4.1 Operating Principles and Characteristics 8.4.2 Performance Comparison to Traditional Controllers 8.4.2.1 UPFC vs. Series Reactive Compensators 8.4.2.2 UPFC vs. Phase Shifters (Angle Regulators) 8.5 Control Principles and Structures 8.5.1 Functional Control Scheme 8.5.1.1 Operating Modes and Requirements of the Shunt Converter 8.5.1.2 Operating Modes and Requirements of the Series Converter 8.5.2 Control Schemes for Active and Reactive Power Flow Control 8.5.2.1 Both Converters with DC Voltage Independent Voltage Control 8.5.2.2 Only Series Converter with Independent Voltage Control

A-20

8.5.3 Achievable Dynamic Performance 8.6 Converter Power Circuit Topology and Output Control Options 8.6.1 Harmonic Generation and Possible Filtering Requirements 8.6.1.1 Harmonic Current Injection by Shunt Converter 8.6.1.2 Harmonic Voltage Injection by Series Converter 8.6.2 Losses 8.7 Rating and Protection Considerations 8.7.1 Shunt Converter 8.7.2 Series Converter 8.7.3 Possible By-Pass Needs and Protection Approaches under Line Faults 8.8 Operation Consideration and Performance under Line Fault Conditions 8.9 Use of Phase Shifter to Tailor UPFC Rating to Given Operating Range and Performance Requirements 8.10 Application and Installation Considerations 8.10.1 Coordination Between Control and Power Circuit Reconfigurations to Implement the Required Operating Options 8.10.2 Coordination Between System Control and UPFC Control to Implement Operating Options 8.11 Applications 8.12 Application Example: AEP Inez Installation 8.12.1 Application Background and Requirements 8.12.2 Power Converter Topology and Structure 8.12.3 Functional Capabilities 8.12.4 Control System 8.12.5 Test Results 8.12.6 Operating experience 8.13 Selected References

A-21

CHAPTER 9 THE INTERLINE POWER FLOW CONTROLLER (IPFC) 9.1 Basic Concept of the Interline Power Flow Controller (IPFC) 9.1.1 Conceptual Representation 9.1.2 Electronic Implementation 9.2 Operating Principles and Characteristics 9.2.1 Active Power Exchange Between Lines with Individual Series Reactive Compensation 9.2.2 Reactive Power Flow Balancing Between Lines with Independent Active Power Flow Control 9.2.3 Complementary P-Q Characteristic 9.3 Basic Control Scheme 9.3.1 Control Principles and Constrains 9.3.2 Functional Control Structure 9.5.3 Achievable Performance 9.4 Converter Power Circuit Topology Considerations 9.4.1 Harmonic Generation 9.4.2 Losses 9.5 Rating and Protection Considerations 9.6 Installation Considerations 9.7 Installation Example: NYPA Marcy Installation 9.8 Potential Applications 9.9 Selected References

A-22

CHAPTER 10 CONVERTIBLE AND EXPANDABLE STATIC COMPENSATORS 10.1 Need for Convertibility in Evolving Utility Environment 10.2 Concept of Convertible Static Compensator (CSC) 10.2.1 Converter-Based Functional Building Block Structure 10.2.2 Structure for Selectable Application Function and Operating Mode Control 10.3 Application Example: NYPA CSC Installation at Marcy Substation 10.3.1 Application Requirements and Expected Benefits 10.3.2 Power Converter Topology and Structure 10.3.3 Functional Capabilities 10.3.4 Control System 10.3.5 Installation 10.3.6 Operating experience 10.4 Functional and Rating Expandability via Converter-Based Building Block Structure 10.4.1 Functional Expansion from Single-Line to Multi-Line Transmission Control 10.4.2 Rating Expansion by Modular Approach: Issues and Trade-offs 10.5 Selected References

A-23

CHAPTER 11 VOLTAGE-SOURCED CONVERTER-BASED BACK-TO-BACK (BtB) SYSTEM INTERTIES 11.1 Application Motivation (As Contrasted to Line-Commutated Approach) 11.1.1 Capability to Provide Capacitive and Inductive Vars for Terminal Voltage Regulation 11.1.2 Capability to Supply Weak Systems and Passive Loads 11.1.3 Capability to Operate it Both as an Asynchronous Link or a Synchronous Tie with Controllable Synchronizing Torque 11.2 Operating Principles and Characteristics 11.2.1 Conceptual Representation 11.2.2 Equivalent Circuit 11.2.3 Practical Implementation 11.3 Functional Characteristics 11.3.1 Transmittable Power vs. Reactive Power Generation/Absorption 11.3.2 Transmittable Power vs. Frequency 11.3.3 Synchronizing Torque vs. Frequency Deviation 11.4 Control Principles and Structures 11.4.1 Functional Control Schemes to Meet Application Requirements and Limitations 11.4.1 Active Power Transmission, Reactive Power and Voltage Regulation 11.4.2 Torque Control 11.4.3 Power Oscillation Damping 11.4.4 Achievable Dynamic Performance 11.5 Converter Power Circuit Topology and Output Control Options 11.5.1 Requirements and Trade-offs 11.5.1.1 Voltage vs. Current 11.5.1.2 Structure vs. MVA Rating 11.5.1.3 Modularity and Redundancy vs. Availability and Cost
A-24

11.5.2 Harmonic Generation and Filtering 11.5.3 Losses 11.6 Rating and Protection Considerations 11.6.1 Surge Rating Requirements 11.6.2 System Faults 11.6.3 Internal Converter Faults 11.7 Performance During and Following Line Faults 11.8 Installation and Insulation Coordination Considerations 11.9 Installation and Performance Comparison to Conventional LineCommutated BtB Power System Ties 11.10 BtB Application Examples 11.10.1 Application Background and Requirements 11.10.2 Power Converter Topology and Structure 11.10.3 Functional Capabilities 11.10.4 Control System 11.10.5 Test Results 11.10.6 Operating Experience 11.11 Selected References

A-25

CHAPTER 12 VOLTAGE-SOURCED CONVERTER-BASED DC TRANSMISSION SYSTEMS 12.0 Introduction 12.1 Differences in Features and Characteristics from Conventional HVDC 12.1.1 Technology 12.1.2 Operating Features 12.1.3 Application Benefits and Disadvantages 12.2 Application Areas 12.2.1 Interconnection of Geographically or Otherwise Isolated Networks 12.2.2 Interconnections between Weak Power Systems 12.2.3 Reinforcement of Weak AC Tie-Lines for Stability Improvement 12.2.4 Connection of Distant Loads (Off-Shore Oil and Gas Platforms) 12.2.5 Connection of Remote Wind-Parks 12.3 Basic Operating Principles and Characteristics 12.3.1 Conceptual Representation 12.3.2 Equivalent Circuit of a VSC Transmission Scheme 12.3.3 Steady-State Operating Characteristics of a VSC Transmission Scheme 12.3.4 Dynamic Capabilities and Application Potential A. First Swing Stability B. Rotational Oscillation Damping C. Voltage Stability and Prevention of Voltage Collapse D. Power Quality 12.4 Practical Implementation 12.4.1 Two- and Three-Level Converter Structure 12.4.1.1 Two-Level Converter 12.4.1.2 Three-Level Converter 12.4.1.3 Other Converter Arrangements 12.4.2 High-Frequency Pulse-Width-Modulation 12.4.3 High-voltage Turn-off Valves
A-26

12.4.4 Harmonic Filtering 12.4.5 Phase Reactor 12.4.6 DC Capacitor 12.4.7 Interface Transformer 12.4.8 Other VSC Transmission Substation Equipment 12.4.9 DC Cables 12.5 Basic Operational Limits 12.6 Converter Rating and Protection Considerations 12.6.1 Overvoltage Stresses and Protection 12.6.2 Overcurrent Stresses and Protection 12.7 Practical Operating Characteristics 12.7.1 Terminal Voltage and Current Relationships 12.7.2 Active Power vs. Reactive Power Generation/Absorption Capability 12.7.3 Operating Losses vs. Transmitted Power 12.8 Control Principles and Structures 12.8.1 Control Structure 12.8.1.1 I/O Interface 12.8.1.2 PQU Order Control 12.8.1.3 Current Order Control 12.8.1.4 Current Control 12.8.1.5 Firing Control 12.8.2 Control Strategies for Different Applications 12.8.2.1 Power Transmission Between Weak AC Networks 12.8.2.2 Feeding of Passive Network 12.8.2.3 Connection of a Wind Farm 12.8.2.4 Multi-terminal Schemes 12.8.3 Performance During and Following Line Faults 12.8.4 VSC Transmission Protection

A-27

12.9 Installation Considerations 12.10 Modularity 12.11 Reliability, Availability, and Maintainability 12.12 VSC Transmission Application Examples 12.12.1 Gotland, Sweden 12.12.2 Tjreborg, Denmark 12.12.3 Direct Link, Australia 12.12.4 Murray Link 12.12.5 Cross Sound, USA 12.12.6 Troll, Norway 12.12.7 Estlink, Estonia-Finland 12.13 Acknowledgment 12.14 Selected References

A-28

CHAPTER 13 HARMONICS 13.1 Harmonic Sources 13.1.1 Industrial and Consumer Loads 13.1.2 Power Electronics Based Controllers, Compensators and AC-DC Converters 13.2 Possible Effects of Harmonics 13.2.1 Initiation of Oscillation at Network Resonant Frequencies 13.2.2 Overvoltages 13.2.3 Overcurrents 13.2.4 Increased Operating Temperature, Reduced Lifetime 13.2.5 Telephone and Radio Frequency Interference 13.3 Typical Utility Specified Harmonic Limits 13.3.1 Individual and Total RMS Harmonic Voltage 13.3.2 Individual and Total RMS Injected Harmonic Current 13.3.3 Telephone Interference TIF and IT Factors 13.4 Harmonic Reduction Techniques for Power Electronics Equipment 13.4.1 Circuit Topology A. MultiPulse Structure B. Multi-Level Structure 13.4.2 Various PWM Techniques 13.4.3 Filtering 13.4.4 Combinations of the Above 13.5 Passive Filter Design Consideration 13.5.1 Network Impedance Frequency Characteristic A. Actual Measure B. Simplified Worst Case Assumption 13.5.2 Avoidance of Resonances A. Selecting Impedance Zero and Pole Frequencies

A-29

B. Damping C. Filter Attenuation vs. Frequency Characteristic 13.6 Active/Hybrid Filtering and Damping Possibilities. 13.7 Selected References

A-30

CHAPTER 14 STUDIES FOR TRANSMISSION CONTROLLER SELECTION AND SPECIFICATION PREPARATION 14.1 Main Objectives of Studies 14.1.1 Identify System Needs and Candidate Solutions 14.1.2 Identify Possible Locations for the Installation 14.1.3 Compare Technical and Economic Benefits of Candidate Solutions 14.1.4 Prepare Specification for Equipment of Selected Solution(s) 14.2 Technical Information Needed for Controller Specification 14.2.1 Equipment Type and Operating Environment 14.2.1.1 Application Objectives 14.2.1.2 Controller Type(s) Wanted (or Acceptable) 14.2.1.3 Connection Point to the Transmission System A. Bus Voltage B. Max and Min Short Ckt. MVA C. Max Unbalance (Negative Sequence) D. Max Fault Clearance Time E. Range and Rate of Frequency Change F. Transient Overvoltage Levels and Durations 14.2.1.4 Continuous MVA/MW/MVar Ratings 14.2.1.5 Normal Operating Regions 14.2.1.6 Possible Abnormal and Emergency Operating Conditions 14.2.1.7 Required Short Term Ratings and Operating Limits 14.2.2 Operating Modes, Control Functions and Performance Requirements 14.2.2.1 Identify Required Operating Modes and Control Functions 14.2.2.2 Define Steady-State Operating Characteristics 14.2.2.3 Define Dynamic and Transient Performance A. During Small Disturbances B. During and Following Large Disturbances C. Coordination of Equipment Protection with Dynamic
A-31

Performance Requirements 14.2.2.4 Local and Remote Control Operation, Alarm Monitoring, Diagnostic, and SCADA Requirements 14.2.3 Harmonic Performance 14.2.3.1 Existing Harmonic Environment at the Location 14.2.3.2 System Impedance Characteristic at the Bus (Measured or Worst Case Estimate) 14.2.3.3 Acceptable Harmonic Generation by Equipment A. Voltage B. Current C. Telephone Interference 14.2.4 Criterion for Equipment Loss Evaluation 14.2.5 Acoustic Noise Restrictions 14.2.6 Availability and Reliability Requirements 14.2.6.1 Redundancy and Modularity in the Power Electronic Circuit 14.2.6.2 Full or Partial Control Redundancy 14.2.6.3 Spare Part Short- and Long-Term Availability 14.2.7 Acceptance and Commissioning Tests 14.2.7.1 Control System Operational and Performance Tests 14.2.7.2 Factory Type Tests 14.2.7.3 On Site Commissioning Tests 14.3 Main System Studies 14.3.1 Load Flow Studies for A. Equipment Ratings (Steady-State and Short-term) and Location B. Determine Steady-State Operating Point and Range C. Determine Contingencies and Abnormal Operating Conditions 14.3.2 Small Disturbance (Steady-State Stability) Studies for A. Determining Existing System Damping B. Finding Effective Control Signal and Equipment Damping
A-32

Capability 14.3.3 Large Disturbance (Transient Stability) Studies for A. Confirming Location and Adequacy of Rating B. Determining Voltage and Power Excursion Levels at Required Damping C. Verifying Stability Improvement Controller Can Provide D. Finalizing Controller Control Mode, Control Variables, Operating Range and Limits E. Determining if There is Any Interaction Between Controller and Existing Power System Constituents 14.3.4 Voltage Stability Studies for Determining Voltage Levels and Recovery After Disturbances and System Contingencies with Existing Slow-Acting Voltage Regulators and Voltage Dependent Loads 14.3.5 Harmonic Studies for A. Determining Acceptable Levels of Voltage and Current Distortion Generated by the Controlled to Meet Utility Standards B. Detecting Possible Harmonic Interactions C. Detecting Possible Resonances with System and Controller Filters and Other Reactive Components. 14.4 Selected References

A-33

CHAPTER 15 ARCHIVE REFERENCE FOR SELECTED INSTALLATIONS Note: Whereas the purpose of the previous chapters included in the Reference Book are selfevident, that for this chapter requires some explanation. The idea is to provide potential users with systematic, summary information on noteworthy installations in all categories of power electronics-based transmission controllers. The information would include application background and objectives, equipment description and site details with relevant technical data, operating modes and features, control functions, and achieved performance. The aim of these systematically organized installation descriptions is to provide a selection of applications precedents to new users to help their equipment selection and evaluation process. The following Controller Groups, to include the individual installation descriptions, are proposed: 1. Shunt Compensators 1A Static Var Compensator (SVC) 1B Static Synchronous Compensator (STATCOM) for Transmission Applications 1C Static Synchronous Compensator (STATCOM) for Arc Furnace Compensation 2. Series Compensators 2A Thyristor Controlled/Switched Series Capacitor (TCSC and TSSC) 2B Static Synchronous Series Compensator (SSSC) 3. Thyristor Controlled Phase Shifters (TCPS) and Phase Angle Regulators (TCPAR) 4. Thyristor Controlled Voltage Regulators (TCVR) 5. Multi Function Transmission Controllers 5A Unified Power Flow Controller (UPFC) 5B Interline Power Flow Controller (IPFC) 5C Convertible Static Compensator (CSC) 6. Voltage-Sourced Converter Based Back-to-Back (BtB) Ties 7. Voltage-Sourced Converter Based DC Transmission Systems

The following general description structure is proposed for each installation within its relevant Controller group: 1. Title and Location of the Installation 2. Brief Background Description of the User Utility System 3. Reasons for, and Selection Process of the Installation

A-34

3.1 Planning and/or Contingency Studies 3.2 Reasons for, and Objectives of the Installation 3.3 Candidate Equipments for the Installation 3.4 Selection Criteria 3.5 Main Specification Requirements 3.5 Reason for the Selection Result 4. Description of the Equipment Installed 4.1 Singe Line Diagram of the Total Installation 4.2 Simplified Layout of the Installation 4.3 Rating and Structure of the Power Electronic Circuit Nominal Rating Short-Term Rating Transmission Voltage Power Converter/Circuit Output/Operating Voltage Coupling Transformer Winding Arrangement and Voltages Type of Converter/Circuit Topology Pulse Number Number of Converters/Circuits Used to Provide the Output Method Used to Combine the Outputs of Individual Converters/Circuits Type of Valves Used Type of Semiconductor Used in the Valve Number of Semiconductors used in Series and/or Parallel in the Valve Voltage and Current ratings of the Individual Semiconductors DC Capacitor (if used) Nominal DC Voltage Rating in kJ Cooling System 4.3 Functional Structure and Operating Features of the Control System 4.4 Summary of Commissioning and other Performance Tests 4.5 Results and Benefits of the Installation 4.6 Selected References

A-35

DRAFT CHAPTER 1, ALTERNATING CURRENT (AC) TRANSMISSION SYSTEM

B-1

CHAPTER 1 ALTERNATING CURRENT (AC) TRANSMISSION SYSTEM

OVERVIEW FOR THE APPLICATION OF POWER ELECTRONICS-BASED COMPENSATORS AND CONTROLLERS

1.1 Historical Perspective.1 1.2 Fundamental Relationships of AC Power Transmission...3 1.2.1 Analytical Characterization..4 1.2.1.1 Ideal (lossless) line4 1.2.1.2 Lossless inductive line...5 1.2.2 Transmission Via Multiple Paths in Intertied Power Systems8 1.2.3 Inherent Steady-State Limits of AC Power Transmission..9 1.2.4 Extension of Power Transmission Limits by Line Compensation and Power Flow Control...9 1.2.4.1 Shunt Compensation...10 1.2.4.2 Series Compensation...12 1.2.4.3 Transmission Angle Control13 1.2.4.4 Independent Control of Active and Reactive Power Flow..15 1.2.5 Dynamic Limits of AC Power Systems...16 1.2.6 Dynamic Compensation for Stability Enhancement..16 1.2.6.1 Transient Stability Improvement..18 1.2.6.2 Power Oscillation Damping.20 1.2.6.3 Increase of Voltage Stability Limit...21
1.3 Transmission Problems Electronics..21 and Needs, The Role of Power

1.3.1 Problems and needs..22 1.3.2 The role of power electronics...22 1.3.3 The Objectives of FACTS23 1.4 Transmission Compensators and Controllers...24 1.4.1 Theoretical concepts.....24 1.4.2 Power electronics-based realization of basic transmission compensator and controllers..26 1.4.2.1 Shunt Compensators: SVC and STATCOM.............................................................26 1.4.2.2 Series Compensators: TCSC and SSSC..........27 1.4.2.3 Phase Shifters and Power Flow Controllers: TCPAR and UPFC, IPFC, BtB....28

1.5 Issues and Options of Future Transmission Control.34 References.36

2 CHAPTER 1 ALTERNATING CURRENT (AC) TRANSMISSION SYSTEM

OVERVIEW FOR THE APPLICATION OF POWER ELECTRONICS-BASED COMPENSATORS AND CONTROLLERS

1.1 Historical Perspective The electric power industry began to evolve in the 1880s. Almost from the very beginning two competitive systems started to emerge: Direct current (dc) power generation and transmission strongly pursued by Thomas Edison, and alternating current (ac) power generation and transmission initiated in Europe and transformed into a practical scheme with Nikola Teslas inventions. This scheme, implemented in the U.S.A. by industrialist George Westinghouse, decisively won the early competition in 1896 when the famous Niagara hydro power generation project convincingly demonstrated viable long distance ac power transmission over a 20 mile, 11 kV high voltage line from Niagara Falls to the city of Buffalo, N.Y. The success of the prestigious Niagara project fueled the universal acceptance and rapid development of ac power systems. The key to this acceptance was the technical feasibility to step up the alternating generator voltage by highly efficient magnetic transformers for transmission in order to minimize conductor size and losses, and step it down for the consumer to meet domestic and industrial load requirements. The ac power systems with high voltage transmission capability allowed the use of remote power generation and, ultimately, also the intertying separate power systems into a large area power grid characterizing modern supply systems today. This approach is in contrast to Edisons original concept of dc power system, which, due to transmission limitations, envisioned a large number of distributed and independent dc central (generation) stations, each supplying no more than a few square miles of distribution network for local loads. The U.S. electric power industry has been one of the 20th Centurys most phenomenal growth industries. From the very beginning, the unparalleled expansion of industrial activity, rapidly growing population, and available energy sources contributed to this growth. The number of utilities reached 3,620 by 1902, and was more than 6000 by the early 1920s. Privately owned utilities dominated the production of electric power, but municipal organizations also participated in retail power distribution. Originally, the utilities typically were small and locally owned, with very limited geographical service areas. However, by the late 1920s, a few large utility holding companies controlled over 80% of the industrys power generation capacity. The largely unregulated (and often abused) holding company era was brought to an end by the Public Utility Holding Company Act of 1935, and the creation of the Federal Power Commission. The industry concentration was significantly decreased and regulation of interstate electric rates and other legislative measures were introduced. In addition, the economic depression of the 1930s prompted the government to participate in the production of electric power for regional economic development. This led to the establishment of a number of public utility entities and cooperatives, most of which were primarily distribution companies. Still by 1950, the total power supplied by the private sector decreased to about 80%. The unprecedented technological developments after the Second World War with rapid industrial growth resulted in dramatic increase in the demand of electric power and the industry capacity expanded nearly tenfold from the 1950s to the early 1970s. This huge increase of power demand was answered by major

3 expansion of generation and transmission facilities, and by the formation of regional power pools and increasing interconnection of individual power systems. The socioeconomic conditions had unexpectedly begun to change during the 1970s with the utility industry facing a set of difficult economic, environmental, and social problems. The oil embargo in the mid 1970s, public opposition to nuclear power, social focus on clean air and other environmental issues led to considerable increase in operating cost and governmental intervention. The national energy legislation, various environmental initiatives, and other restrictive regulations went into effect. Alternate energy development plans for solar, geothermal, oil shale and others were initiated. At the same time the U.S. manufacturing industry went through major restructuring: large, concentrated manufacturing facilities were closed down and production was distributed to smaller facilities at different geographic locations. This, combined with pronounced demographic changes (people moved from cold to warm climate), resulted in a considerable geographical shift in power demand. All these would have required the relocation or construction of new generation facilities and transmission lines relatively quickly to match the geographically different power demand profile and accommodate a volatile fuel cost structure. Neither the strong economic base, nor the previous freedom of action existed for utilities to adopt these conventional solutions. Indeed, the increasing public concern about environment and health, the cost and regulatory difficulties in securing the necessary rights-of-way for new projects, have often prevented or excessively delayed the construction of many generation facilities and transmission lines needed by the utilities. The problems imposed by the new socioeconomic conditions fueled the further growth of interconnection among neighboring utility systems to share power with other regional pools and be part of a growing national grid. The underlying reason for this integration has been to take advantage of the diversity of loads, changes in peak demand due to weather and time differences, the availability of different generation reserves in various geographic regions, shifts in fuel prices, regulatory changes and other factors which may manifest themselves differently in other time and geographic zones. The U.S. power system, evolving from originally isolated utility suppliers to regional power pool groups, did not have a flexible enough transmission grid to cope with the rapidly changing requirements under rapid economical and environmental changes. In the interconnected system contracted power was to be delivered some times from a distant generation site, often by wheeling it through the transmission systems of several utilities, to the designated load area. These arrangements inevitably led to uncontracted and undesired parallel- and loop-flows of power (since part of the line current from the sending-end flowed through each available parallel path in proportion to its admittance), which often overloaded some lines causing thermal and voltage variation problems. The receiving-end was also exposed to difficulties caused by the contingency loss of imported power and the consequent heavy overload condition on the local system, leading to severe voltage depression with the danger of possible voltage collapse. Also, with the growing interconnected system it was increasingly difficult to maintain the traditional (conservative) stability margins without sufficient transmission reinforcement. The voltage support and transient stability requirements of the expanding interconnected network, and the prevailing restrictions for new line construction, as well as economic considerations, led to the increasing applications of the first generation of power electronics-based the Static Var Compensators, in the late 1970s and more advanced transmission controllers in the 1990s. The Notice of Proposed Ruling (NOPR) by the Federal Energy Regulatory Commission (FERC) in 1995 for a more competitive electric industry presented a new challenge, with implied structural change, for the utility industry. The main objective of NOPR was to facilitate the development of a competitive market

4 by ensuring that wholesale buyers and sellers can reach each other through non-discriminatory, open access transmission services. The implication of open access is that power generation and transmission must be functionally unbundled. It is evident that the unbundled power system structure considerably increases the demand on the transmission network. The main economic emphasis of deregulation is to reduce the cost of electricity, that is, to minimize the cost of electric power generation by free (non-discriminatory) competition. This, as an opportunity, increases the number of independent power producers. Also, it may keep moving the geographical locations of lowest cost power generation, and vary the local generation levels, according to the relative cost of different fuels and other changing factors affecting the cost of energy production (e.g., environmental protection). The compulsory accommodation of the contracted (usually the least expensive) power by the transmission network can aggravate the parallel- and loop-flow problems, causing unpredictable line loading (thermal limits), voltage variation and the potential decrease of transient stability. The unbundling of power generation from transmission may also worsen these problems by eliminating the incentive for equipping the generators with - often costly - functional capabilities, such as the effective control of reactive power generation and absorption (high-response excitation systems, power system stabilizers, etc.) to aid power transmission. The consequent possible decrease of coordinated reactive power support provided traditionally by the generators can further aggravate the transmission problems in the areas of voltage variation, transient and dynamic stability. The potential effect of all these on the reliability and security of the overall power system, without effective counter measures, could result in economically damaging forced load shedding and blackouts. In the last few years, both the U.S.A. and several European countries have been exposed to these problems. The traditional solution to the above problems would be a massive reinforcement of the transmission network with new lines to reestablish, by the method of brute force, the conventional voltage limit, stability, and thermal margins under greatly expanded contingency scenarios. Such a major undertaking probably would not be possible due to the prevailing environmental and regulatory constraints, and likely opposition by affected public, nor would it be economically savvy. Although transmission systems do need new lines to handle the increasing loads and flexibility of energy transportation routs, these can be combined with the increasing application of power electronics-based, and real time computer-controlled, compensators and controllers to provide cost effective, high tech solutions to the prevailing problems and growing flexibility requirements with economically desired utilization of transmission assets. In the late 1980s, the Electric Power Research Institute initiated a major program for the development of various power electronics-based Controllers for the practical realization of the Flexible AC Transmission Systems (FACTS), in which these Controllers regulate transmission voltage and power flow and, through rapid control action, mitigate dynamic disturbances. As a result of this program the Thyristor-Controlled Series Capacitor (TCSC), and a new family of converter-based Controllers, the Static Synchronous Compensator (STATCOM), Static Synchronous Series Compensator (SSSC), the Unified Power Flow Controller (UPFC), the Interline Power Flow Controller (IPFC), and Back-to-Back Ties (BtB) (BtB) with four-quadrant operation were developed and successfully commissioned through the 1990s. 1.2 Fundamental Relationships of AC Power Transmission It is recognized that the majority of readers are familiar with the physical laws characterizing ac transmission lines and the various mathematical techniques may be applied in their analysis. It is not the intention here to dwell into this theory, but to establish, with a systematic review and physical explanation,

5 those limiting factors of ac power transmission which can be effectively influenced by power electronics based Controllers to increase the transmittable power, system security, and asset utilization. The main constituents of an ac power system are: generators, transmission (sub-transmission), and distribution lines, and loads, with their related auxiliary support and protection equipment. The generators are rotating synchronous machines. The transmission, sub-transmission, and distribution lines are essentially distributed parameter, dominantly reactive networks designed to operate at high, medium, and low, alternating voltages, respectively. The loads may be synchronous, non-synchronous and passive, consuming in general both active and reactive power. The modern transmission system is a complex network of transmission lines interconnecting all the generator stations and all the major loading points in the power system. These lines carry large blocks of power which generally can be routed in any desired direction on the various links of the transmission system to achieve the desired economic and performance objectives. Separate ac systems are often synchronously intertied with ac transmission lines to form a power pool in which energy can be transported between the individual systems. In this arrangement, at a given time some system may be importing and others exporting power, while some systems just providing the service of wheeling power through their transmission network to facilitate particular transactions. The power pools with individual systems can further be tied together with synchronous ac, or asynchronous dc links to form a large power grid to provide highly reliable electric power supply at minimum cost. The main characteristic of todays transmission system is an overall loop structure, as illustrated with a simple power system schematic in Figure 1.1, which provides a number of path combinations to achieve the functional versatility desired. This is in contrast to early day transmission (and present day sub-transmission and distribution systems), which were (are) mostly radial, supplying power from generator to defined loads.
Generator Transformer

Bus coding 2 L1 L2
L6
L3
3

Bus Circuit breaker

L7

To bus 6

6
Tie -line to neighboring System 1

L4

L5 Load 4
L9

L8

Load

3
Tie-line to neighboring System 2

Load

Load

To bus 10

Figure 1.1 Electric power system structure.

1.2.1 Analytical Characterization In spite of the generally complex nature of an actual power system, the basic analytical relationships of power transmission can be derived by a simple so-called two machine model, in which a sending-end generator is interconnected by a transmission line with a receiving-end generator (which is sometimes considered as an infinite power voltage-bus). For the sake of generality, the sending-end and receiving-end generators in the model may also represent two independent ac systems (each represented by a

6 hypothetical resultant machine and impedance) which are inter-tied by a transmission link for power exchange. 1.2.1.1 Ideal (lossless) line An ac transmission line is characterized by its per-mile distributed circuit parameters: the series resistance and inductance, and the shunt conductance and capacitance. The characteristic behavior of the line is primarily determined by the reactive circuit elements, the series inductance l and shunt capacitance c. With a customary lumped-element representation of the ac transmission line, the two machine transmission model is shown in Figure 1.2. (Bold-faced letters represent voltage and current phasors [rotating vectors].)

Is

line inductance line capacitance

Ir

Vs

Vr

Figure 1.2 Representation of a lossless transmission line.

The transmittable electric power of the system shown in Figure 1.2 is defined by the following equation1:
P= VsVr sin d Z 0 sin q

(1.1)

in which Vs is the magnitude of the sending-end (generator) voltage phasor, Vs Vr is the magnitude of the receiving-end (generator) voltage phasor, Vr d is the phase angle between Vs and Vr (transmission or load angle), Z0 is the surge or characteristic impedance given by
Z0 = l c

(1.2)

q is the electrical length of the line expressed in radians by

q=
2p a = ba l

(1.3)

where l is the wavelength and b is the number of complete waves per unit line length, i.e.,

b=

2p = w lc = 2pf lc l

(1.4)
6

7 and a is the length of the line. The lossless line considered exhibits an ideal power transmission characteristic at the surge impedance or natural loading, at which the transmitted power is:
P0 = V 02 Z0

(1.5)

where V0 (= Vs = Vr) is the nominal or rated voltage of the line. Conclusions on power transmission by lossless line: 1. At natural (surge impedance) loading the amplitude of the voltage remains constant and the voltage and current stay in phase with each other (but rotated together in phase) along the transmission line. Consequently, the transmission power, P0, is independent of the length of the line. At surge impedance loading the reactive power exchange within the line is in perfect balance, and the line provides its own shunt compensation. That is, the reactive power demand of the series line reactance is precisely matched by the reactive power generation of the shunt line capacitance. 2. Economic considerations and system operation requirements rarely allow surge impedance loading. At lighter loads the transmission line is capacitively over compensated. The voltage increase across the series line reactance, due to the charging current of the shunt line capacitance, is greater than the voltage drop caused by the load current. As a result, the transmission line voltage increases along the line, reaching its maximum at the mid-point. Of course, this surplus charging current also flows through the sending-end and receiving end generators (or ac systems) forcing them to absorb the corresponding (capacitive) reactive power. At greater than surge impedance loading the transmission line is under compensated. That is, the voltage increase resulting from the shunt line capacitance is insufficient to cancel the voltage drop across the series line reactance due to the load current. Therefore, the voltage along the line decreases, reaching the minimum at the mid-point. In this case, the net reactive power demand of the line (inductive) must be supplied by the sending-end and receiving-end generators. 1.2.1.2 Lossless inductive line Equation (1.1) provides a generalized expression characterizing the power transmission over a lossless, but otherwise accurately represented line. For the explanation of the major transmission issues, and for the introduction of relevant power electronics based Controller concepts, it is convenient to use an approximate form of Equation (1.1) characterizing electrically short transmission lines, for which sin q @ q = ba = w a lc . Then Z0q = w a transmitted power becomes:
lc l / c = w al = wL = X, the series inductance of the line, and the

P@

VsVr sin d X

or

P@

V2 sin d X

(1.6)

This simplified equation neglects the shunt capacitance of the line. The effect of shunt capacitance on the transmission for lines shorter than 100 miles is indeed negligibly small. Moreover, although the line capacitance, as explained above, can cause over-voltage problems for under loaded lines, but, since sinq q and thus Z0 sinq X, it always tends to increase the transmittable power. Thus, the neglect of shunt line capacitance represents a worse than actual case from the standpoint of maximizing the (steady-state or

8 transient) transmittable power, but does not falsify the main considerations governing power flow control (which also determine the possible utilization of transmission assets and the application of power electronics technology). The voltage related problems of open circuited and underloaded lines are usually handled satisfactorily by permanently connected or switched shunt reactors in combination with the excitation control of generators.
X/2
I X/2

Vs
(a)

Vm

Vr

Vx = j X I

Vs

Vm

Vr

I (b)

? ??
P,Q 2 Pmax

?? ?

Q=

2 V 2 ( 1-cos ) ? X

P (c) 0 ??? ? P= V2 sin ? X

Figure 1.3 Model of a two machine power system with inductive line (a), corresponding phasor diagram (b) and power transmission vs. angle characteristic (c).

Equation (1.6) can also be derived from the elementary techniques of ac circuit analysis, using complex phasors, an approach which allows a very effective treatment and illustration of different compensation and control concepts employed in power electronics based Controllers. Considering again the simple two machine model, but assuming a purely inductive transmission line with zero shunt capacitance, as shown in Figure 1.3a with the corresponding phasor diagram in Figure 1.3b. If the sending- and receiving-end voltages are defined by
Vs = Ve
jd / 2

= V (cos

d d + j sin ) 2 2

(1.7)

and
Vr = Ve
- jd / 2

= V (cos

d d - j sin ) 2 2

(1.8)

then the midpoint voltage is

Vm = Vs + Vr d = Vm e j 0 = V cos 2 2

(1.9)

and the current through the line is given by

9
I= Vs - Vr 2V d = sin jX X 2

(1.10)

In the case of lossless line assumed, power is the same at both ends (and at the midpoint), i.e.,
V2 P = VmI = sin d X

(1.11)

which is, of course, identical to that given by equation (1.6). The reactive power provided for the line at each end is
Qs = - Qr = VI sin

d V2 = ( 1 - cos d ) 2 X

(1.12)

The relationships between real power P, reactive power Q, and angle d are shown plotted in Figure 1.3c. As seen, at a constant voltage (Vs=Vr=V) and fixed transmission system (X=const) the transmitted power is exclusively controlled by angle d. Note also that real power, P, cannot be controlled without also changing the reactive power demand on the sending- and receiving-ends. A different illustration using voltage and current phasors to provide a physical visualization for the coupled variation of real power, reactive power, and of voltage along the transmission line, as a function of angle d, is presented in Figure 1.4a and 1.4b. Figure 1.4a shows that the voltage phasor Vx across the series line reactance increases, together with line current phasor I , as angle d is increased. At the same time, the (line to neutral) voltage along the line decreases from the ends towards the middle, reaching a minimum at the actual midpoint. It can also be observed that the relative angular position of the voltage along the line from the two ends is continuously changing (in the opposite direction) until at the midpoint it falls precisely in phase with the line current I (which is at a fixed 90! angle with respect to the voltage Vx). Thus, the product of the midpoint voltage and the line current, VmI, yields the transmitted power P. It is evident from this relationship that the progressive increase of angle d will not progressively increase the power P. This is because for d>90!, the midpoint voltage will decrease more rapidly than the line current increases and, consequently, their product will decrease from its maximal value and ultimately reach zero at d=180!.

10
VS

Vx
X /2

Vr

X /2 Mid -point

Vx P = Vm I
I increases with ??? V x 2V ? sin = I = 2 X X Vm decreases with ?? Vm = V cos ? 2 Vs = Vr = V

I
Vm

Vs

Vr

? ? 2 2

P = V I cos

? 2

? Q = V I sin 2

Figure 1.4a Variation of mid-point voltage and line current with transmission angle d.

Figure 1.4b illustrates the variation of the in-phase and quadrature components of the line current with respect to the (receiving-)end voltage, as angle d is varied from zero to 180! (i.e., d/2 varied from zero to 90!). The quadrature component of the line current with respect to the sending-end, or the receiving-end, voltage progressively increases with d until it becomes the total line current at d=180!. By contrast, the in-phase component of the line current with respect to the end voltage increases with d in the 0<d<90! interval, and then it decreases to zero as d reaches 180!.
Line current: I =

? 2V sin 2 X
V15 o
V30 o

I 90o I 75o I 60o I 45 o

? Receiving -end voltage: V ( ) 2 V45o

I 30o
Pmax

V60 o

P = V I cos
V75 o

? 2

I 15o

Angle

? 2
0

Locus of Q Q max Locus of P

Q = V I sin
V90 o

? 2

Figure 1.4b Variation of the transmitted real power P and the (receiving-end) reactive power Q with transmission angle d. 1

11 Conclusions on power transmission by a lossless inductive line: 1. At fixed end voltages and line impedance, the transmission angle controls the power flow in the line by changing the voltage across the line (measured from the sending- to the receiving-end). 2. The voltage along the transmission line decreases from the ends towards the midpoint of the line with increasing transmission angle. The midpoint voltage is maximum at zero degree, and zero at 180 degrees. 3. The line current increases with increasing transmission angle. The product of the midpoint voltage and the line current determines the transmitted active power at any given transmission angle. The active power is zero at 0 and 180 degrees; maximum at 90 degrees. 4. The variation of the transmission angle changes not only the line current, but also the angles between the end voltages and the line current. Thus, with increasing transmission angle the reactive power flow in the line starts to increase rapidly: it is zero at zero transmission angle, reaches the value of maximum transmittable active power at 90 degrees, and then increases to the double of that at 180 degrees. 5. The adjustment of the transmitted active power by angle control is always associated with a mathematically defined reactive power flow change in the line. 6. The reactive power in the line is supplied by the generators. 7. The transmittable electric power at a given system voltage is a function of the electrical length of the line, i.e., the value of the effective series line reactance X. Once the theoretical limit of steady-state power transmission is reached at 90 degrees, the transmitted power would decrease with increasing line length (increasing X) unless either the line voltage is increased or the effective line impedance is decreased. 1.2.2 Transmission Via Multiple Paths in Intertied Power Systems In a modern power system a load area usually receives power through two or more paths, some of which may be corridors of several lines. Although in this case the power in each line is still characterized by the previously derived analytical formulae with the associated observations, nevertheless some other factors important from the standpoint of system imposed transmission limits and line utilization, should also be considered. Indeed, these considerations will further expand the area of the applications for power electronics based Controllers to achieve optimized system operation with effective utilization of system assets. A simple case of power flow through three paths is illustrated in Figure 1.5a. In a general case, the transmission lines representing the three paths could differ both in length and character (transmission voltage and unit-length impedance); consequently, the total impedance (inductive and resistive) determining the power transmission (Figure 1.5b) over the three independent paths between the two buses at a given transmission angle would be different. If the impedance difference is not correlated with the power transmission capacity of the line, then, evidently, one line could become overloaded, and another one underloaded. For example, one path could be a long, high impedance line with high transmission capacity and another one a short, low impedance line with much lower transmission capacity. Thus, although the three paths may represent the necessary transmission capacity between the two buses, the actual transmittable power would be limited by the low impedance line.

12

Vs

Q s1

P1

Qr1

Vr

System "s"

Qs2

P2

Qr2

System r" Qs3

P3 Q r3

(a)
Vs I 1
jX1

R1

Vr

I2

jX2

System "s"

R2

System r"

I3

jX3

R3
(b)

Figure 1.5 Multi-path power transmission (a), and corresponding circuit elements determining power flow distribution (b).

In addition to the main potential problem of improper (inductive) impedance ratio between the three transmission paths, disproportionate load sharing could also be caused by the different values of the reactive to real impedance ratios (X/R) in the three lines. That is, with different X/R ratios the currents flowing in the lines have different angles with respect to the bus voltages, resulting in circulating currents through the loops formed by the lines. In an actual power system, many buses are interconnected by transmission lines of different operating voltage, impedance and length to provide an arrangement in which power can freely flow from areas of surplus generation capacity to areas of power deficit. Such an arrangement practically always results in a number of major and many minor power loop flows, due to the differences in reactive and real impedances of the lines providing the various interconnecting paths. Uncontrolled loop power flow may become a limiting factor between certain points of the overall power system; it can hinder the proper utilization of system assets, and increase transmission losses. Conclusions for multiple-paths transmission 1. Power will flow according to the (normalized) reactive impedance of the paths, and not according to the power transmission capacity of the lines. 2. The natural utilization of the transmission lines is generally uneven; some may be fully loaded and some others grossly underloaded. 3. Internal circulating loop power flow may be caused by different X/R ratios of the lines forming the loop. 4. In a meshed power system with many interconnected buses the power flow between certain areas may take place over a number of major and many minor power loops, some of which could stretch out over long distances. 5. Loop power flows may limit transmittable power, hinder asset utilization, and increase transmission losses. 1.2.3 Inherent Steady-State Limits of AC Power Transmission

13 The relationship between the transmitted power and the transmission parameters given by Equation (1.11) indicates that the maximum power, P max=V2/X , transmittable over a lossless line at a given transmission voltage, is totally determined by the line reactance X and thus sets the theoretical limit for steady-state power transmission. A practical limit for an actual line with resistance R may be imposed by the I 2R loss that heats the conductor. At a certain temperature the physical characteristics of the conductor would irreversibly change (e.g., it could get deformed with permanent sag). This sets a thermal limit for the maximal transmittable power. Generally, for long lines X, and for short lines R would provide the main transmission limitation. AC loads are generally sensitive to the magnitude, and may be sensitive to the frequency of the applied alternating voltage. AC power systems are generally operated at a substantially constant (typically 50 or 60 Hz) frequency. The voltage levels in ac systems may moderately vary, but are not allowed to exceed well defined limits (typically +5 and -10%). This tight voltage tolerance may impose the primary transmission limitation for long radial lines (no generation at the receiving end) and for the so-called tapped-lines, which feed a number of (relatively small) loads along the transmission line. In a system with multiple transmission paths the steady-state transmission limits may be reached or exceeded as a result of parallel and loop power flows. These flows, as previously discussed, often occur in a multi-line, interconnected power system, as a consequence of basic circuit laws which define current flows by the impedance rather than the current capacity of the lines. They can result in overloaded lines with thermal and voltage level problems. Conclusions for steady-state transmission limits 1. The theoretically maximum transmittable power of a natural (uncompensated) line at a given transmission voltage is determined by the reactive line impedance (reactance), ultimately the length of the economically acceptable transmission line. 2. The line resistance determines the thermal limit for line. This limit is absolute, determined by line design (primarily conductor size). Whereas the effective line reactance can be changed by appropriate compensation to increase the limit for the transmittable power, or the distance of transmission, similar compensation could not change the actual line losses although it could decrease the resistive voltage drop over the total transmission path. 3. Allowed bus voltage variation of +5 and -10 per cent, also imposes a limit on transmittable power, particularly for radial lines. 1.2.4 Extension of Power Transmission Limits by Line Compensation and Power Flow Control It follows from the reactive character of ac transmission lines (and the corresponding analytical expressions) that the steady-state transmittable power can be increased and the voltage profile along the line controlled by appropriate reactive compensation. The purpose of this reactive compensation is to change the natural electrical characteristics of the transmission line to make it more compatible with the prevailing load demand. Thus, shunt-connected, fixed or mechanically switched reactors can be applied to minimize line overvoltage under light load conditions, and shunt connected, fixed or mechanically switched capacitors can be used to maintain voltage levels under heavy load conditions. In the case of long transmission lines, series capacitive compensation is often employed to establish a virtual short line by reducing the inductive line impedance and thereby the electrical length of the line. In some multi-line system configurations, it can happen that the transmission angle imposed naturally on a particular line is inappropriate for the power transfer planned for that line. In this case, a phase angle regulator (or
1

14 phase shifter) may be employed to control the angle of this line independent of the prevailing overall transmission angle. In the following sub-sections, basic concepts for increasing the transmittable power by ideal shuntconnected reactive compensation, series compensation, and phase angle regulation, as well as for controlling active and reactive power flow in the line, will be reviewed. These basic approaches will provide the foundation for power electronics-based compensation and control techniques capable not only to increase steady-state power flow, system imposed transmission limits but also to improve the stability and overall dynamic behavior of the system. 1.2.4.1 Shunt Compensation Figure 1.6a shows the simple two machine (two-bus) transmission model in which an ideal var compensator is shunt-connected at the midpoint of the transmission line. This compensator is represented by a sinusoidal ac voltage source (of the fundamental frequency), in-phase with the midpoint voltage, Vm, with an amplitude identical to that of the sending- and receiving-end voltages (Vm = Vs = Vr = V) The midpoint compensator in effect segments the transmission line into two independent parts: the first segment, with an impedance of X/2, carries power from the sending end to the midpoint, and the second segment, also with an impedance of X/2, carries power from the midpoint to the receiving end. Note that the mid-point var compensator exchanges only reactive power with the transmission line in this process. The relationship between voltages, Vs, Vr,, Vm, (together with Vsm, Vrm,), and line segment currents Ism and Imr is shown by the phasor diagram in Figure 1.6b.
X /2
Ism Imr
X /2

(a)

Vs

Ideal Comp. (P=0)

Vm

Vr

j X Is m
2

j X Imr
2

Vs
(b) Ism

Vsm

Vm

Vmr

Vr
V s = V r = Vm = V

Imr

???
P,Q 4 Pmax (c) 2 Pmax Pmax

??? 2

Qp=

4V X

(1- cos ? ? ?)
sin ? ? ?

Pp=

2V 2

X
2

P = V sin ? X

? ??

Figure 1.6 Two machine power system with an ideal mid-point reactive shunt compensator (a), corresponding phasor diagram (b), and transmitted power vs. transmission angle characteristic (c).

For the lossless system assumed, the real power is the same at each terminal (sending-end, mid-point, and receiving-end) of the line, and it can be derived readily from the phasor diagram of Figure 1.6b using a similar computational process demonstrated in the previous section [see equations (1.7) through (1.11)].
4

15 With

d Vsm = Vmr = V cos ; 4

the transmitted power is

I sm = I mr = I =

4V d sin X 4

(1.13)

P = Vsm I sm = Vmr I mr = Vm I sm cos

d d = VI cos 4 4

(1.14)

or
V2 d P=2 sin X 2

(1.15)

Similarly
Q = VI sin

d 2V 2 d = ( 1 - cos ) 4 X 2

(1.16)

The relationship between real power P , reactive power Q , and angle d for the case of ideal shunt compensation is shown plotted in Figure 1.6c. It can be observed that the mid-point shunt compensation can significantly increase the transmittable power (doubling its maximum value) at the expense of a rapidly increasing reactive power demand on the mid-point compensator (and also on the end-generators). The concept of transmission line segmentation can be expanded to the use of multiple compensators, located at equal segments of the transmission line, as illustrated for four line segments in Figure 1.7. Theoretically, the transmittable power would double with each doubling of the segments for the same overall line length. Furthermore, with the increase of the number of segments, the voltage variation along the line would rapidly decrease, approaching the ideal case of constant voltage profile. Ultimately, with a sufficiently large number of line segments, an ideal distributed compensation system could theoretically be established, which would have the characteristics of conventional surge impedance loading, but would have no power transmission limitations, and would maintain a flat voltage profile at any load.

16

X /4 Isl

X /4 Ilm

X /4 Imn

X /4 Inr

Vs

Vl

Vm

Vn

Vr

j X I sl 4

j X Ilm 4

j X Imn 4

j X I nr 4
Vr

Vl
Vs

Vm

Vn

I sl I lm

Inr Imn

Figure 1.7 Two machine system with ideal reactive shunt compensators providing multiple line segmentation, and associated phasor diagram

It is to be appreciated that such a distributed compensation hinges on the instantaneous response and unlimited var generation and absorption capability of the shunt compensators employed, which would have to stay in synchronism with the prevailing phase of the segment voltages and maintain the predefined amplitude of the transmission voltage, independently of load variation. To visualize the operation and control coordination complexity of a generalized compensation scheme exhibiting ideal transmission characteristics, consider the lossless but otherwise correctly represented system of Figure 1.2. Assume that the line is provided with a sufficient number of shunt connected ideal var compensators. At no load (zero transmission), the voltages of all compensators would be in phase and they would be absorbing the capacitive vars generated by the distributed line capacitance. With increasing load (increasing d ), the relative phase angle between the voltages of adjacent compensators would increase, but their var absorption would continuously decrease up to the natural (surge impedance) loading, where it would become zero. With further increasing load, beyond the surge impedance loading, the compensators would have to generate increasing amount of capacitive vars to maintain the flat voltage profile. However, at sufficiently heavy loads, the relative phase angle between two adjacent compensators could become too large, resulting in a large voltage sag, at which the power transmission could not be maintained, regardless of the var generation capacity of the compensators, unless additional compensators would be employed to increase further the segmentation of the line. From the above discussion it is evident that the controlled shunt compensation scheme approximating an ideal line, whose surge impedance is continuously variable so as to maintain a flat voltage profile over a load range stretching from zero to several times the actual surge impedance characterizing that line, would be too complex, and far too expensive, to be practical, particularly if stability and reliability requirements under appropriate contingency conditions are also considered. However, it is interesting to note historically that the practicability of limited line segmentation, using 12, 300 Mvar thyristor-controlled static var compensators, has been demonstrated by the major, 600 mile long, 735 kV transmission line of the Hydro-Quebec power system built to transmit up to 1200 MW power from the James Bay hydrocomplex to the City of Montreal and to neighboring US utilities. More importantly, the transmission
6

17 benefits of voltage support by controlled shunt compensation at strategic locations of the transmission system have been demonstrated by numerous installations in the world. Of course, there are many applications in which mechanically-switched capacitors are applied to control transmission line voltage where there are slow, daily and seasonal load variations. Although, reactive shunt compensation is discussed above in relation to a two-machine system, this treatment can easily extended to the more special case of radial transmission. Indeed, if a passive load, consuming power P at voltage V, is connected to the midpoint in place of the receiving-end part of the system (which comprises the receiving-end generator and transmission link X/2), the sending-end generator with the X/2 impedance and load would represent a simple radial system. Clearly, without compensation the voltage at the mid-point (which is now the receiving-end) would vary with the load (and load power factor). It is also evident that with controlled reactive compensation the voltage could be kept constant independent of the load. Shunt compensation in practical applications is often used to regulate the voltage at a given bus against load variations, or to provide voltage support for the load when, due to generation or line outages, the capacity of the sending-end system becomes impaired. Conclusions on shunt compensation 1. Reactive shunt compensation can maintain constant midpoint voltage at increasing (varying) transmission angle, thereby in effect segmenting the transmission line into two independent lines of half length and thus doubling the theoretically transmittable power. 2. With multiple shunt compensation (multiple segmentation) the theoretical, steady-state transmittable power can be multiplied. 3. Shunt compensators are also effective in maintaining (regulating) the voltage at the end of a radial line. 4. Under steady-state conditions, shunt compensators used for voltage regulation or control do not exchange active power with the ac system (neglecting the possible internal losses of practical compensators). 1.2.4.2 Series Compensation The basic idea behind series capacitive compensation is to decrease the overall effective series transmission impedance from the sending-end to the receiving-end. The conventional view is that the impedance of the series connected compensating capacitor cancels a portion of the actual line reactance and thereby the effective transmission impedance is reduced as if the line was physically shortened. An equally valid physical view, helpful to the understanding of power flow controllers, is that, in order to increase the current (and thereby the transmitted power) through the series impedance of a physical line (which clearly cannot be changed), the voltage across this impedance needs to be increased. This can be accomplished by a series-connected (passive or active) circuit element that produces a voltage opposite to the prevailing voltage across the series line reactance. The simplest such element is a capacitor, but, as will be seen later, controlled voltage sources can also accomplish this function in a much more generalized manner that can ultimately facilitate full control of real and reactive power flow in the line. The familiar two-machine model with a series capacitor compensated line, composed of two identical segments for the clarity of illustration, shown at Figure 1.8a. The corresponding voltage and current phasors are shown at Figure 1.8b. Note that the magnitude of the total voltage across the series line reactance, Vx = 2Vx/2, increased by the magnitude of the opposite voltage, VC, developed across the series

18 capacitor.
XC 2
X 2

X 2

XC 2

P,Q

Vs

Vm

Vr

k=

XC
X

(a)

Qs C =
k = 0.4
2P max
X -j C I 2

2 V 2k (1-cos? ) X (1- k) 2
V sin ? X (1-k)
2

Vx
X -j C I 2

Ps =

Vs

Vm
I

Vr

P max

k = 0.2 k=0

??2
(c)

(b)

Figure 1.8 Two machine power system with series capacitive compensation (a), corresponding phasor diagram (b), and transmitted power vs. transmission angle characteristic (c).

The effective transmission impedance Xeff with the series capacitive compensation is given by Xeff = X - XC or Xeff = (1 - k) X where k is the degree of series compensation, i.e., k = XC/X 0 k <1 (1.19) (1.18) (1.17)

The current in the compensated line and the real power transmitted, per Equations (1.10) and (1.11) are:
I= 2V d sin (1 - k )X 2 V2 sin d (1 - k )X

(1.20)

P = VmI =

(1.21)

The reactive power supplied by the series capacitor can be expressed by

QC = I X C
2

2V 2 k = ( 1 - cos d ) X (1 - k )2

(1.22)

The relationship between the real power P , series capacitor reactive power Q C, and angle d is shown plotted at various values of the degree of series compensation k in Figure 1.8c. It can be observed that, as

19 expected, the transmittable power rapidly increases with the degree of series compensation k . Similarly, the reactive power supplied by the series capacitor also increases sharply with k and varies with angle d a similar manner as the line reactive power. Series capacitors have been used extensively in the last 50 years throughout the world for the compensation of long transmission lines. It should be noted that series reactive compensation can be an effective method for balancing power flow (and thereby avoiding line overloading), when power transmission involves multiple paths with lines of different length and character, by adjusting the impedance of the individual lines according to their loading capacity. Conclusions on series compensation 1. Reactive series compensation reduces the effective reactive impedance and thereby the electric length of the line. 2. Series compensation provides the most effective method for economical long distance ac power transmission. 3. Series compensation is an effective technique to balance power flow in multi-path power transmission. 4. Under steady-state conditions, reactive series compensators do not exchange active power with the ac system (neglecting the possible internal losses of practical compensators). 1.2.4.3 Transmission Angle Control In practical power systems it occasionally happens that the transmission angle required for the optimal use of a particular line would be incompatible with the proper operation of the overall transmission system. Such cases would occur, for example, when power between two buses is transmitted over parallel lines of different electrical length or when two buses are intertied whose prevailing angle difference is insufficient to establish the desired power flow. In these cases a phase shifter or phase angle regulator is frequently applied. The basic concept is explained again in connection with the two machine model in which a phase shifter is inserted between the sending-end generator (bus) and the transmission line, as illustrated in Figure 1.9a. The phase shifter can be considered as a sinusoidal (fundamental frequency) ac voltage source with controllable amplitude and phase angle. In other words, the effective sending-end voltage Vseff becomes the sum of the original sending-end bus voltage Vs and the voltage Vs provided by the phase shifter, as the phasor diagram shown in Figure 1.9b illustrates. The basic idea behind the phase shifter is to keep the transmitted power at the desired level, independent of the prevailing transmission angle d , in a predetermined operating range. Thus, for example, the power can theoretically be kept at its peak value after angle d exceeds p/2 (the peak power angle) by controlling the amplitude of quadrature voltage Vs so that the effective phase angle (d-s) between the sending- and receiving-end voltages stays at p/2. In this way, the actual transmitted power may be increased significantly, even though the phase-shifter per se does not increase the steady-state power transmission limit.

20

V?

(a)

Vs

Phase shifter

Vseff

Vr

-V ?
+V?

Vx ( ? ? )

Vx (??? ? )

Vx ( ? ? )

Vs
(b)

Vseff (? ? )

Vr

Vseff (? ? )

??? ?
?? ??

???
P a=
V 2 sin(?? ? ) X

P P max
(c)

?? ???

??

?? ?

Figure 1.9 Two machine power system with a phase shifter (a), corresponding phasor diagram (b), and transmitted power vs. transmission angle characteristic (c).

With the above phase angle control arrangement the effective phase angle between the sending- and receiving-end voltages becomes (d-s), and the transmitted power P can therefore be expressed as:
P= V2 sin(d - s ) X

(1.23)

The relationship between real power P and angles d and s is shown plotted in Figure 1.9c. It can be observed that, although the phase-shifter does not increase the transmittable power of the uncompensated line, theoretically it makes it possible to keep the power at its maximum value at any angle d in the range p/2 < d < p/2+s by; in effect, shifting the P versus d curve to the right. It should be noted that the P versus d curve can also be shifted to the left by inserting the voltage of the phase shifter with an opposite polarity. In this way, the power transfer can be increased and the maximum power reached at a generator angle less than p/2 (that is, at d = p/2-s). In contrast to the previously-investigated shunt and series compensation schemes, the phase shifter generally has to handle both active and reactive powers. The VA throughput of the phase shifter (viewed as a voltage source) is VA = "Vs - Vseff""I" ="V s""I" = VsI (1.24)

Assuming that the magnitude of the original sending-end voltage, Vs, and that of the effective sending-end voltage, Vseff, are equal, and using the expression given for the line current in Equation (1.10), VA can be written in the following form:

21
VA = 4V 2 d -s s sin sin X 2 2

(1.25)

In Equation (1.25), the multiplier sin[(d -s )/2] defines the current, at a given system voltage and line impedance, flowing through the phase shifter, and the multiplier sin(s/2) determines the magnitude of the voltage injected by the phase shifter. Phase shifters employing a shunt connected excitation transformer with a mechanical tap-changer and a series connected insertion transformer to provide adjustable series voltage injection for phase angle control are often employed in transmission system to control steady-state power flow and prevent undesired parallel and loop power flows. Conclusions on transmission angle control 1. Transmission angle control can adjust the prevailing power flow, but cannot increase the maximum transmittable power characterizing the controlled line. 2. Transmission angle control changes both the active and reactive line power in a mathematically defined manner. 3. In the case of conventional angle regulators (tap-changing transformers), both the active and reactive power exchanged in the process of angle control is supplied by the ac system. 1.2.4.4 Independent Control of Active and Reactive Power Flow Independent control of active and reactive power flow in the line is attainable by the generalized power flow controller, which can inject in series with the line the compensating ac voltage with a phase angle fully controllable from zero to 360 degrees with respect to a chosen reference (bus voltage or line current). The two-machine system with the generalized power flow controller is shown in Figure 1.10a and the corresponding phasor diagram in Figure 1.10b. Similarly to the previously discussed phase shifter, the power flow controller adds voltage phasor Vpq, with controllable magnitude, Vpq, and phase angle, r, to the original sending-end bus voltage phasor Vs and thereby producing the effective sending-end voltage phasor Vseff for the line. Since in addition to the adjustable magnitude Vpq, angle r is freely variable from 0 ! to 360 !, the voltage across the transmission line and, with that, the line current, is controllable both in magnitude and phase angle with respect to the receiving-end bus voltage (or other selected voltage reference). The capability of adjusting both the magnitude and phase angle of the line current enables the generalized power flow controller to control independently the active and reactive power in the line.

22

Figure 1.10 Two machine power system with a generalized power flow controller (a), corresponding phasor diagram (b), control region of attainable active and reactive line power (c), and transmitted power vs. transmission angle characteristic (d).

The active and reactive control capability of the generalized power flow can be conveniently illustrated in the {Q,P} plane where the normalized reactive power Q (p.u.) is plotted against the normalized active power P (p.u.) of the simple two-machine model shown in Figure 1.10a. For the uncompensated system, with V 2/X=1 (p.u.) stipulation, the function Q=f(P) can be readily written from Equations (1.11) and (1.12):
Q = -1 - 1 - P 2

(1.26)

or
( Q + 1 )2 + P 2 = 1

(1.27)

Equation (1.27) shows that the Q=f(P) function for the uncompensated system describes a circular arc drawn with a radius of 1.0 around the center defined by coordinates Po=0, Qo=-1.0, shown by dashed-line in Figure 1.10c. Each point defined by its P and Q coordinates on this arc corresponds to a particular transmission angled; e.g., coordinates P= 0, Q =0 corresponds to d =0; P=1.0, Q =-1.0 corresponds to d =90 ! , etc., which identifies the starting control point of the power flow controller. Assuming, as illustrated, that the prevailing transmission angle, d , is 30! , then, the full (360! ) rotation of the compensating voltage phasor Vpq, with its maximum magnitude Vpq(=0.5 p.u.), will describe a circle in the {Q,P} plane with a radius of 0.5 around its center defined by coordinates P30! =0.5 and Q30! =-0.134, which characterize the uncompensated system at d=30!. The area within this circle defines all P and Q values obtainable by controlling the magnitude and angle of compensating voltage phasor Vpq. In other words, this circle in the {Q,P} plane defines all P and Q transmission values attainable with the given rating of the power flow controller for the power system considered. With changing transmission angle, the control area with the same circular boundary would simply move up or down along the arc

23 characterizing the uncompensated system. The transmittable power against the transmission angle d between the boundaries determined by the magnitude Vpq of the injected voltage is shown in Figure 1.10d. Similarly to the phase shifter, and in contrast to the series reactive compensator, the generalized power flow controller has to handle both active and reactive powers. The VA throughput of the phase shifter (viewed as a voltage source) is VA = "Vs - Vseff""I" ="V pq""I" = VpqI (1.28)

The conventional, phase shifter type implementation of the generalized power flow controller (e.g., using a mechanical tap-changer with in-phase and quadrature voltage steps) would necessitate the power system (sending-end bus) to supply both the active and reactive power, i.e., the total VA, exchanged in the course of the desired control action. However, as will be seen later, the converter-based implementation would internally generate all the reactive power exchanged, and thus the system would supply only the active power exchanged (which actually transmitted to the receiving-end bus). The basic concept of independent active and reactive power flow control, that is, in effect the insertion of a voltage source with controllable magnitude and phase angle in series with the line, can also be applied for the minimization of active and reactive loop power flows in multi-path and meshed systems. As illustrated in Figure 1.11a, if the angle of the inserted voltage is in quadrature with the current in a substantially inductive line, then the magnitude of the inserted voltage will control only the magnitude of the line current, and thus primarily the active power flow in the line. This means essentially that series reactive compensation primarily changes active power flow.
Vq
Pq I

VL

Qq

VLmax VLo Vq (+90 )

o

Vq = + - j Vq(? )

_ I I
I
o

XL

Vq (-90 o)

VLmin I max

(a)

Vs

Vq (+90 )

Vq (-90 )

Vr

Vs

Io Vr I min

Reactive power compensation changes the magnitude of the line current and thus primarily controls real power
Vp

? /2
VL
(+P )

? /2

Pp I

VL

Qp

+ Vp
VLo

_ I +Vp Vp = + - Vp (? ) I I

XL
+R

- Vp
Vr

(b)
Vs
- Vp
-R

Vs
I (-P)

VL( -P)

Vr

Io

I (+P)

Real power compensation changes the angle of the line current and thus primarily controls reactive power

? /2

? /2

Figure 1.11 Illustration for effect of line compensation: (a) reactive compensation primarily changes active power, and (b) active power compensation primarily changes reactive power flow.

Similarly, Figure 1.11b illustrates that if the inserted series voltage is in phase with the line current, then it will primarily control the angle of this current, and thereby the reactive power flow in the line. This means
2

24 that series active power compensation primarily changes reactive compensation in the line. From the above it follows that active loop power flow can be minimized by quadrature voltage injection providing essentially series reactive compensation; reactive loop flow can be minimized by in-phase voltage injection providing essentially series real power compensation. Conclusions on independent active and reactive power flow control 1. The active and reactive power flow in the line at a given transmission angle can be controlled independently by a compensating voltage of adjustable magnitude and fully controllable phase angle (0!360!) injected in series with the line. The phase angle of the injected voltage of appropriate magnitude is adjusted so as to force the line current to assume the angular position with respect to the selected (receiving-end) voltage that results in the desired active and reactive power flow. 2. The injected voltage in series with the line generally results in both active and reactive power exchange between the power flow controller and the ac system. With possible conventional implementation (inphase and quadrature tap-changing transformers) both the active and reactive power exchanged would be supplied by the ac system; with converter-based implementation, only the active power would be supplied by the ac system, the reactive power would be internally generated. 3. Injected series voltage with controllable angle can be used for the minimization of either, or both, the active and reactive loop power flows in the line by providing controllable reactive and/or active power compensation. 1.2.5 Dynamic Limits of AC Power Systems AC power systems employ rotating synchronous machines for electric power generation. (They may also employ rotating synchronous compensators (condensers) for reactive power compensation.) It is a fundamental requirement of useful power exchange that all synchronous machines in the system operate in synchronism with each other maintaining a common system frequency. However, power systems are exposed to various dynamic disturbances (such as line faults, equipment failures, various switching operations), which may cause a sudden change in the real power balance of the system and consequent acceleration and deceleration of certain machines. The ability of the system to recover from disturbances and regain the steady-state synchronism under stipulated contingency conditions becomes a major design and operating criterion for transmission capacity. This ability is usually characterized by the transient and dynamic stability of the system. A transmission system is said to be transiently stable if it can recover normal operation following a specified major disturbance. Similarly, the system is said to be dynamically stable if it recovers normal operation following a minor disturbance. The dynamic stability indicates the damping characterizes the system. A dynamic (or oscillatory) instability means that a minor disturbance may lead to increasing power oscillation and the eventual loss of synchronism. During and after major disturbances the transmission angle and transmitted power may significantly change from, and oscillate around their steady-state values. Consequently, a power system cannot be operated at, or even too close to its steady-state power transmission limit. An adequate margin is needed to accommodate the dynamic power swings while the disturbed machines regain their synchronism in the system. The phenomenon of escalating decrease and an eventual collapse of the terminal voltage as a result of an incremental load increase is referred to as voltage instability. Voltage collapse is the result of a complex

25 interaction between induction motor type loads and certain voltage regulators, such as tap-changing transformers, which may take several seconds to minutes. The essence of this process is that decreasing terminal voltage results in increasing load current and poorer load power factor (induction motors) which tend to further decrease the terminal voltage. The voltage regulators (tap-changing transformers) are not able to change the character of this process and under sufficiently severe conditions (low system voltage and heavy loading) it degenerates (in a positive-feedback manner) into a voltage collapse. The voltage stability limit identifies for a given system the specific V and P condition at which the next increment of load causes a voltage collapse. Summary on dynamic transmission limits 1. The main synchronous machines of the power system must operate in synchronism to maintain a substantially constant, common system frequency. 2. Transient stability means that the system can recover normal operation following a major disturbance. 3. Dynamic stability means that the system can recover normal operation following a minor disturbance. Dynamic stability in effect means that the system has sufficient damping. 4. Voltage instability means an escalating process which results in the progressive decrease and the eventual collapse of the terminal bus-voltage after an incremental load increase. Voltage stability limit specifies the voltage and load limits for the terminal bus beyond which the process of voltage collapse would begin. 1.2.6 Dynamic Compensation for Stability Enhancement As shown in Section 1.2.4, shunt and series line compensation, as well as transmission angle and power flow control can significantly change the prevailing transmitted power. Thus, it is reasonable to expect that, with suitable and fast controls, these techniques would become able to vary the power flow in the system so as to counteract disturbances and thereby increase the transient stability limit and provide effective power oscillation damping. It is also predictable that appropriate shunt and series compensation would be effective for increasing the voltage stability limit. In the following two sub-sections, the potential effectiveness of shunt and series compensation, angle and power flow control for transient stability improvement and power oscillation damping are explored and compared. In the subsequent third sub-section, the use of shunt and series capacitive compensation for the increase of voltage instability limit for a radial transmission line is discussed. 1.2.6.1 Transient Stability Improvement The potential effectiveness of shunt and series compensation, angle and power flow control on transient stability improvement can be conveniently evaluated by the equal area criterion. The meaning of the equal area criterion is explained with the aid of the simple two machine (the receiving-end is an infinite bus), two line system shown in Figure 1.12a and the corresponding P versus d curves shown in Figure 1.12b. Assume that the complete system is characterized by the P versus d curve a and is operating at angle d1 to transmit power P1 when a fault occurs at line segment 1. During the fault the system is characterized by P versus d curve b and thus, over this period, the transmitted electric power decreases significantly while mechanical input power to the sending-end generator remains substantially constant. As a result, the generator accelerates and the transmission angle increases from d1 to d 2 at which the protective breakers

26 disconnect the faulted line segment 1 and the sending-end generator absorbs accelerating energy, represented by area A1. After fault clearing, without line segment 1 the degraded system is characterized by P versus d curve c. At angle d2 on curve c the transmitted power exceeds the mechanical input power and the sending end generator starts to decelerate; however, angle d further increases due to the kinetic energy stored in the machine. The maximum angle reached at d3, where decelerating energy, represented by area A2, becomes equal to the accelerating energy represented by area A1. The limit of transient stability is reached at d3 = dcrit, beyond which the decelerating energy would not balance the accelerating energy and synchronism between the sending-end and receiving-end could not be restored. The area Amargin, between d3 and dcrit, represent the transient stability margin of the system.

Figure 1.12 Illustration of the equal area criterion for transient stability: (a) power system model, and (b) transmitted power vs. transmission angle characteristic under fault condition.

From the above general discussion it is evident that the transient stability, at a given power transmission level and fault clearing time, is determined by the P versus d characteristic of the post fault system. Since, as previously shown, shunt and series compensation and angle control improve the natural transmission characteristic of the system, it can be expected that the judicious employment of these techniques would be highly effective in increasing the transmission capability of the post-fault system and thereby enhancing transient stability. For comparison, consider the four basic two-machine (sending-end generator, receiving-end infinite bus) systems, with no compensation, mid-point shunt compensation, series compensation, phase angle control, and generalized, active and reactive power flow control, as shown in Figures 1.3, 1.6, 1.8, 1.9 and 1.10. For clarity, the above introduced equal-area criterion is applied here in a greatly simplified manner, with the assumption that the original single line model represents both the pre-fault and post-fault systems. (The impracticality of the single line system and the questionable validity of this assumption have no effect on this qualitative comparison.) Suppose that in the uncompensated and all four compensated

27 systems the steady-state power transmitted is the same. Assume that all four systems are subjected to the same fault for the same period of time. The dynamic behavior of the four systems is illustrated in Figures 1.13a through 1.13e. Prior to the fault, each of the four systems transmits power Pm (subscript m stands for mechanical) at angles d1, d p1, d s1, d a1, and dpq1, respectively. (Subscripts p, s , a, and pq stand for parallel, series, angle and active and reactive power flow.) During the fault, the transmitted electric power (of the single line system considered) becomes zero while the mechanical input power to the generators remains constant (Pm). Therefore, the sending-end generator accelerates from the steadystate angles d1, d p1, d s1, d a1, and dpq1 to angles d2, dp2, ds2, da2, and dpq2 at which the faults clears. The accelerating energies in the four systems are represented by areas A1, A p1, As1, Aa1, and Apq1. After fault clearing, the transmitted electric power exceeds the mechanical input power and the sending-end machine decelerates, but the accumulated kinetic energy further increases until a balance between the accelerating and decelerating energies, represented by areas A 1, A p1, A s1, A a1, A pq1 and A 2, A p2, A s2, A a2, Apq2, respectively, is reached at d3, dp3, ds3, da3, and dpq3. The difference between angles d3, dp3, ds3, da3, and dpq3, representing the maximum angular swings, and the critical angles dcrit, dpcrit, dscrit, dacrit, and dpqcrit determines the margin of transient stability, that is, the unused and still available decelerating energy, represented by areas Amargin, Apmargin, Asmargin, Aamargin, and Apqmargin.

Figure 1.13 Equal-area criterion to illustrate the transient stability margin for a simple two machine system, (a) without compensation, (b) with an ideal mid-point compensator, (c) with a series capacitor, (d) with a phase shifter, and (e) with a generalized power flow controller.

Comparison of Figures 1.13a through 1.13e clearly shows a substantial increase in the transient stability margin that the various compensation approaches can provide through the control of different system parameters. The shunt-connected ("parallel') var compensation method provides the improvement by segmenting the transmission line and regulating the midpoint voltage. The series capacitive compensation approach reduces the effective transmission impedance and minimizes the transmission angle. Phase angle control keeps the transmission angle, d-s, at p/2 for maintaining maximum power transmission while the generator angle d swings beyond this value. The active and reactive power flow controller forces the
2

28 desired power flow by the injecting the necessary voltage in series with the line. The use of any one of the four compensation approaches obviously can increase the transient stability margin significantly over that of the uncompensated system. However, it should be noted that the diagrams illustrate ideal cases, the potential of various compensation techniques for improving transient stability. The theoretically attainable (and practically unnecessarily large) increase in the transient stability margin represented by the yellow-colored areas in the figures could be obtained only at economically unacceptable MVA ratings, particularly for the shunt and series capacitive compensators. Still, the improvements in transient stability usually required in practice can be achieved most of the time at reasonable MVA rating and at competitive cost. If the uncompensated system has a sufficient transient stability margin, the compensation techniques described can considerably increase the transmittable power while maintaining the necessary stability margin. In the explanation of the equal area criterion at the beginning of this section, a clear distinction was made between the pre-fault and post-fault power system. It is important to note that from the standpoint of transient stability, and thus of overall system security, the post-fault system is the one that counts. That is, power systems are normally designed to be transiently stable, with defined pre-fault contingency scenarios and post-fault system degradation, when subjected to a major disturbance (fault). Because of this (sound) design philosophy, the actual capacity of transmission systems is considerably higher than that at which they are normally used. Thus, it may seem technically plausible (and economically savvy) to employ fast acting compensation techniques, instead of overall network compensation, specifically to handle dynamic events and increase the transmission capability of the degraded system under the contingencies encountered. Conclusions on transient stability 1. Equal area criterion is an effective tool for the visual understanding and comparative evaluation of methods for transient stability improvement. 2. All shunt and series compensation methods, as well as angle and power flow control are effective for transient stability improvement with sufficiently fast control action. 3. Because of the powerful effect of reactive series compensation and power flow control with directly injected forcing voltage, these can be expected to give greater increase in transient stability at a given equipment rating that reactive shunt compensators and phase angle regulators. 1.2.6.2 Power Oscillation Damping In an under-damped power system any minor disturbance can cause the machine angle to oscillate around its steady state value at the natural frequency of the total electromechanical system. The angle oscillation, of course, results in a corresponding power oscillation around the steady-state power transmitted. The lack of sufficient damping can be a major problem in some power systems and, in some cases it may be the limiting factor for the transmittable power. Until the late 1970s, the excitation control of rotating synchronous machines was the available active means for power oscillation damping. Later technological developments made it possible to realize rapidly controllable reactive shunt and series compensators, transmission angle and power flow controllers to provide highly effective power oscillation damping. (Since compensations can be used for power flow control, and angle or power flow control can be considered as compensation of specific system
2

29 parameters, the general term of compensator or Controller is often used for brevity to refer to the group of these equipments and compensation for their action.) Since power oscillation is a sustained dynamic event, it is necessary to vary the applied compensation so as to achieve consistent and rapid damping. The control action required is in principle the same for all of compensation approaches. That is, when the rotationally oscillating generator accelerates and angle d increases (dd/dt > 0), the electric power transmitted must be increased to compensate for the excess mechanical input power. Conversely, when the generator decelerates and angle d decreases (dd/dt < 0), the electric power must be decreased to balance the insufficient mechanical input power. (The mechanical input power is assumed to be essentially constant in the time frame of an oscillation cycle.) The requirements of output control, and the process of power oscillation damping, by the four compensation approaches, are illustrated in Figure 1.14a through 1.14f. Waveforms at a show the undamped and damped oscillations of angle d around the steady-state value d0. Waveforms at b show the undamped and damped oscillations of the electric power P around the steady-state value Po. (The momentary drop in power shown in the figure represents an assumed disturbance that initiated the oscillation.)
?
(a)

Undamped

?o
P

t
Undamped

(b)

Po

Qp
(c)

t t

k
(d)

?
(e)

0
0

t
t

Vpq
(f)

Figure 1.14 Waveforms illustrating power oscillation damping by shunt compensator, series compensation, phase shifter, and generalized power flow controller: (a) generator angle, (b) transmitted power, (c) var output of a shunt compensator, (d) degree of series compensation, (e) angle variation of a phase shifter, and (f) active power variation by generalized flow controller.

Waveform c shows the reactive power output Qp of a shunt-connected var compensator. The capacitive (positive) output of the compensator increases the midpoint voltage and the transmitted power when dd/dt > 0, and it decreases those when dd/dt < 0. Waveform d shows the required variation of k=XC/X for series capacitive compensation. When dd/dt > 0, k is increased and thus the line impedance is decreased. This results in the increase of the transmitted power. When dd/dt < 0, k is decreased (in the illustration it becomes zero) and the power transmitted is decreased to that of the uncompensated system. Waveform e shows the variation of angle s produced by the phase shifter. (For the illustration it is
2

30 assumed that s has an operating range of -smax s smax, and d is in the range of 0 < d < p/2.) Again, when dd/dt > 0, angle s is negative making the power versus d curve (refer to Figure 1.9c) to shift to the left, which increases the angle between the end terminals of the line and, consequently, also the real power transmitted. When dd/dt < 0, angle s is made positive, which shifts the power versus angle curve to the right and thus decreases the overall transmission angle and transmitted power. Waveform f shows the injected forcing Vpq voltage to increase active power flow during machine acceleration (dd/dt > 0) and decrease it during deceleration (dd/dt < 0). During damping, the angle r of the injected voltage is kept at p/4, at which the forcing voltage controls exclusively the active power flow. As the illustrations show, a bang-bang type control (output is varied between minimum and maximum values) is assumed for all four compensation approaches. This type of control is generally considered the most effective, particularly if large oscillations are encountered. However, for damping relatively small power oscillations, a strategy that varies the controlled output of the compensator continuously, in sympathy with the generator angle or power, may be preferred. To detect accurately machine acceleration and deceleration to control the output of the compensators is usually location dependent and the selection of the useful signals for control may require studies and experimentations. Conclusions on power oscillation damping 1. Power oscillation damping can be achieved with all compensators and power flow controllers by varying their output so as to increase active power transmission when the involved machines are accelerating, and decreasing it when they are decelerating. 2. The derivation of useful control signal indicating machine acceleration and deceleration may require studies and experimentations in practice. 1.2.6.3 Increase of Voltage Stability Limit Voltage collapse can be a potential problem primarily for buses supplying load areas with heavy motor loads. Radial lines are particularly prone to this problem. The solutions for all cases are similar and illustrated here for the simple radial system with feeder line reactance of X and load impedance Z, shown in Figure 1.15a, together with the normalized terminal voltage Vr versus power P plot at various load power factors, ranging from 0.8 lag and 0.9 lead. The nose-point at each plot given for a specific power factor represents the voltage instability corresponding to that system condition. It should be noted that the voltage stability limit decreases with inductive loads and increases with capacitive loads.

31

Figure 1.15 Variation of voltage stability limit with load and load power factor (a), extension of voltage stability limit by reactive shunt compensation (b), and extension of voltage stability limit by series capacitive compensation (c).

The inherent circuit characteristics of the simple radial structure, and the Vr versus P plots shown, clearly indicate that both shunt and series capacitive compensation can effectively increase the voltage stability limit. Shunt compensation does so by supplying the reactive load and regulating the terminal voltage (VVr=0) as illustrated in Figure 1.15b. Series capacitive compensation does so by canceling a portion of the line reactance X and thereby in effect providing a stiff voltage source for the load, as illustrated for a unity power factor load in Figure 1.15c. Conclusions on voltage stability 1. The problem of voltage instability and collapse occur usually at buses supplying loads with heavy motor loads. 2. Both shunt and series compensation ban be used for voltage stability increase. The shunt compensation provides closer regulation, the series compensation a smoother, more manageable transition from stable to unstable state.
1.3 Transmission Problems and Needs, - The Role of Power Electronics 1.3.1 Problems and needs

The physical laws of ac power systems impose basic limitations on the transmittable power, apart from the current conducting capacity of the line. These limitations are due partly to the electro-mechanical characteristic of the ac machines, and partly to the electrical LC lattice network type characteristic of the ac transmission lines. The nature of generator imposed limitation is maintainable rotational stability, to prevent either a run-away speed and frequency change (transient instability) or periodic frequency variation with corresponding oscillation of the transmitted power between the generators and other synchronous elements of the system (dynamic instability). This necessitates the limitation of the
3

32 transmitted power to a level at which the generators can recover normal synchronous operation following a defined (worst case) major and minor disturbances in the system. The transmission line imposed limitation is acceptable voltage variation along the line since the series line inductance causes increasing voltage droop with increasing load (and/or increasing lagging load power factor), and the line capacitance causes growing voltage increase with decreasing load. The voltage variation, primarily voltage decrease, sets the limit for the line-length dependent transmittable power. As shown in the previous sections, the natural limits of ac power transmission can be increased, theoretically to any extent by line compensation and power flow control techniques, which change the natural impedance characteristic of the line and control the otherwise free, impedance-defined power flow. The line impedance caused steady-state voltage variation can be minimized by fixed and mechanically switched series and shunt reactive compensation, however, the management of the dynamic rotational and voltage variation problems requires fast control actions. The application of fast controls in the execution of shunt- and series-compensation, as well as angle and power flow control, makes it possible to reduce dynamic voltage and frequency (machine speed) variation and thereby increase the transmittable power without decreasing the necessary limits established for secure system operation. In other words, fast controls capable of counteracting dynamic system variations allow higher utilization of transmission assets. Conclusions on transmission problems and system needs 1. Transmission limitations are caused by the rotational stability problems of generators due to disturbances, and to voltage variations with changing load due to the reactive line impedance. 2. Steady-state transmission limits can be increased by fixed, or mechanically switched reactive compensation; the improvement of dynamic rotational and voltage stability requires fast control actions. 1.3.2 The role of power electronics Dynamic voltage variation and rotational stability problems of ac power transmission, which had necessitated the under-utilization of lines and other system assets, provided the incentives in the late 1970s to introduce power electronics-based control for reactive compensation. This normal evolutionary process has been greatly accelerated by subsequent developments in the utility industry in the subsequent decades, which have aggravated the early problems and highlighted the structural limitations of power systems in a greatly changed socio-economical environment (see Section 1.1 Historical Perspective). The desire to offer economical, and rapidly executable solutions to these problems by means of increased utilization of system assets, led to focused technological developments under the Flexible AC Transmission System (FACTS) initiative of the Electric Power Research Institute (EPRI) to provide power electronicsbased Controllers for real time control of voltage and power flow in ac transmission systems. It is to be noted that the aim of the power electronics-based Controllers is higher utilization of transmission assets and greater freedom in managing power transmission, particularly under heavy power demand or emergency conditions. The application of power electronics is, in general, not a substitute for new lines, but a powerful means to achieve higher utilization of all lines, old and new, in the system. Conclusion on role of power electronics
3

33 Power of electronics by means of providing rapidly controllable compensation and power flow control of the line can increase the transmittable power and the utilization of the transmission assets. 1.3.3 The Objectives of FACTS Modern transmission networks are increasingly expected to facilitate the transfer of electric power from any supplier to any consumer over a large geographic area under market forces (supply vs. demand) controlled, and thus continuously varying, pattern of contractual arrangements. The realization of such a network is significantly constrained by construction cost, right-of-way, and environmental restrictions. Thus, the optimization and full utilization of existing transmission assets, together with systematically planned and executed construction of new facilities, appears to be the only realistic approach for meeting the complex transmission requirements of modern socio-economic system. The Electric Power Research Institute, after years of supporting the development of high power electronics for reactive compensation, formalized the broad concept of Flexible AC Transmission System (FACTS) in the late 1980s. The acronym FACTS identifies alternating current transmission systems incorporating power electronics-based controllers to enhance the controllability and increase power transfer capability. The FACTS initiative was originally launched to solve the emerging system problems in the late 1980s due to restrictions on transmission line construction, and to facilitate the growing power export/import and wheeling transactions among utilities, with two main objectives: (1) To increase the power transfer capability of transmission systems, and (2) to keep power flow over designated routes. The first objective implies that power flow in a given line could be increased up to the thermal limit by forcing the necessary current through the series line impedance if, at the same time, stability of the system is maintained via appropriate real-time control of power flow during and following system faults. This objective of course does not mean to imply that the lines would normally be operated at their thermal limit loading (the transmission losses would be unacceptable), but this option would be available, if needed, to handle severe system contingencies. However, by providing the necessary rotational and voltage stability via FACTS Controllers, instead of large steady-state margins, the normal power transfer over the transmission lines is estimated to increase significantly (in some cases up to about 50%, according to some studies conducted). The second objective implies that, by being able to control the current in a line (by, for example, changing the effective line impedance), the power flow can be restricted to selected (contracted) transmission corridors (which have been contracted and have the capacity) while parallel and loop-flows can be mitigated. It is also implicit in this objective that the primary power flow path must be rapidly changeable to an available secondary path under contingency conditions to maintain the desired overall power transmission in the system. It is easy to see that the achievement of the two basic objectives would significantly increase the utilization of existing (and new) transmission assets, and could play a major role in facilitating deregulation with minimal requirements for new transmission lines. The implementation of the above two basic objective requires the development of high power compensators and controllers. The technology needed for this is high power (multi-hundred Mva) electronics with real-time operating control. However, once a sufficient number of these fast compensators and controllers are deployed over the system, the coordination and overall control to provide maximum system benefits and prevent undesirable interactions with different system configurations and objectives,

34 under normal and contingency conditions, present a different technological challenge. This challenge is to develop appropriate system optimization control strategies, communication links, and security protocols. The realization of such an overall system optimization control can be considered as the third, and still a future, objective of the FACTS initiative. Conclusions on the objectives of FACTS The overall objective of FACTS is to economically solve transmission problems, increase transmission capacity by higher utilization of system assets, and provide improved flexibility for power export and import among utilities, using state-of-the-art power electronics-based compensators and controllers. To this end, FACTS is to (1) increase power transmission capability, up to thermal limit if needed, with the necessary transient, dynamic and voltage stability, and (2) restrict power flow along defined transmission corridors, and eliminate (minimize) capacity limiting loop flows. 1.4 Transmission Compensators and Controllers 1.4.1 Theoretical concepts Shunt compensators are primarily used to regulate the transmission voltage at critical system locations under varying load and system conditions. Thus, an ideal shunt compensator is functionally a controllable ac current source that can force the necessary current through the effective line impedance to maintain the desired bus voltage by negating the voltage drop caused by the prevailing line (load) current, as illustrated in Figure 1.16a. Since the line impedance is substantially reactive, the voltage regulation can be achieved by purely reactive (capacitive or inductive) compensating current. There are two approaches to the realization of controllable (reactive) current sources which can be implemented by practical power electronic circuits: one is controllable reactive admittance (Figure 1.16b) and the other is a controllable voltage source in series with a tie reactance (Figure 1.16c). The adjustment of the admittance changes the compensating current drawn by the compensator from the bus; and similarly, the magnitude of the voltage source with respect to the prevailing bus voltage determines the voltage across its tie reactance and thereby the compensating current drawn from the bus.

35
I
X1

X2
Icomp

(a)

Vs

Vreg
Controllable reactive current source

X1

X2

Icomp

(b)

Vs

V reg

? Y comp
Controllable reactive admittance

X1

X2

I comp

(c)

Vs

X comp

V reg
Controllable reactive voltage source

Figure 1.16 Representation of an ideal shunt compensator by a controllable reactive current source (a) and its realization by a controllable reactive admittance (b), and by a controllable voltage source with reactive tie impedance (c).

Series compensators are primarily used to control transmission line current and thereby the transmitted power. Thus, an ideal series compensator is functionally a controllable ac voltage source, as illustrated in Figure 1.17a, that can inject the compensating voltage in-phase, or anti-phase, with the prevailing voltage drop across the line impedance to increase or decrease the line current, as if the effective line impedance was changed. Since the series line impedance is inductive, the voltage across is in quadrature with the line current. Therefore, the compensating voltage injected to increase or decrease the line current has also to be in quadrature with the line current. Therefore, the series compensator does not exchange active power with the line and thus it can be realized by a controllable reactive impedance in series with the line, or by a controllable voltage source whose phase angle is kept rigidly at 90! with respect to the line current, as shown in Figures 1.17b and 1.17c, respectively.

36
Vcomp

X1

X2
Controllable reactive voltage source

(a)

Vs

Vcomp
I

X1

X2

? Xcomp
(b)
Vs

Controllable reactive impedance

X1

Vcomp

X2

(c)

Vs
Controllable reactive voltage source

Figure 1.17 Representation of an ideal series compensator by a controllable reactive voltage source (a) and its realization by a controllable reactive impedance (b), and by a controllable reactive voltage source with transformer coupling (c).

Phase Shifters (Angle Regulators) and general active and reactive power flow controllers insert the compensating voltage with respect to a selected bus voltage to control the corresponding effective transmission angle. For phase shifting, the injected voltage is adjusted so as to establish the desired transmission angle between the sending- and receiving-end voltages, regardless of the phase angle it may have with respect to the line current. For the independent control of active and reactive power flow, the injected voltage is controlled so as to establish the necessary magnitude and phase angle for the line current with respect to the bus voltage of interest (typically the receiving-end), at which the desired active and reactive line power is defined. It is evident that the source of the injected voltage used for phase shifting, as well as that employed for power flow control, must be able to exchange both reactive and active power with the ac system. Thus, the realization of these by controllable impedance, similar to the series reactive compensator) would require both a variable reactive impedance (capacitance or inductance) and variable positive or negative resistance (power sink and source) in series with the line. Of course, in a practical implementation only the reactive power exchange can be supported by external elements (capacitor, reactor, or reactive power generator like a synchronous compensator), the active compensating power exchanged through series compensation must be supplied or absorbed by the ac system itself. Two conceptual implementations functionally suitable for both phase shifting and power flow control are shown in Figures 1.18a and 118.b. A tap-changing transformer arrangement with ( ) in-phase and quadrature voltage injection capability (Figure 1.18a), can, with sufficient number of taps, rotate the inserted voltage from zero to 360! and change its magnitude from zero to a maximum value. In this case, however, both the active and reactive power exchanged through series compensation is supplied or absorbed by the sending- (or receiving-) end bus via the tap-changing transformers. An idealized motorgenerator set with four-quadrant operation (Figures 1.18b) is an equivalent conceptual solution in which, however, only the active power exchanged in the compensation is provided by the ac system, the reactive power is supplied by the generator itself.
3

37
I
X1
Vpq

X2

(a)

Vs
-

In -phase and quadrature Tap -changer

X1

Vpq

X2

(b)

Vs

Voltage control

Ideal synchronous motor-generator set

Voltage control

Figure 1.18 Conceptual realizations of an ideal phase shifter or generalized power flow controller by in-phase and quadrature tap-changers (a), and by an ideal synchronous motor-generator set (b).

It should be noted that the shown realizations of the general power flow controller can be restricted to those of specific types of phase shifters. In particular, if the magnitude of the phase-shifted voltage is not required to stay constant, then the in-phase voltage insertion can be omitted, which leads to considerable circuit simplification for the tap-changing transformer type of realization. (However, it does not change the conceptual motor-generator arrangement.) Using only quadrature voltage insertion, the phase-shifted voltage increases with the angle of shift, s, i.e., Vshifted =V/coss. This type of phase-shifter is often referred to quadrature booster. It should also be noted that the tap-changing transformer with in-phase voltage insertion is often used separately in practical applications for voltage regulation. The conceptual representations of the basic transmission compensator and controllers illustrated in Figures 1.16 through 1.18 can, as will be seen, be realized functionally faithfully by power electronic circuits. Conclusions on compensator and power flow concepts 1. Reactive shunt compensation increases transmittable power by maintaining transmission voltage, series compensation by decreasing line impedance, and phase shift by controlling transmission angle. 2. Ideal shunt compensator is a controllable reactive current source that can be realized by a controllable reactive admittance (susceptance), or by a controllable reactive voltage source in series with a fixed reactive tie impedance (reactance). 3. Ideal series compensator is a controllable reactive voltage source with respect to the line current that can be realized by a controllable series reactive impedance, or by a controllable reactive voltage source. 4. Ideal phase shifter or power flow controller is functionally a voltage source with controllable magnitude and phase angle that can exchange both active and reactive power with the ac system. It can be realized by combination of in-phase and quadrature tap-changing transformers, or by a coupled arrangement of two voltage sources functioning as an ideal, four-quadrant motor-generator set.

38 5. A quadrature booster is a phase-shifter, whose injected voltage is maintained in quadrature with respect to the controlled bus voltage. 1.4.2 Power electronics-based realization of basic transmission compensators and controllers The development of power electronics-based compensators and controllers has followed two distinctly different technical approaches, as the conceptual representations in the previous section suggest, both resulting in a comprehensive group of controllers able to address targeted transmission problems. The first group employs reactive impedances or tap-changing transformers with conventional thyristor switches (can be turned on by gate control, but turns off only if the current becomes zero) as controlled-elements; the second group uses static switching-converters with gate turn-off semiconductors (GTO, GCT, IGBT, etc.), as controlled voltage sources. The thyristor-controlled compensators/controllers have a common characteristic in that the necessary reactive power required for the compensation is generated or absorbed by traditional capacitor or reactor banks (shunt and series reactive compensators), or drawn from the ac system (phase-shifter), and the thyristor switches (valves) are used only to set the combined reactive impedance these banks present to ac the system or connect the necessary voltage steps for phase shifting. By contrast, the compensator/controller approach based on static switching-converters, operated as synchronous voltage sources, is able to generate internally, without capacitors and inductors, the reactive power needed for reactive compensation. In addition, they can also exchange active power with ac system and thus also facilitate real power compensation and asynchronous power transfer. 1.4.2.1 Shunt Compensators: SVC and STATCOM Static Var Compensator (SVC). A typical shunt-connected static var compensator (SVC), composed of thyristor-switched capacitors (TSCs) and thyristor-controlled reactors (TCRs), is shown in Figure 1.19a. With proper coordination of the capacitor switching and reactor control, the var output can be varied continuously between the capacitive and inductive ratings of the equipment.

Transmission line

VT
Coupling transformer
PT

Transient rating

Thyristor valves

Control

Vref
Auxiliary inputs

L
Reactor banks

L
Parameter setting

IC ICmax
Capacitive 0

IL ILmax
Inductive

Capacitor banks

(a)

(b)

39
Figure 1.19 Static Var Compensator employing thyristor-switched capacitors and thyristor-controlled reactors (a), and its V-I regulating characteristic (b).

The compensator is normally operated to regulate the voltage of the transmission system at a selected bus. The V-I characteristic of the SVC, shown in Figure 1.19b, indicates that regulation with a given slope around the nominal voltage can be achieved in the normal operating range defined by the maximum capacitive and inductive currents of the SVC. However, the maximum obtainable capacitive current decreases linearly (and the generated reactive power in quadrature) with the system voltage since the SVC becomes a fixed capacitor when the maximum capacitive output is reached. Therefore, the voltage support capability of the conventional thyristor-controlled static var compensator rapidly decreases with decreasing system voltage. Static Synchronous Compensator (STATCOM). The STATCOM is analogous to an ideal, rotating synchronous compensator (condenser), which is, as known, a synchronous machine generating a balanced set of (three) sinusoidal voltages at the fundamental frequency, with controllable amplitude and phase angle. This ideal machine has no inertia, its response is practically instantaneous and it can internally generate reactive (both capacitive and inductive) power by excitation control. The basic switching converter equivalent of the rotating machine is a matrix of solid-state switches which connect a dc voltage source sequentially to the three output terminals so as to generate a set of three balanced voltages. This type of converter is called voltage-sourced converter (VSC). The simplest realization of VSC, as a synchronous voltage source, is shown in Figure 1.20a in the functional role of synchronous compensator (condenser). (It is to be noted that this simple switch matrix would only produce a quasi-square wave output. In practice more complex switch arrangements and/or special modulation techniques are used to produce outputs that can sufficiently approximate sine-waves.) Since in this role the converter-based voltage source is used strictly for reactive shunt compensation, like the rotating synchronous compensator or the conventional static var compensator, thus it exchanges no active power (except for losses) and the dc voltage can be sustained by a relatively small dc capacitor. The converter itself keeps the capacitor charged to the required voltage level by making the ac output voltage of the converter to slightly lag the system voltage, and thereby absorb a small amount of active power from the ac system to replenish the internal operating losses. This method can also be used to increase or decrease the capacitor voltage, and thereby the amplitude of the output voltage of the converter, for the purpose of controlling the var generation or absorption in the same way as done by the excitation control of a rotating synchronous machine, as illustrated in Figure 1.20b.

40

AC system V T

?Vg

Coupling Transformer

~ Iq (+I dc )
Vg
Generator ac terminal

AT AC TERMINAL
Supplies "Q"

Transient rating
Transient rating

Vg > VT +Iq
PT

V gc

Vga

+ Vg- V T Vg

VT

V gb

Vg < VT -Iq

-? Vg

+? Vg

Absorbs "Q"
Vref
Auxiliary inputs

Control

AT DC TERMINAL

Vdc
C

Figure 1.20 STATCOM employing a voltage-sourced converter (a), var output control by output voltage variation (b), and V-I regulating characteristic (c).

The V-I characteristic of the STATCOM is shown in Figure 1.20c. As illustrated, the STATCOM can provide both capacitive and inductive compensation and is able to control its output current over the rated maximum capacitive or inductive range independently of the ac system voltage. That is, the STATCOM can provide full capacitive output current at any system voltage, close to zero. By contrast, the SVC can supply only diminishing output current with decreasing system voltage, as Figure 1.19b illustrates. 1.4.2.2 Series Compensators: TCSC and SSSC Thyristor-Controlled Series Capacitor (TCSC). There are two basic schemes of thyristor-controlled series capacitors: one uses thyristor-switched capacitors, as shown schematically in Figure 1.21a, and the other employs a fixed capacitor in parallel with a thyristor-controlled reactor, as shown Figure 1.21b. (Note that the former often referred to as Thyristor-Switched Capacitor.)

Idc
+

Switching converter

I dc
Parameter setting

+ Vdc 0

IC ICmax
Capacitive

IL
0

DC terminal
Energy source or storage

ILmax
Inductive

(a)

(b)

(c)

41

C1
CT
CF

Cn

(a)

Transmission line

P (pu)

Tyristor valve
PT

2.0

Ps =

V2 X L (1 -k)

sin ?

Control

Parameter setting Control inputs

1.5
k = 0.33
1.0

k = 0.2

C
CT
CF

(b)
Tyristor valve

Transmission line

k=

XC

k=0

Thyristor -Controlled Reactor

Parameter setting Control inputs

XL

PT

Control

???
(C)

Figure 1.21 Controllable series compensator scheme using thyristor-switched series capacitors (a), or a thyristor-controlled reactor parallel with a fixed capacitor (b), and transmitted power vs. transmission angle characteristic with variable series capacitive compensation (c).

In the thyristor-switched capacitor scheme of Figure 1.21a, the degree of series compensation is controlled by increasing or decreasing the number of capacitor banks in series. To accomplish this, each capacitor bank is inserted or bypassed by a thyristor valve (switch). To minimize switching transients and utilize natural commutation, the operation of the thyristor valves is coordinated with voltage and current zero crossings. Since, the voltage across the series capacitor is a direct function of the line current, the prevention of damaging overvoltage during faults and other surge current conditions usually necessitates the use of a ZnO type voltage clamping device or other by-pass arrangement in parallel with the thyristorswitched capacitor banks. In the fixed-capacitor, thyristor-controlled reactor scheme of Figure 1.21b, the degree of series compensation in the capacitive operating region is increased (or decreased) by increasing (or decreasing) the thyristor conduction period, and thereby the current in the TCR. Minimum series compensation is reached when the TCR is off. The TCR may be designed to have the capability to limit the voltage across the capacitor during faults and other system contingencies of similar effect. The transmitted power versus transmission angle as a function of series capacitive compensation is shown in Figure 1. 21c. The two schemes may be combined by connecting a number of basic TCR with fixed parallel capacitor modules in series in order to achieve greater control range and flexibility. Static Synchronous Series Compensator (SSSC). The concept of the SSSC is based on the recognition that the function of the series capacitor is simply to produce an appropriate voltage at the fundamental ac system frequency to increase the voltage across the inductive line impedance, and thereby the fundamental line current and the transmitted power. (This of course has the same electrical effect as if the series line inductance was reduced to that of a shorter line.) Therefore, an ac voltage source of fundamental frequency, similar to that employed in the STATCOM, inserted in series with the line with a locked quadrature (lagging) phase relationship to the line current, and whose amplitude is made proportional to that of the line current, i.e., Vc =kXI (where, as before, Vc is the amplitude of the inserted voltage, X is the

42 line impedance, k is the degree of series compensation, and I is the amplitude of the line current), provides series reactive compensation equivalent to that obtained by a series capacitor, as illustrated in Figure 1.22a. However, in contrast to the series capacitor, the SSSC is able to maintain a constant compensating voltage in face of variable line current, or control the amplitude of the injected compensating voltage independent of the amplitude of the line current. (This is analogous to the STATCOM which is able to maintain a constant compensating current in face of grossly decreased bus voltage.)
V2 X sin ? +

Vq

Iline

X
CT
+Vq

Transmission line

P =
P (pu)

? V Vq cos 2 X
Vq = 0.707 Vq = 0.353

Coupling Transformer

1.5

Iline

Iline
Vg
V gc
Vga
V gb
Gen. ac terminal

PT

1.0
-Vq

Vq = 0 V q = -0.353

Vq= -jkX Iline

Vq,ref
Auxiliary inputs

For the emulation of series capacitor

0.5

Vq = -0.707

Control

0
For line current Independent compensation

Vdc
C

Figure 1.22 SSSC employing a voltage-sourced converter (a), and its transmitted power vs. transmission angle characteristic (b

For normal capacitive compensation, the output voltage lags the line current by 90 degrees. However, the output voltage of the SVS can be reversed by simple control action to make it lead the line current by 90 degrees. In this case, the injected voltage decreases the voltage across the inductive line impedance and thus the series compensation has the same effect as if the reactive line impedance was increased. The transmitted power P versus transmission angle characterizing the SSSC is illustrated in Figure 1.22b. Comparison of these plots to those shown in Figure 1.21c for the series capacitor (of comparable compensation range) clearly shows that the series capacitor increases the transmitted power by a fixed percentage of that transmitted by the uncompensated line at a given transmission angle and, by contrast, the SSSC increases it by a fixed fraction of the maximum power transmittable by the uncompensated line, independent of the transmission angle, in the important operating range of 0 through p/2. From the standpoint of practical applications, steady-state power flow control or stability improvements, the SSSC clearly has considerably wider control range than the controlled series capacitor of the same Mva rating. 1.4.2.3 Phase Shifters and Power Flow Controllers: TCPAR and UPFC, IPFC, BtB Thyristor-Controlled Phase Angle Regulator (TCPAR). Although there is no high power, non-mechanical phase-shifter in service, the principles for using a phase-shifting transformer with a thyristor tap-changer are well established. To avoid excessive circuit complexity, the Thyristor-Controlled Phase Angle

Idc
+

Switching converter

DC terminal
Energy source or storage

Parameter setting

I Vq Vq = j I

? /2

-0.5

(a)

(b)

43 Regulator, just as its commonly used conventional counterpart with a mechanical tap-changer, provides quadrature voltage injection, that is, it is in effect a quadrature booster. A Thyristor-Controlled Phase Angle Regulator is shown in Figure 1.23a. It consists of a shunt-connected excitation transformer with appropriate taps, a series insertion transformer and a thyristor valve arrangement connecting a selected combination of tap voltages to the secondary of the insertion transformer. The excitation transformer has three non-identical secondary windings, in proportions of 1:3:9. It can produce a total of 27 steps using only 12 thyristor valves (of three different voltage ratings, i.e., the voltage rating of the valves is also in the 1:3:9 ratio) per phase with a switching arrangement that can bypass a winding or reverse its polarity.
V?
Transmission line

V* Series transformer
Measured variables
2 V? P = V ( sin ? ? cos? ) V X

Excitation transformer
1

V?
V ?
Control

V*

V*

= ( V 2? V?2 )

V? V? V? V?

= 1.0 = 0.66 = 0.33

=0

V? V? V? V?

= -1.0 = -0.66 = -0.33 =0

Reference input

9
Thyristor tap-changer

Parameter setting

? ??
(b)

(a)

Figure 1.23 Thyristor-controlled quadrature booster type of phase shifter (a), and its transmitted power vs. transmission angle characteristic (b).

The phase angle requirements for power flow control can be determined from angle measurements, if available, or from power measurements. The TCPAR could be applied to regulate the transmission angle to maintain balanced power flow in multiple transmission paths, or to control it so as to increase the transient and dynamic stability of the power system. The transmitted power versus transmission angle characteristic of the TCPAR, with the described practical realization of the quadrature booster, is shown in Figure 1.23b. It is to be noted that the phase angle between the voltage injected by the TCPAR (which is, by design, in quadrature with the line to neutral terminal voltage) and the line current is arbitrary, determined by the pertinent parameters of the overall power system. This means that, in general, the TCPAR must exchange, via the series insertion transformer, both active and reactive power with the ac system. Since this type angle regulator cannot generate, or absorb, either real or reactive power, it follows that both the active and reactive power that it supplies to, or absorbs from the line when it injects quadrature voltage must be absorbed from it, or supplied to it, by the ac system. As a consequence, the Mva ratings of the excitation and insertion transformer are substantially the same. The fact that the tap-changing transformer type angle regulator cannot generate or absorb reactive power is a disadvantage in those applications in which the reactive power exchanged has to be transmitted through a line of appreciable length, due to the resulting voltage drop. In these cases, to avoid large voltage drops
4

44 across the line, the TCPAR needs to be complemented with some type of controllable var supply, or located close to the power generator. Converter-Based Angle Regulators and Power Flow Controllers. Converter-based phase-shifters and more general power flow controllers employ the general functional concept of the ideal synchronous motor/generator set, shown in Figure 1.18b, using two voltage-sourced converters in back-to-back configuration with a common dc link as illustrated in Figure 1.24. Such an arrangement is a close approximation of the ideal motor-generator set capable of full four-quadrant operation: Active power flow is fully controllable between the ac terminals in either direction, and each ac terminal can independently supply or absorb reactive power in the defined operating range, independent of the active power transfer within Mva rating of the equipment. With this general arrangement either ac terminal can be coupled in shunt or in series with the transmission line and thus four different power flow controller arrangements, shunt-series, series-series, and shunt-shunt, which are referred to as Unified Power Flow Controller (UPFC), Interline Power Flow Controller (IPFC), and Back-to-Back Tie (BtB) can be realized.

V1
Coupling Transformer

V2
Coupling Transformer

AT AC TERMINAL 1
Vg

Generator 1 ac terminal

Generator 2 ac terminal

AT AC TERMINAL 2 Q2

Vg
V gc

Q1

V ga

Vgb

System variables

V gc

ga

Vgb

P1 < 0 Q1 > 0 P1 < 0 Q1 < 0

P1 > 0 Q1 > 0 P1 > 0 Q1 < 0

P1

P2 < 0 Q2 > 0

P2 > 0 Q2 > 0 P2 > 0 Q2 < 0

P2

Control

P2 < 0 Q2 < 0
+

C
MVA limit
DC terminal

Switching Parameter References Switching converter converter setting Idc

Vdc

Figure 1.24 Realization of ideal synchronous motor-generator set by two voltage-sourced converters in back-to-back connection.

The Unified Power Flow Controller scheme is shown with the conventionally used converter symbols in Figure 1.25a. In this back-to-back configuration of "Converter 1" and "Converter 2", as indicated before, the active power can freely flow in either direction between the ac terminals of the two converters, and each converter can independently generate (or absorb) reactive power at its own ac output terminal.

C
DC terminal

MVA limit

P1 = -P 2

Vdc

45

Transmission Line

V pq

V +V pq

Supply transformer

Converter 1

Converter 2

Series Transformer

Vpq
?
+ Vdc
AC

AC

V V+Vpq

P,Q
P

Control
Measured Variables Parameter Settings

Po

Ppqmax

(Vpq = 0)

Qo

90

180

270

360

V Ref

Z Ref ? Ref

Q
(Vpq = 0 )

Qpqmax

or

P Ref Q Ref

(a)

(b)

Figure 1.25 UPFC scheme using two back-to-back voltage-sourced converters (a), variation of active and reactive line power as a function of control angle r (b), attainable transmitted active power and reactive line power vs. transmission angle characteristics by control (c), and circular control region for P and Q at a given transmission angle (d).

Converter 2 provides the main function of the UPFC by injecting a voltage v pq (represented by phasor Vpq) with controllable magnitude Vpq and phase angle r (0 r 2p) in series with the line via an insertion transformer. This injected voltage acts as a synchronous ac voltage source exchanging active and reactive power with the ac system as the line current flows through it. The reactive power exchanged at the ac terminal (i.e., at the terminal of the series insertion transformer) is generated internally by the converter. The active power exchanged at the ac terminal is converted into dc power which appears at the dc link as a positive or negative real power demand. The basic function of Converter 1 is to supply or absorb the real power demanded by Converter 2 at the common dc link to support the active power exchange resulting from the series voltage injection. This dc link power demand of Converter 2 is converted back to ac by Converter 1 and coupled to the transmission line bus via a shunt-connected transformer. In addition to the real power need of Converter 2, Converter 1 can also generate or absorb controllable reactive power and thereby provide independent shunt reactive compensation for the line. It is important to note that, in contrast to the Thyristor-Controlled Phase Angle Regulator, the reactive power exchanged through the series and shunt transformers is selfsufficiently supplied or absorbed by the corresponding converters themselves, and therefore it does not have to be transmitted by the line. It follows from the unrestricted capability of the UPFC to control the angle of the injected voltage without any restriction (0 r 2p) that it can emulate the operation of the conventional power flow controllers (voltage regulation, series reactive compensation and phase shifting), simply by controlling angle r in specific ways. For terminal voltage regulation the injected voltage is kept in phase or anti-phase with the bus voltage, that is, angle r is kept 0 or p. For series reactive compensation the injected voltage is kept in quadrature with the prevailing line current by the appropriate control of angle r. For angle regulation using the quadrature booster scheme, angle r is kept at 90o with respect to the bus voltage. For ideal phase shift, the magnitude of the injected voltage Vpq is varied with angle r so as to keep the magnitude of the

46 phase-shifted voltage constant independent of the amount of phase-shift. In addition to these specific operation modes, the UPFC is also able to provide the combined control action of these specific controllers, as if each of them was individually coupled in series with the line. This could be visualized as deriving the combined compensation by determining the resultant voltage v pq, with its amplitude Vpq angle r, by summing all the constituent voltages providing the selected individual compensations. In phasor form, this means a simple vectorial summation, i.e., Vpq=DVm (voltage regulation)+Vq (series compensation)+Vs(phaseshifting). The multi-functional compensation capability of the UPFC is based on the independent magnitude and unrestricted angle control of the injected compensating voltage, which is also the basis for executing independent active and reactive power flow control in the line. Since this is generally the real objective of power transmission control, the conventional terms of series compensation and phase shifting become irrelevant, and the UPFC can be viewed simply with the functional objective of controlling the magnitude and phase angle of the injected voltage so as to force the magnitude and angle of the line current, with respect to a selected voltage (e.g., the receiving-end), to such values which yield the desired active and reactive power flow in the transmission line. The variation of the active and reactive line power, P and Q with angle r around the steady-state transmitted power Po and reactive power Qo at a given transmission angledo is shown in Figure 1.25b, the complete active and reactive power versus transmission angle characteristics in Figure 1.25c, and the attainable active and reactive power by the control of the injected voltage Vpq at the given steady-state transmission angle do in Figure 1.25d. As seen in Figure 1.25b, as angle r is varied between 0 and 2p, Ppq( r ) and Q pq(r ) varies sinusoidally (with 90o phase difference) around the steady-state values of transmitted power Po and reactive power Qo obtained at the given transmission angle do. The attainable maximum of Ppq and, similarly, that of Qpq is determined by the magnitude of the injected voltage Vpq , that is, P pqmax= VVpq/X and Q pq= VVpq/X. Consequently, as Figure 1.25c illustrates, the transmittable active power P(d) is controllable at any transmission angle d (0 d p) between

47

Po (d ) -

VV pq X

P(d ) Po (d ) +

VV pq X

and the reactive power Q(d) is between

Qo (d ) VV pq X Q(d ) Qo (d ) + VV pq X

The variation of Ppq(r) and Q pq(r) with angle r shown in Figure 1.25b defines a circular region whose center is defined by the coordinates of the steady-state active and reactive power Po(do ) and Qo(do), that is, 2 VV pq max 2 2 {P(r )- Po (do )} + { } = Q(r )- Qo (do ) X This {Ppq,Qpq} circular control region is shown around the steady-state P o(do ) and Qo(do) coordinate values at the assumed transmission angle of do=30o in Figure 1.25d. The control of P and Q within the region can be executed independently within the restriction of P pq2 + Q pq2 = (VVpq/X)2. Note that the {Ppq,Qpq} control region remains the same at any value of the transmission angle d. The unique capabilities of the UPFC summarized above provide powerful, hitherto unavailable means to maintain or vary the real and reactive power flow in the line to satisfy load demand and improve system stability with optimized asset utilization. The Interline Power Flow Controller represents a novel concept with the objective of providing a flexible power flow control scheme for a multi-line power system, in which two (or more) lines would require series reactive compensation. The IPFC scheme provides, together with independent SSSC type
4

48 controllable series reactive compensation for each line, a capability to transfer active power between the compensated lines. This capability makes it possible to equalize both real and reactive power flow between the lines, to transfer power demand from overloaded to underloaded lines, to compensate against resistive line voltage drops and the corresponding reactive line power, and to increase the effectiveness of the compensating system for dynamic disturbances (transient stability and power oscillation damping). In general, the IPFC provides a highly effective scheme for power transmission management at a multi-line substation. The basic Interline Power Flow Controller scheme, also using of two back-to-back converters, is shown in Figure 1.26a. Each converter is coupled to a different transmission line via its own series insertion transformer and is able to provide independent series reactive compensation to its own line. (Note that the two lines can be totally different in voltage, transmission capacity and loading.)
V1 Vpq1 V 1 + V pq1 Series Transformer 1 I1 Transmission line 1

Transmission line 2

I2

V2

Vpq2

V2 + V pq2

Converter 1

Converter 2

Series Transformer 2

+ Vdc

AC

AC

Measured Variables 1 Measured Variables 2

Control

Parameter Settings

P 1Ref P 2Ref

Primary line selector

Q Ref

(a)

Figure 1.26 Basic IPFC scheme for two lines (a), and active and reactive power vs. transmission angle characteristics attainable for the two lines (b).

Each converter produces controllable ac output voltage at the fundamental frequency, which is synchronized to the voltage of the transmission line which that converter controls. The injected voltage in each line generally has one component that is in quadrature and another that is in phase with the relevant line current. The quadrature component provides series reactive compensation for the line to control active power flow, and the in-phase component defines the active power exchanged with the line to control the reactive flow. Since each converter is self-sufficient in generating or absorbing reactive power, the quadrature voltage components in each line can be independently varied to control the active power transmitted over the line. However, since the active power exchanged by a converter at its ac terminal has to be supplied to, or absorbed from its dc terminal, the in-phase output voltage component of each of the two converters must be controlled so as to ensure a net zero active power balance at their common dc terminals. In other words, the active power compensation demand of one line must be fully supplied (or absorbed) by the other line. This restriction means that changing the reactive power in one direction in one line (e.g., decreasing it), it will have to be changed in the other direction in the other (e.g., increasing it). In
4

49 other words, the reactive power flow can be optimized in one (the weaker or overloaded) line by transferring the burden of reactive power flow to the other (stronger or underloaded) line. From the optimized, prime line viewpoint, the IPFC provides the same full, active and reactive power flow control that characterizes the UPFC. The active power flow in the other line can be controlled by reactive compensation in the same way that characterizes the SSSC. However, the reactive power flow in this line, in addition to the prevailing line load, will be determined by the active power compensation demand of the other line due to the fundamental operating condition that the active power exchanged by the two converters must sum to zero, that is, P1pq = - P2pq. The practical effect of this constraint is illustrated by the active and reactive power versus transmission angle characteristic of the IPFC at its two ac terminals in Figure 1.26b. For this illustration it is assumed that the two lines are identical and thus the IPFC with two identically rated converters has the same control range of power flow control for each line. It can be observed in the figure that the active power flow control range is the same at both terminals, i.e., the reactive compensation can be carried out on each line independently either to increase or decrease the active power flow P(d). However, the reactive power flow control ranges are mirror images of each other for the two lines. That is, if active power is supplied to one line in order to decrease reactive power flow, then the same amount of active power must be absorb from the other line which will increase the reactive power flow in that line. This requirement is illustrated in the Q( d ) versus d characteristic plots by showing the attainable control areas of Q1pq(d) and Q2pq(d) above and below the Qo (d) curve in different colors as mirror images of each other at the two ac terminals of the IPFC.

Figure 1.26 Basic IPFC scheme for two lines (a), and active and reactive power vs. transmission angle characteristics attainable for the two lines (b).

Figure 1.26b clearly show the added flexibility the IPFC configuration provides for series compensation: it is able to control not only the active power flow in a two- or multi-line transmission system, but also can equalize or control the reactive power flow in the lines. The IPFC provides an excellent tool to solve economically power flow problems in a multi-line transmission system in which the actual power flows are not proportional to the capacities of the corresponding lines or to their desired power transmissions,
4

50 or in which the desired real power transmission in some lines is hindered by relatively high reactive power flow. It is evident from the above discussion that the IPFC concept, characterized so far only for two lines, can be extended to multiple (n) lines as illustrated in Figure 1.27a. The underlying idea of this generalized IPFC approach is that the strong or under-loaded lines are forced to help the weaker or over-loaded lines in order to optimize the utilization of the whole transmission system. The operation of a multi-line IPFC requires that the sum of the active power exchanged by the total number of converters must be zero. However, the distribution of the positive and negative power exchanged with the individual lines by the individual converters within this overall constraint is arbitrary (as long as the Mva rating of the individual converters is not exceeded). This arrangement would provide a UPFC-type two-dimensional compensation for some lines and an SSSC-type reactive compensation for the others.

Line n Line 2 Line 1

Line n Line 2
Line 1

SSSC 1
+
DC bus

SSSC 2
+ -

SSSC n
+ -

STATCOM
+
DC bus

SSSC 1
+ -

SSSC 2
+ -

SSSC n
+ -

(a)

(b)

Figure 1.27 Multi-line line IPFC scheme (a), and generalized multi-line IPFC with line independent active power support and bus voltage regulation (b).

The constraint for keeping the sum of the active power exchanged with the n -lines zero can be circumvented by adding a shunt-connected converter to the multi-converter IPFC as illustrated in Figure 1.27b. This arrangement is particularly attractive in those cases in which the active power compensation requirement of the "weak" lines exceeds the real power that can be absorb from the "strong" lines without appreciably impacting their own power transmission or when shunt reactive compensation at the substation is required anyway for voltage support. The Back-to-Back Tie scheme is shown with the conventionally used converter symbols in Figure 1.28. Through the DC link, the tie between AC System 1 and AC System 2 can be asynchronous with large and a variable angle difference or even with different system frequencies. It can also be synchronous, in which case the tie is able to provide synchronizing torque in either direction. Generally, from the viewpoint of either system, the tie behaves like a perfect synchronous generator, providing rapid control for active power with effective current limitations, and also executing bus-voltage regulation by fast reactive compensation. This generator-like characteristic is illustrated at both terminals of the tie in Figure 1.28 by two perfect four-quadrant circular Q versus P plots, within which any combination of simultaneous active and reactive power output is achievable at each side up to the limit determined by the rating of the equipment: P2+Q2MVArating. Of course, the fundamental operating requirement, P1+ P2=0, i.e., P1= -P2,
5

51 must be maintained at any MVA output. In actual operation, generally the required active power transmittal setting would have the priority, and with the maintenance of this, the control would regulate the bus voltages at the two sides independently by the adjustment of the reactive output according to the V-I regulation characteristics shown in Figure 1.28.
AC SYSTEM 1
V1

V2

AC SYSTEM 2

At AC System 1bus Q1
P2 < 0 Q2 > 0 P2 < 0 Q2 < 0 P2 > 0 Q2 > 0 P2 > 0 Q2 < 0

At AC System 2 bus

Coupling Transformer 1
Converter 1

P1 = -P 2

Coupling Transformer 2

Q2
P2 < 0 Q2 > 0

Converter 2

P2 > 0 Q2 > 0 P2 > 0 Q2 < 0

P1

+ Vdc
AC
AC

P2 < 0 Q2 < 0
MVA limit

P2

MVA limit

V1

V2

Measured Variables

Control

Parameter Settings

IC1 IC1max

IL1 I1Lmax

V 1Ref

P Ref

V 2Ref

IC2 IC2max

IL2 IL2max

Figure 1.28 Asynchronous/synchronous tie scheme using two back-to-back voltage-sourced converters, and the corresponding power transmission and voltage regulation terminal characteristics.

Summary on power electronics-based transmission compensators and controllers 1. There are two approaches to the realization of power electronics-based compensators and controllers: One employs thyristor-switched capacitors and reactors, and tap-changing transformers, the other voltage-sourced converters as synchronous voltage sources. The first approach has resulted in the Static Var Compensator (SVC), the Thyristor-Controlled Series Capacitor (TCSC), and the ThyristorControlled Phase Shifter. The second approach has produced the Static Synchronous Compensator (STATCOM), the Static Synchronous Series Compensator (SSSC), the Unified Power Flow Controller (UPFC), the Interline Power Flow Controller (IPFC), and the Back-to-Back Tie (BtB). 2. The two groups of Controllers operate in distinctly different manners: # The thyristor-controlled group employs capacitor and reactor banks with thyristor valves in traditional shunt or series circuit arrangements. The thyristor valves control the on and off periods of the fixed capacitor and reactor banks realizing a variable reactive (shunt or series) impedance. The voltage-sourced converter-based group, using power semiconductors with gate turn-off capability (GTO, GCT, IGBT, etc.), can emulate a synchronous voltage source that internally generate reactive power for transmission line compensation, without the use of ac capacitors or reactors. The converter, if supported by a dc power supply or energy storage device, can also exchange real power with the ac system.

3. The different operations result in different application characteristics:

52 # Shunt compensators. The maximum attainable maximum capacitive compensating current of the SVC, operating as a variable reactive admittance, decreases linearly with the system voltage. The STATCOM, functioning as a synchronous voltage source, exhibits a superior V-I characteristic by being able to maintain the maximum capacitive output current at any system voltage down to zero. Series compensators. The compensating voltage provided by the Thyristor-Switched/Controlled Series Capacitor is a function of the line current. The compensating voltage of the SSSC, as a reactive voltage source, can be maintained at any value independent of the line current, or can be varied with the line current to emulate a compensating capacitor. Moreover, the SSSC can inherently provide capacitive as well as inductive compensation, which makes it highly effective in power flow control, as well as in power oscillation damping. Phase shifters. Thyristor-Controlled Phase-Shifters based-on tap-changer transformers draw both the active and reactive power from the ac system they exchange through series voltage injection in executing angle control. The back-to-back converter arrangement, emulating the operation of an ideal motor-generator set, internally generates or absorbs all the reactive power the phase shifting requires. This minimizes line voltage variation and losses.

4. The back-to-back arrangement of two voltage-sourced converters emulates the operating characteristics of an ideal motor-generator set with four-quadrant operation in the reactive versus active power plane. The use of this versatile arrangement with shunt and series, series and series, and shunt and shunt coupling results in three different and extremely powerful transmission controllers: the UPFC, IPFC, and BtB Tie. # The UPFC with a shunt and a series coupled converter can provide concurrent or individual voltage, impedance, and angle regulation or, alternatively, independent real and reactive power flow control. With this operating flexibility it can readily adapt to particular short term contingencies or future system modifications. The IPFC with two or more back-to-back connected converters, each coupled in series with a different line, provides means to transfer real power among lines, and execute independently controllable reactive compensation for each line. This capability makes it possible to equalize power flow among lines or manage power flow according to demand and line capacity. In general, the IPFC provides a highly effective scheme for the optimized use of a multi-line transmission system. The two voltage-sourced converters in back-to-back connection with shunt coupling at each side provides a perfect arrangement for tying two power system asynchronously or synchronously with fully controllable power transfer and terminal voltage regulation at both sides.

1.5 Issues and Options of Future Transmission Control The majority of installed power electronics-based (FACTS) Controllers are conventional thyristorcontrolled Static Var Compensators (SVCs) and a much smaller number of Thyristor-Controlled Series Capacitors (TCSCs). Since the 1990-s, there has also been installed an appreciable number of STATCOMs and also some prototype installations of SSSC and multi-function power flow controllers, UPFC, IPFC, and BtB. In contrast to these newly developed converter-based Controllers, ThyristorControlled Var Compensator, Series Capacitors and Phase Shifters, do not have multi-functional capability. They can execute only the single function (i.e., shunt compensation, series compensation, or
5

53 angle control) for which they were installed. They may, however, have the limited capability to change their operating mode, e.g., the SVC may be set up to regulate voltage or reactive power, the TCSC to control line impedance or power flow, etc. The single function operating capability is also true for the STATCOM and SSSC with the significant difference that these can be expanded to multi-function transmission control capability by the addition of one or more converters in back-to-back configuration. The UPFC and IPFC are generally characterized by multi-functional transmission control capability, e.g., simultaneous voltage regulation, active and reactive power flow control, as well a s by operational convertibility, e.g., to be operated as independent STATCOM and SSSC, IPFC or UPFC or even as BtB Tie. Moreover, within any selected operating mode, they offer a number of selectable control functions, e.g., transmission angle and impedance or direct active and reactive power control. The customary and still prevailing utility practice has been to install a dedicated transmission Controller and operate it in a fixed mode with substantially steady-state reference signals controlling local system variables (voltage, current, impedance etc.). In other words, the Controller usually receives no other indication of system problems and contingencies than the change in some locally sensed variables. Its response to the received indication is fast and automatic, trying to correct the detected system discrepancy. However, this may not be the needed, or even a compatible, response to the problem under a possibly different post fault conditions and system parameters (e.g., system impedance). Moreover, although some SVC installations can initiate the operation of mechanically switched capacitor banks for steadystate compensation, the Controllers would generally have no direct information of the actions of other controlled power system devices. In fact, it could preclude the operation of those with slower response, or interact with those of comparable speed. There are possible control strategies measures and protection schemes to provide an acceptable steady-state solution to these problems. However, in order to get predictable and proper dynamic response, as well as maximal steady-state benefits, the direct coordination of fast Controllers and other devices, located close enough to interact, in addition to real time information on prevailing system conditions, will not only be operationally advantageous, but absolutely necessary with the ever growing number, and increasingly powerful transmission Controllers deployed. Beyond the interaction prompted coordination and real time system information, the anticipated deployment of ever increasing number transmission Controllers, particularly the converter-types with multiple functional capabilities, provides a new basis for re-conceptualizing the control of the bulk power transmission system as a dynamic entity. Within this entity, power electronics-based Controllers could be employed to establish optimal power flow patterns for economic benefits consistent with network security requirements under prevailing system conditions and handle dynamic disturbances by concerted control action. From the technical standpoint the most plausible system control structure to manage the transmission grid as a dynamic entity is hierarchical, built up with layers of controls from the level of each individual Controller ultimately to the top level central control that strategically coordinates the operation of the overall network. This hierarchical control, illustrated in Figure 1.29, could be envisioned in several ways. One extreme would be to mandate that the top level central controller direct the operation of each operational Controller in real time. The opposite extreme would be to operate the Controllers independently from local system data and providing coordination from local system operators in the form of steady-state reference signals (i.e., the method utilities presently use). The first extreme would clearly raise serious questions regarding the security of the total system since errors in the data collection and processing, malfunctions in the central control or in the communication system, could result in major and potentially system wide outages. The second extreme would not completely utilize the inherent control and operating flexibility of the power electronics-based Controllers of the system for full economic or
5

54 security benefits. A reasonable solution may be similar to that used in organizational managerial structures in which the overall strategic decisions are made at the top level, the tactical decisions at mid-level and the implementation details are executed at local levels. In this scenario one could visualize that the central control would determine the main transmission paths for the overall system according to demand and economic objectives, taking into account equipment availability and other constraining factors, and would coordinate the operation of neighboring transmission areas. The area controls would optimize the defined power flow through their system by setting, via appropriate communication links, their Controllers into the desired operating mode and by providing reference signals for them for the selected system variables controlled. The area control would supervise the operation of the area transmission system by collecting system status data as well as critical operating and availability information from each Controller. Under area contingency or system flow change demand, the area control would re-optimize its system operation by appropriately changing the functional operating mode of the Controllers and their reference inputs. In the case of an area disturbance, the area control would dynamically change the reference inputs to the appropriate Controllers in order to achieve a coordinated counter action to reestablish rapidly rotational and voltage stability. Within this scheme, the individual Controller would operate self-sufficiently from locally measured system variables and its output would be controlled, in the operating mode defined, by the reference input the area control system provides. In the case of an area control outage or malfunction, the local Controllers would automatically fall back to independent operation determined by the prevailing system steady-state references. Of course, under all conditions, the individual Controllers would accept overriding instruction from authorized operating personnel.

Central Control

Area Control 1
Area Control 2
Area System 1 with FACTS Controllers

Area Control n

Intertie

Area System 2 with FACTS Controllers

Area System n with FACTS Controllers

Figure 1.29 Conceptual hierarchical control scheme for interconnected power systems using power electronics-based transmission controllers.

The establishment of a viable hierarchical system control for a large transmission grid would be a large undertaking. It would require fundamental changes in control strategies, development of new system security procedures and wide-area measurement capabilities together with highly reliable communication links and protocol. It would also require the development of analytical (software) tools to process real time information for system security, voltage and rotational stability assessments.

55 In addition to the many challenging technical issues, the hierarchical control of power electronics-based power transmission also poses ownership and responsibility problems for the still evolving utility system structure. There could be many questions posed and many different answers offered, based on different transmission network and ultimate organizational structures assumed. It would be speculative to pursue these here. However, the incentives are clearly established: the broad application of power electronicsbased Controllers with other appropriate advanced technologies promises to facilitate competitive electric power system with modern, highly flexile transmission network that fully utilizes system assets while maintaining the required security and reliability.

References [1] Miller, T.J., Editor, Reactive Power Control In Electric Systems, Chapter 2, John Wiley & Sons, 1982. [2] Gyugyi, L., Power Electronics in Electric Utilities: Static Var Compensators, Proceedings of the IEEE, Vol. 76, No. 4, April 1988. [3] Hingorani, N.G., High Power Electronics and Flexible AC Transmission System, IEEE Power Engineering Review, pp. 3-4,July, 1988. [4] Gyugyi, L., Solid-State Control of Electric Power in AC Transmission Systems, International Symposium on Electric Energy Conversion in Power Systems, Invited paper, No. T-IP. 4, Capri, Italy, 1989. [5] Gyugyi, L., A Unified Power Flow Control Concept for Flexible AC Transmission Systems, IEE Fifth International Conference on AC and DC Power transmission, London, Conference Publication No. 345, pp 19-26, 1991, also IEE PROCEEDINGS-C, Vol. 139, No. , July 1992. [6] Gyugyi, L., Dynamic Compensation of AC Transmission Lines by Solid-State Synchronous Voltage Sources, IEEE Transactions on Power Delivery, Vol. 9, No. 2, April 1994. [7] Gyugyi, L., et al., Static Synchronous Series Compensator: A Solid-State Approach to the Series Compensation of Transmission Lines, IEEE Transactions on Power Delivery, Vol. 12, No. 1. January 1997. [8] Edris, A., et al., Controlling the Flow of Real and Reactive Power, IEEE Computer Applications in Power, January 1998. [9] Gyugyi, L., et al., The Interline Power flow Controller Concept: A New Approach to Power Flow Management in Transmission Systems, IEEE/PES Summer Meeting, Paper No. PE-316-PWRD-0-071998, San Diego, July 1998. [10] Song, Y.H., and Johns, A.T., Editors, Flexible ac transmission systems (FACTS), Chapter 1, The Institution of Electrical Engineers, 1999. [11] Hingorani, N.G., and Gyugyi, L., Understanding FACTS, The Institute of Electrical and Electronics Engineers, Inc., 2000.

56 List of Figure Titles

Figure 1.1 Electric power system structure. Figure 1.2 Representation of a lossless transmission line. Figure 1.3 Model of a two machine power system with inductive line (a), corresponding phasor diagram (b) and power transmission vs. angle characteristic (c). Figure 1.4a Variation of mid-point voltage and line current with transmission angle d. Figure 1.4b Variation of the transmitted real power P and the (receiving-end) reactive power Q with transmission angle d. Figure 1.5 Multi-path power transmission (a), and corresponding circuit elements determining power flow distribution (b). Figure 1.6 Two machine power system with an ideal mid-point reactive shunt compensator (a), corresponding phasor diagram (b), and transmitted power vs. transmission angle characteristic (c). Figure 1.7 Two machine system with ideal reactive shunt compensators providing multiple line segmentation, and associated phasor diagram Figure 1.8 Two machine power system with series capacitive compensation (a), corresponding phasor diagram (b), and transmitted power vs. transmission angle characteristic (c). Figure 1.9 Two machine power system with a phase shifter (a), corresponding phasor diagram (b), and transmitted power vs. transmission angle characteristic (c). Figure 1.10 Two machine power system with a generalized power flow controller (a), corresponding phasor diagram (b), control region of attainable active and reactive line power (c), and transmitted power vs. transmission angle characteristic (d). Figure 1.11 Illustration for effect of line compensation: (a) reactive compensation primarily changes active power, and (b) active power compensation primarily changes reactive power flow. Figure 1.12 Illustration of the equal area criterion for transient stability: (a) power system model, and (b) transmitted power vs. transmission angle characteristic under fault condition. Figure 1.13 Equal-area criterion to illustrate the transient stability margin for a simple two machine system, (a) without compensation, (b) with an ideal mid-point compensator, (c) with a series capacitor, (d) with a phase shifter, and (e) with a generalized power flow controller. Figure 1.14 Waveforms illustrating power oscillation damping by shunt compensator, series compensation, phase shifter, and generalized power flow controller: (a) generator angle, (b) transmitted power, (c) var output of a shunt compensator, (d) degree of series compensation, (e) angle variation of a phase shifter, and (f) active power variation by generalized flow controller. Figure 1.15 Variation of voltage stability limit with load and load power factor (a), extension of voltage stability limit by reactive shunt compensation (b), and extension of voltage stability limit by series capacitive compensation (c). Figure 1.16 Representation of an ideal shunt compensator by a controllable reactive current source (a) and its realization by a controllable reactive admittance (b), and by a controllable voltage source with reactive tie impedance (c). Figure 1.17 Representation of an ideal series compensator by a controllable reactive voltage source (a) and its realization by a controllable reactive impedance (b), and by a controllable reactive voltage source with transformer coupling (c). Figure 1.18 Conceptual realizations of an ideal phase shifter or generalized power flow controller by in-phase and quadrature tap-changers (a), and by an ideal synchronous motor-generator set (b). Figure 1.19 Static Var Compensator employing thyristor-switched capacitors and thyristor-controlled reactors (a), and its V-I regulating characteristic (b). Figure 1.20 STATCOM employing a voltage-sourced converter (a), var output control by output voltage variation (b), and V-I regulating characteristic (c). Figure 1.21 Controllable series compensator scheme using thyristor-switched series capacitors (a), or a thyristor-controlled reactor parallel with a fixed capacitor (b), and transmitted power vs. transmission angle characteristic with variable series capacitive compensation (c). Figure 1.22 SSSC employing a voltage-sourced converter (a), and its transmitted power vs. transmission angle characteristic (b). Figure 1.23 Thyristor-controlled quadrature booster type of phase shifter (a), and its transmitted power vs. transmission angle characteristic (b). 5

57
Figure 1.24 Realization of ideal synchronous motor-generator set by two voltage-sourced converters in back-to-back connection. Figure 1.25 UPFC scheme using two back-to-back voltage-sourced converters (a), variation of active and reactive line power as a function of control angle r (b), attainable transmitted active power and reactive line power vs. transmission angle characteristics by control (c), and circular control region for P and Q at a given transmission angle (d). Figure 1.26 Basic IPFC scheme for two lines (a), and active and reactive power vs. transmission angle characteristics attainable for the two lines (b). Figure 1.27 Multi-line line IPFC scheme (a), and generalized multi-line IPFC with line independent active power support and bus voltage regulation (b). Figure 1.28 Asynchronous/synchronous tie scheme using two back-to-back voltage-sourced converters, and the corresponding power transmission and voltage regulation terminal characteristics. Figure 1.29 Conceptual hierarchical control scheme for interconnected power systems using power electronics-based transmission controllers.

DRAFT CHAPTER 12, VOLTAGE-SOURCED CONVERTER-BASED DC TRANSMISSION SYSTEMS

C-1

12

VOLTAGE SOURCED CONVERTER BASED DC TRANSMISSION - VSC TRANSMISSION

12.0 Introduction
HVDC Transmission was first put into commercial service in 1954 and has since been used extensively for the interconnection of asynchronous ac networks and for the transmission of power over long distances. The technology used for the initial HVDC transmission schemes relied on the presence of an ac voltage in the network for its correct operation, and is known as line commutated converter (LCC) HVDC technology. In this chapter, HVDC schemes using this technology will be referred to as LCC HVDC. This technology is still used extensively for HVDC transmission. The total rating of LCC HVDC schemes installed by the end of 2005 was in excess of 60 GW, the largest scheme having a rating of 6300MW and operating at 600 kVdc. The use of voltage sourced converters for dc power transmission (VSC Transmission) was introduced with the commissioning in 1997 of the 3MW, 10kVdc technology demonstrator at Hellsjn, Sweden. CIGR has given this new type of dc transmission the name VSC Transmission, and this term will be used in this chapter [1]. Figure 12-1 shows a simplified diagram of a VSC Transmission scheme connecting two ac grids.
ac grid A U LA Z LA X Conv A VSC A DC transmission line Rdc I dc +U dA UdB + VSC B ac grid B U LB X Conv B Z LB

~ =

= ~

Sending End
Figure 12-1 Simplified diagram of VSC Transmission scheme

Receiving End

The demonstrator proved the feasibility of the technology and subsequently, a number of commercial schemes using this technology have been installed. At the end of 2005 ABB was the only manufacturer supplying VSC Transmission schemes. The total rating of VSC Transmission schemes in service at the end of 2005 was 890MVA, a growth rate significantly exceeding that of the early LCC HVDC technology. The largest VSC Transmission scheme in service at the end of 2005 had a rating of 330 MW and operated at 150 kVdc.

12-1

12.1 Differences in Features and Characteristics from Conventional HVDC

The following subsections provide a brief overview of the differences in features and characteristics between a VSC Transmission scheme and a LCC HVDC scheme. Further descriptions and details will be provided about VSC Transmission in subsequent sections. 12.1.1 Technology VSC Transmission is, as its name implies, based on the use of Voltage Sourced Converters (VSC) [2][3]. An ideal VSC has a constant dc voltage at its dc terminals, and creates an alternating voltage at its ac terminals by the switching of the converter. The Line Commutated Converter as used in LCC HVDC is normally a Current Sourced Converter (CSC). An ideal CSC has a constant direct current at its dc terminals and creates an alternating current at its ac terminals by the switching of the converter. In practice the supposedly constant voltage or current will in both cases have some variation during operation, as the voltage/current source will have finite impedance. The variation in voltage/current has to be taken into account in the determination of the actual wave shapes. VSCs have been used for many years in motor drives, and the development of powerful semiconductors, with turn-off as well as turn-on capability, and with high speed switching capability, has enabled modern motor drives to become compact and versatile. The semi-conductor used most frequently today in Voltage Sourced Converter applications is the Insulated Gate Bipolar Transistor (IGBT), but several other promising devices are also available or under development [4]. The maximum device rating commercially available at the end of 2005 is 6500V and a current capability of up to 2kA. The IGBT is a controlled device, which means that the impedance between its collector and emitter is determined by the control signal applied to its gate. This feature is used to enable the device to turn off at times other than those corresponding to a naturally occurring current zero, in particular, a VSC can deliver power to a passive network. The IGBT is capable of blocking voltage only in the forward direction. Therefore, a diode is used in parallel to the IGBT. This diode prevents the dc voltage from reversing polarity, and enables current to flow in the opposite direction. When changing the direction of power flow between the two terminals in a VSC Transmission scheme the direction of the direct current flow is reversed. The direction of power flow can be smoothly changed through zero, and there is no constraint on the duration of operation at any power within the rating of the converters. The power semi-conductor used in todays LCC HVDC schemes is the thyristor, which can be triggered into conduction by a low power gate signal when the device is forward biased, and which will turn off when the current through the device attempts to reverse. Since the thyristor cannot be turned off on command, a LCC HVDC scheme requires a rotating EMF in the network for satisfactory operation. The thyristor needs a short time interval (a fraction of a millisecond) from the instant it has turned off until it can withstand forward voltage again. Devices capable of being turned on by a light signal are now available, in addition to devices requiring an electrical signal. The power rating of thyristors available for LCC HVDC schemes is very large, with individual devices having a capability of 8.5kV withstand voltage and 4kA current capability. Devices with voltage capability up to 12kV have been developed, but the current capability is then significantly reduced. Since the thyristor can withstand voltage in both directions, the dc voltage can be reversed. When changing the direction of power flow between the two terminals in a LCC HVDC scheme the polarity of the direct voltage is changed, since the thyristors can 12-2

conduct current only in one direction. An LCC HVDC scheme is not normally able to operate continuously at a direct current below about 5% of the rated current, because the harmonics results in discontinuous current operation. As stated above, the semi-conductors in the VSC can be turned on and off irrespective of the state of the current flow through the converter. Therefore, it is possible to control the converter such that both the active and the reactive power flow at the converter ac terminals is determined in accordance with operational requirements. This means that the VSC can be operated in all 4 quadrants of the P and Q diagram. In other words, the VSC can be operated as a rectifier or an inverter, and as a generator or absorber of reactive power, and in any combination of active and reactive power, within the rating of the converter. Naturally, the active power flow at the terminals of a VSC Transmission scheme must be coordinated, such that the active power is balanced. The converters in a LCC HVDC scheme operate with a lagging power factor, typically absorbing reactive power equivalent to approximately 50% of the active power. Thus, the converters in a LCC HVDC scheme can only be operated in 2 quadrants of the PQ diagram. AC harmonic filters and shunt capacitors are used to offset the converters reactive power absorption, such that the overall power factor is acceptable to the ac network. The ac harmonic filters and shunt capacitors are normally switched in and out by means of circuit breakers as the dc load varies. It is possible to use a capacitor in series with the ac terminals of the converter, and this can significantly reduce both the magnitude and variability of the reactive power absorbed by an LCC HVDC converter. However, unless the value of the series capacitor can be varied, the converters reactive power absorption cannot be changed without also changing the active power, unless ac harmonic filters or shunt capacitors are switched. Different topologies have been used or proposed for the converters in a VSC Transmission scheme. The topology for the first schemes was a 2-level converter, in which the ac terminals are switched between the two dc terminals. Some later schemes use a 3-level converter, in which the ac terminals are also switched to the potential at the midpoint of the dc capacitor. The 3level topology reduces the voltage across each valve, and also reduces the base harmonic content in the output voltage. All LCC HVDC schemes are based on the use of a 3-phase, 2-level converter, which is known as a 6-pulse converter. Modern LCC HVDC schemes use basic building blocks of two such 6-pulse converters, combined in series on the dc side and in parallel on the ac side. Converter transformers providing a 30 degree phase shift between the two basic 6-pulse groups can be used to provide harmonic cancellation, leaving only main ac side harmonics of the order 12n1, where n is an integer. The turn-off and high switching speed capability of the VSC valves in a VSC makes it possible to turn each valve on and off several times in each power frequency cycle. This feature can be used to eliminate or significantly reduce the low order harmonic content of the voltage on the ac and the dc side of the VSC. A number of different control methods can be used for the reduction of lower frequency harmonics, leaving primarily the higher frequency harmonics, which can be reduced to acceptable levels more easily by passive filters than lower frequency harmonics. Due to the controllability of reactive power a VSC Transmission scheme does not require switchable ac harmonic filters. This means that the site area of a VSC Transmission terminal is 12-3

dictated largely by the size of the converter, which tends to be of similar size to that of an LCC HVDC converter. Accordingly, the overall site area of a VSC Transmission terminal is significantly smaller than the site area for a LCC HVDC terminal, typically occupying 40% or less of the area of a LCC HVDC terminal. 12.1.2 Operating Features In order for a LCC HVDC scheme to operate satisfactorily the short circuit level of the ac system at the point of connection needs to have a minimum level. The actual minimum level depends on the characteristics of the converter control and on the design of the converter itself. Effectively, the issue is one of ac network voltage stability during system dynamics. For a conventional LCC HVDC transmission scheme to operate satisfactorily the minimum short circuit level typically needs to be about 2.5 times the rating of the HVDC scheme. A low commutating impedance for the LCC HVDC converter, e.g. by using low transformer impedance and/or a series capacitor, enables satisfactory operation at a lower short circuit level at the point of connection. Similarly, arranging the converter control characteristics such that a drop in ac voltage results in a reduction in reactive power absorption, enable the converter to operate satisfactorily at lower short circuit level at one of the terminals. A VSC Transmission scheme can provide reactive power support at both terminals independently of the active power transmitted between the terminals. Therefore, it can stabilize the ac network voltage and a minimum short circuit level at the point of connection is not required. As a consequence the terminals of the VSC Transmission scheme can be located at weak point in the ac network, e.g. remote from the core of the network. Naturally, system studies must be performed to determine the acceptability of the resulting load flows and voltage profiles in the ac networks. As the power flow on a LCC HVDC scheme varies from minimum power to maximum power and back again, ac harmonic filters and shunt capacitor banks are switched in and out. The switching control should at all times and power flows keep the harmonic distortion at an acceptable level, and the steady state ac voltage at the point of connection at an appropriate level for the power flow in the ac network. The size of ac harmonic filters and shunt capacitors must be selected such that the voltage steps caused by filter switching do not cause unacceptable flicker in the ac network. If the power flow on the HVDC scheme changes very frequently, then the number of operations of the circuit breakers can be relatively large, resulting in the need for frequent inspection and maintenance of the circuit breakers. A VSC Transmission scheme typically uses a fixed ac harmonic filter, and the converter is rated to provide the varying reactive power requirements for the envisaged operating conditions. In this case circuit breaker operation is required only at start-up, shut-down, and fault clearance, and the inspection and maintenance needs will typically be less than for a normal ac substation circuit breaker. The breaker switched ac harmonic filters and shunt capacitor banks required for steady state ac voltage control in an LCC HVDC scheme can result in large temporary over-voltages if power transmission suddenly stops, e.g. during dc line faults or converter faults. For some events the converters can continue to circulate direct current to absorb reactive power, thereby reducing the amplitude of the over-voltage, but for other events the current must be stopped immediately, e.g. 12-4

to prevent equipment damage or remove dangerous conditions. The over-voltage could also be reduced by disconnection of the filters and capacitor banks, but this would delay resumption of power transmission until the capacitors had been discharged, typically several minutes. Since a VSC Transmission scheme uses only a relatively small ac harmonic filter, and the healthy converter can continue to control reactive power without circulation of direct current. Faults within the converter would normally result in tripping of the main circuit breaker, thereby removing not only the converter but also the ac harmonic filter. Therefore, the ac over-voltage caused in the event of a sudden stop in power transmission is typically very small. In the event of a sudden drop or phase shift in the ac voltage amplitude, the LCC converter may not be able to complete the commutation of the inverter before the voltage across the outgoing valve becomes forward biased, and therefore the valve will continue to conduct. As the valves continue to be fired, a short circuit between the dc terminals of the 6-pulse bridge will be caused. This event is called a commutation failure, and results in a dip in the ac voltage and temporary dc over-current, both of which are quickly reduced by control at the rectifier. Typically normal commutation and power transmission can be restored within two power frequency cycles, and the event has no significant impact on the operation of the ac system. Since the VSC valves can be turned off as well as being turned on, the operation of a VSC Transmission scheme is not significantly influenced by voltage dips or other transient ac disturbances. In particular, a VSC Transmission scheme does not suffer commutation failures. In the event of a short circuit on the dc side of a LCC HVDC scheme, the converters can quickly stop the current flowing into the fault, merely by stopping the firing of the thyristor valves. The quick reaction of the converter control and protection will minimize any damage caused by the fault. If the fault is on a dc overhead line, then the power transmission is stopped for a sufficient period of time to allow de-ionization of the arc, typically 200-300ms, and transmission is then resumed. Several re-start attempts, with increasing de-ionization periods, can be used if desired. The diodes used in a VSC would continue to conduct current into a fault on the dc side, even when the IGBTs are turned off. A direct current circuit breaker could be used to stop the current flow, but such breakers are expensive and not generally used. Therefore, the ac breakers at both ends of the VSC Transmission scheme must be opened in the event of a fault on the dc line. The scheme then has to be re-started once the fault has been removed. All commercial VSC Transmission schemes have so far used cables rather than overhead lines, which reduces the number of faults on the dc side. 12.1.3 Application Benefits and Disadvantages The technical and operational differences described above in 12.1.1 and 12.1.2 results in a VSC Transmission scheme having a number of application benefits compared with a LCC HVDC scheme. These application benefits are briefly described as follows: The ability of the VSC to control reactive power independently of the active power flow between terminals means that it can support ac voltage at its connection point. This can provide significant benefits to the ac network, e.g. improving voltage stability and increasing load-ability of the network.[5] [6].

12-5

The rapid speed of control of active and reactive power resulting in part from PWM operation can be used to improve the power quality at the point of connection, e.g. the reduction of flicker from industrial plant. [7] The self commutation and reactive power control means that a VSC Transmission scheme can be connected at very weak points in the ac network, and can be the sole supply of power to a network which does not have synchronous generation or compensation. This capability can be useful also in a conventional ac network, where a VSC Transmission terminal can provide black start (passive network start) capability for the restoration of the network after a major blackout. The land area occupied by the VSC Transmission terminals are considerably smaller than that occupied by the LCC HVDC converter stations of equivalent rating. The constant polarity of the direct voltage means that multi-terminal VSC Transmission schemes can be designed with greater operational flexibility than when using LCC HVDC, where the polarity of a converter would have to be changed if its role were changed from a rectifier to an inverter, or vice versa. The constant polarity of the direct voltage means that dc cables do not have to be capable of withstanding polarity reversal. This enables a variety of polymeric cables to be used. Polymeric cables are smaller and lighter than oil impregnated cables, particularly when heavy mechanical protection or armoring is not necessary [8].

These advantages will be further elaborated in section 12.2. The main application disadvantage of a VSC Transmission scheme compared with a LCC HVDC scheme is the higher power loss in the converter (see section 12.7.3). However, the VSC Transmission schemes reactive power control capability may reduce the reactive power flow in the ac network, thereby reducing the overall power loss.

12.2 Application Areas

HVDC transmission has been used since its first introduction in 1954 to provide interconnections between ac networks and the transmission of bulk energy over long distances. The benefits of interconnections include: Better utilization of the installed generation in the two networks by taking advantage of the diversity in loads and the characteristics of generation. Reduction in overall spinning reserve. Emergency power exchanges between the networks. The power flow on the interconnector is fully controlled. This feature means that loop power flows can be avoided. It is possible to interconnect asynchronous networks.

The benefits arising from the use of HVDC as an interconnector include:

12-6

When the interconnector electrical distance is very long, e.g. >800km overland or >70km submarine, HVDC transmission may provide a more economic solution, both in terms of capital cost and power loss.

VSC Transmission can in principle be used in all the applications for which LCC HVDC are currently used. However, unless the VSC Transmission technology develops further to enable the use of dc overhead lines and higher transmission voltage, the long distance overland bulk power transmission market, e.g. 1000MW+ and 800km+, will continue to be provided only by LCC HVDC schemes. Some application areas, in which the technical characteristics of VSC Transmission may be of particular benefit, are described further in the following subsections. 12.2.1 Interconnection of Geographically or Otherwise Isolated Networks Communities in sparsely populated areas, such as the north of Canada or Alaska, or on small islands, may rely on diesel generation for their electricity supply. The cost of electrical energy in such locations may be very high, because advantage cannot be taken of the benefits of largescale generation available in major networks. Additionally, generation of the energy using small diesel generators is more damaging to the environment, than using large-scale generation in a major network, with its more efficient generators and operating modes. Therefore, transmission of energy from the large network may be economically and environmentally attractive. VSC Transmission may offer a good solution to such power transmission if ac overhead line or submarine cable transmission is not feasible, e.g. if the distance is large, or if an underground cable solution is considered advantageous. An underground cable solution rather than an overhead line, may be advantageous for several reasons: It is less intrusive on the landscape than an overhead line, particularly if the cable can be installed along the route of an existing road, or the land above the cable can be returned to its original use after installation of the cables (e.g. farming). The lower visibility of a cable installation is likely to be more acceptable to the public, thus resulting in planning approval being granted more quickly. It is not subject to flashover due to pollution problems, e.g. due to salt fog near to the coast, or to fertilizer application in farming areas. It is likely to be more reliable than an overhead line, because it has fewer exposed parts. It does not produce electric fields or varying magnetic fields, and the magnetic field produced by the direct current typically results in a steady field with a magnitude significantly less than the Earths magnetic field.

Installation of the polymeric cables can be achieved very quickly if the ground conditions are such that the cables can be ploughed directly into the ground using special machinery. The installation of the two 70km long cables for the Gotland project was achieved in 90days. Figure 12-2 shows the cable laying operation for the Murray Link VSC Transmission scheme in Australia.

12-7

Figure 12-2 Cable Laying for the Murray Link VSC Transmission scheme

VSC Transmission is advantageous compared to a LCC HVDC solution for the supply of power to an isolated/passive network, because synchronous compensators are not necessary for the operation of the VSC. This means that significantly less maintenance will be required, not only because the VSC solution has fewer components than a LCC HVDC scheme, but also because of the omission of a rotating machine. Additionally, whilst the power loss of a VSC Transmission scheme can be 2 or 3 times higher than that of a LCC HVDC scheme, the difference would be reduced in this application by the relatively large power loss associated with the operation of a synchronous compensator. It should be noted that the frequency variation in the network may be significantly larger than when the network is fed from its own diesel generation. This is because a fault causing a large reduction in the ac voltage in the main network will reduce the interconnectors transmission capability. The VSC scheme can temporarily reduce the ac voltage to reduce the power flow, but if large synchronous or asynchronous machines are used in the network, these may slow down very rapidly, as they start to act as generators. The re-acceleration of the machines may impose a large duty on the VSC, and may need to be taken into account in the rating of the scheme. Appropriate control and protection strategies need to be determined taking into account the duration of faults and the characteristics of the loads in the network. 12.2.2 Interconnections between Weak Power Systems For weak ac network applications the control of the ac network voltage is particularly important. When a LCC HVDC scheme is used, the large ac harmonic filters and shunt capacitors used to provide reactive power compensation actually reduce the ac voltage stability of the system, since the reactive power support reduces as the ac voltage reduces. Therefore, the application of LCC HVDC schemes is typically limited to systems where the short circuit power at the point of connection of the LCC HVDC terminal is at least 2.5 times the rating of the scheme. When the 12-8

short circuit level is lower than this ratio special control methods can be used, but these typically require a higher rating of the converter station equipment. Alternatively, additional reactive power support, in the form of SVCs or STATCOMs, can be applied at the point of connection, to improve the ac voltage stability. As a VSC Transmission scheme is capable of controlling the reactive power at its ac terminals independently of the active power flow between terminals (subject to overall rating limitations), it provides a good solution for the interconnection of networks, where the ac network at one or both terminals is weak. 12.2.3 Reinforcement of Weak AC Tie-Lines for Stability Improvement When an ac transmission line is imbedded in a geographically large ac network with major load and generation centers at its extremities, a fault within either of the centers can cause major power oscillations on the ac tie-line. If the tie line is weak (e.g. because of its length or because of its rating relative to the network), then the power oscillations may exceed the over-current setting of the line protection, resulting in a trip of the line. The power oscillations may also have a significant impact on the ac voltage along the line and at the termination points. The power oscillations can be damped by the insertion of controlled series compensation, or, less effectively, by the use of controlled shunt reactive power compensation. A more effective solution would be the installation of a parallel VSC Transmission scheme. Studies have shown that a parallel VSC Transmission scheme can increase the stability limit on a weak ac tie line by more than the rating of the VSC Transmission scheme. This is achieved by suitable modulation of the active power transfer of the VSC Transmission scheme AND the reactive power at its terminals. [5] 12.2.4 Connection of Distant Loads (Off-Shore Oil and Gas Platforms) As an oil or gas reservoir is emptied, the power required to extract and transport the oil or gas increases. Where the production platform is located off shore, and at significant distance from the coast, diesel or gas turbine generators located on the production platform have typically produced the electrical power required for the operation. However, as the power requirement increases, additional generating plant becomes necessary and consideration may be given to obtaining the electrical power from the on-shore ac network. The advantages of obtaining the power from the shore include: The maintenance intensive generation plant on the platform can be removed. CO2 emissions are reduced since the efficiency of on-shore generation plant is more efficient, and may include renewable generation plant. This may result in substantial tax benefits.

When the distance to the shore is large, transmission by means of ac may not be feasible, and HVDC transmission may be considered. The use of VSC Transmission [9] provides the following advantages: The VSC Transmission equipment is compact, making the installation on existing platforms feasible. 12-9

There is no need for synchronous compensators since the VSC Transmission scheme is selfcommutating, and controlled reactive power can be provided, independently of the active power being transmitted. The VSC Transmission scheme can provide variable frequency power supply to the offshore motors, acting as a variable speed motor drive thereby improving the performance of the offshore equipment.

12.2.5 Connection of Remote Wind-Parks Wind generation is currently one of the most economical sources of renewable generation, and is the fastest growing sector of electrical power generation. As more and more wind generators are installed there is a tendency for the wind farms to be located further away from the population centers, partly because of better wind conditions and partly to reduce the visibility and environmental impact of the wind turbines. Wind conditions are often best in offshore locations and several offshore wind farms with ratings of 500MW and more are currently under consideration. When the wind farm can be connected relatively easily to a strong point in the ac network, then an ac interconnection is often the most appropriate solution. However, when the transmission distance to the main network connection point becomes large electrically, or the ac connection point is in a weak part of the ac network, then the most economic solution may be to use VSC Transmission [10][11]. Whilst a HVDC terminal occupies more space and costs significantly more than an equivalently rated ac substation, HVDC Transmission provides the following advantages: The transmission distance can be much larger, enabling connection at a more suitable point in the ac network. The HVDC scheme will provide de-coupling between the main ac network and the wind farm network, thereby improving the ride through capability of the wind farm during faults in the mainland ac network. The frequency of the offshore network can be allowed to vary with the speed of the wind, increasing the efficiency of the wind farm.

In addition to these general advantages, a VSC Transmission scheme provides the following advantages compared with a LCC HVDC scheme The VSC Transmission scheme can provide reactive power support of the wind farm ac network, improving the power quality and stability of the wind farm ac network. Flicker caused by operation of the wind farm will not be transmitted to the mainland ac network. Power can be transmitted to the wind farm ac network during conditions of little or no wind. This is useful, since power is required at all times to satisfy the auxiliary power requirements for both the VSC Transmission scheme and for the wind generators, e.g. control and protection, safety systems, navigation lighting, cubicle heating, etc.

12-10

The optimum overall design of the wind farm and interconnector would be achieved when taking into account the technical capability of the wind generators and the VSC Transmission scheme. For example, doubly fed induction generators enable variable turbine speed and are capable of providing some element of reactive power compensation, and to some extent these characteristics are a duplication of those of a VSC Transmission scheme. Cost savings and satisfactory technical performance might be achieved by combining a VSC Transmission scheme with induction generators, or by using lower ratings of the converters in the doubly fed generators. Naturally, system studies would be required to determine the overall dynamic and transient performance, before finalizing the design and rating of components.

12.3 Basic Operating Principles and Characteristics

12.3.1 Conceptual Representation Conceptually a VSC Transmission scheme is similar to a zero inertia synchronous machine/generator A coupled through an infinitely variable gear/coupling to a zero inertia synchronous generator/machine B. The machine and the generator can transfer active power from the network at A to the network at B or vice versa. In steady state operation the output power at the generator ac terminals will be equal to the input power at the machine ac terminals, less the power loss in the machine and generator and the transmission line. The machine and the generator are each able to provide controlled reactive power independently of the power transmitted between the two terminals. In an actual implementation, the dc capacitors and the cable capacitance provide a small amount of energy storage, typically equivalent to the rated energy for one power frequency cycles or less. This energy provides some smoothing of the power flow, and helps the control and stability of the VSC Transmission scheme. However, the energy storage in the dc capacitor is only a small fraction of the kinetic energy in a typical generator. 12.3.2 Equivalent Circuit of a VSC Transmission scheme Figure 12-3 shows a simplified diagram of a VSC Transmission scheme connecting two ac networks A and B. The essential elements of the VSC Transmission scheme are: The converters VSC A and VSC B, which perform the conversion between the dc voltage and the ac voltage. The dc capacitors, which provide the voltage source, from which the ac voltage is created by switching of the converter. The transmission link, which enables energy to be exchanged between the dc capacitors at VSC A and VSC B. The inductive impedance between the converter and the ac network, which can be found at both VSC A and VSC B, and which is essential to stable operation of the system.

12-11

ac grid A U LA Z LA X

VSC A

DC transmission line Rdc I dc +U dA UdB + -

VSC B

ac grid B U LB

Conv A

~ =

= ~

Conv B

Z LB

Sending End
Figure 12-3 Simplified diagram of VSC Transmission scheme

Receiving End

Most simulation programs (e.g. PSCAD/EMTDC, PSS/E, DigSilent) provide models of VSC Transmission schemes, and can be adapted to provide the relevant control characteristics. Such models are fine for preliminary studies. However, generally speaking, care should be taken when using such models, as they may not include all features of a real control and protection system, and therefore the response during transients and system dynamics may not reflect the actual performance of a real VSC Transmission scheme. 12.3.3 Steady State Operating Characteristics of a VSC Transmission scheme As is the case for a synchronous machine/generator the current flowing at the ac terminals of the VSC depends on the voltage amplitude and phase angle relative to the network EMF, as well as on the impedance between the VSC and the network EMF. Figure 12-4 illustrates the control of active power through the converter line inductance by the variation of the phase angle. When the phase angle of the converter ac voltage Uconv leads the ac system system voltage UL the VSC injects active power in the ac system. Conversely, when the converter ac voltage lags behind the ac system voltage the VSC absorbs active power from the ac system.
Id Us + Rd Ud DC DC resistor capacitor Pconv I conv Uconv = UL UL Uconv - UL ~ Iconv P conv = 0 P conv < 0 Rectifier operation P conv > 0 Inverter operation Uconv UL I conv

Uconv Xconv

Figure 12-4 Active Power Control

12-12

Similarly, Figure 12-5 illustrates the control of reactive power. When the amplitude of the converter ac voltage Uconv is larger than the ac system voltage UL the VSC injects reactive power in the ac system, i.e. it acts as a shunt capacitor. Conversely, when the converter ac voltage is lower than the ac system voltage the VSC absorbs reactive power from the ac system, i.e. it acts as a shunt inductor.
Uconv UL Uconv UL

Uconv = UL Id Us + Rd U d P conv Iconv

Uconv Xconv

UL

~
Qconv = 0 Float Operation

DC DC resistor capacitor

Iconv Q <0 conv Inductive operation

Iconv Qconv > 0 Capacitive operation

Figure 12-5 Reactive Power control

As the amplitude and phase angle of the VSC ac voltages can be controlled independently of each other, subject to possible rating limitation in the converter and the balance of active power of the converters at the two ends, the active and reactive power output of the converter can be controlled to provide active power transmission as well as independent reactive power control at the terminals. The voltage and current capability of the converter components determine the operating range of the converters. Different components determine the limit at different operating points, and therefore, the operating range can be tailored to meet specific requirements by enhancing some of the components. However, enhancement of some components could make the VSC deviate from the manufacturers standard module. More details of the PQ capability of the VSC Transmission scheme will be given in section 12.5. 12.3.4 Dynamic Capabilities and Application Potential The speed of response of a VSC Transmission scheme can be very fast, with the possibility of reversal of power from 100% import to 100% export, and/or a reactive power change from generation to absorption within 1 power frequency cycle. Such fast changes may however not be acceptable to the ac network, and it is then necessary to slow down the speed of response to ensure that the overall performance is acceptable. The overload capability of most VSCs is relatively limited, and thus the fast speed of response of the controls is important in order to keep the VSC within its acceptable duty during disturbances and transients in the ac network.

12-13

The ability of the VSC Transmission scheme to modulate both the active and the reactive power means that it can provide superior assistance to an ac network, compared with a LCC HVDC scheme or a shunt reactive power device, such as a SVC or STATCOM. The VSC Transmission controls typically include a number of control loops aimed at enhancing the performance of the ac network. When the system disturbance, to which the controller is designed to respond, can be measured at the location of the VSC Transmission terminal, it may not be necessary to use telecommunication in order for the converter to provide correct performance. However, when the VSC Transmission scheme can only provide an indirect attenuation of the disturbance, the use of telecommunication may enable a better response of the VSC Transmission terminal to the disturbance. A. First Swing Stability Potential first swing stability problems usually involve one or more generators, which are connected to a remote ac network. The instability is caused when a remote generator is isolated e.g. due to the tripping of the last remaining ac line to the generator. The remote generator will accelerate, as it cannot evacuate the power still being supplied by the prime mover, and resynchronization may not be possible when the line is re-connected. The loss of the large generator may also cause problems in the remaining ac network, particularly if the remaining local generation is much smaller than the local loads, and the shortfall has to be supplied from the main network via a long ac line. In this case, power swings between the local area and the main network may be so large that the ac line trips due to over-current, leaving the local area isolated, with a resulting rapid drop in frequency. The high speed at which the active and reactive power can be controlled by a VSC Transmission scheme may help avoid a first swing stability problem. However, if the ac system voltage at the VSC Transmission terminal has been reduced to near zero during the short circuit to ground, the scheme cannot influence the generator or the ac network until the fault has been cleared. It is necessary to perform detailed studies to determine the control transfer functions that will best suit the situation. The VSC Transmission scheme may be able to assist the ac network by one or more of the following actions: Rapid reversal of the direction of power flow. Rapid restoration of the ac voltage after fault clearance. Providing an asynchronous connection between the affected part of the network and the main network.

One way to provide maximum benefit to the overall network may be to temporarily sever the synchronous links between the parts of the network, leaving only the VSC Transmission link between the two parts. This would allow the frequency in each to swing freely until the networks settle down and the generator governors and VSC Transmission can return the frequencies to the nominal values and the systems can be re-synchronized. Some load shedding may be necessary to achieve power balance, but a total system collapse will be avoided, and the duration of load interruption will be minimized. 12-14

B. Rotational Oscillation Damping A sudden change in a power system, e.g. a short circuit to ground, can result in a temporary unbalance between the electrical power output of generators and their mechanical power input from the prime mover. This is because the time constant associated with a change in mechanical power input is considerably longer than the electrical power output time constant. The result is an acceleration of the mechanical speed of the generators. Depending on the location of generators relative to the short circuit, and their mechanical and electrical characteristics, their speed will change at different rates. The changes to the generators (and loads) cause different power flow in the network, and results in voltage oscillations and power oscillations between the generators. Typically, electrically close generators will exchange electrical energy, some acting as generators and others as machines, and can be considered as one generator, referred to as a generator area. The power swings between the generators cause mechanical wear in the power plants and reduce the power quality in the network, e.g. causing flicker. The power oscillations in the ac network may also cause overloading of transmission lines and equipment, and if the oscillations persist or becomes of too high amplitude, the protection may trip lines, causing further problems in the network. The power and voltage oscillations must be stopped as quickly as possible to minimize wear and the risk of further problems. The generator voltage regulators and turbine governors can be controlled to assist with the damping of the oscillations. However, if a VSC Transmission scheme has one terminal in one generator area subject to inter area oscillations and another in another area, or in another network, it may be able to provide substantial damping of the oscillation. The VSC Transmission scheme can provide damping by: Modulating active power into the oscillating area and keeping the voltage at the connection point constant. Modulating reactive power at the point of connection and keeping the injected active power constant. (SVC type damping). Modulating active and reactive power independently of each other.

The control algorithm must be determined based on a full understanding of the dynamics of the network, generators and loads. This may require extensive dynamic system studies, with and without the simulation of the VSC Transmission scheme. In some cases the dynamics will include several oscillation modes, making the determination of the optimum algorithm more difficult. Since the VSC Transmission scheme can provide two independent degrees of freedom (active and reactive power) in its operation it can provide superior damping compared to a LCC HVDC scheme and/or to a SVC. C. Voltage Stability and Prevention of Voltage Collapse A satisfactory operating voltage profile in a network is typically determined by automatic voltage regulators on machines, operation of tapchangers on transformers, and switching of reactive power elements (reactors and capacitors). As the network loading increases the reactive 12-15

power generated by the transmission lines and cables reduce, and more and more of the voltage supporting equipment will be brought into use. In extreme circumstances equipment and line trips can cause a deficit of reactive power, leading to a reduction in the ac voltage. As the ac voltage falls, the shunt capacitors used for reactive power support becomes less and less effective (their output drops with the square of the ac voltage). If the under-voltage persists for a long time, some equipment will eventually be tripped by under-voltage protection, as a sudden collapse in ac voltage could occur if under-voltage load shedding is not performed. A VSC Transmission scheme can help improve the voltage stability of an area in different ways: By providing reactive power support to the ac network. The reactive power support available from a VSC Transmission scheme is superior to that provided by shunt capacitors, as the reactive power current can be kept at the same amplitude independently of the ac voltage. If the VSC Transmission scheme is in parallel to an ac line it may be possible to divert power flow from the ac line to the VSC Transmission line, thereby reducing the reactive power absorption of the ac line.

D. Power Quality Power Quality issues includes voltage sags and spikes, power interruptions, harmonics, flicker and voltage oscillations. Certain loads, e.g. arc furnaces and rolling mills may cause flicker and other power quality problems for other customers on the network. Faults in the ac network cause voltage sags and phase shifts, the magnitude being larger the closer the fault, and can cause problems for sensitive industrial processes, such as paper mills, semi-conductor processing plants, etc. LCC HVDC schemes can also cause disturbances due to commutation failures, which locally aggravate the disturbance due to remote faults in the ac network. A VSC Transmission scheme can respond very rapidly to voltage sags, increasing its reactive power generation to the maximum permissible level in less than a power frequency cycle. Similarly, the VSC Transmission scheme can respond very quickly to the over-voltage, which typically follows the clearing of a network fault, by increasing its reactive power absorption thereby reducing the temporary over-voltage. The VSC Transmission scheme can also respond and suppress voltage oscillations caused by power swings in the ac network. The ability of the VSC Transmission scheme to reduce voltage sags, temporary over-voltages and to respond to voltage oscillations, can be increased by ensuring that the scheme is normally operated with low reactive power output, such that maximum dynamic capability is available for this duty. To improve the power quality for sensitive industrial processes, consideration can be given to supplying such loads via a VSC Transmission scheme. This could eliminate the voltage sags and over-voltages caused by remote ac system faults. However, in the event of faults close to the VSC Transmission rectifier terminal, the scheme would not be able to maintain power supply, unless significant energy storage were provided on the dc side of the scheme.

12-16

12.4 Practical Implementation A simplified diagram of one terminal of a VSC Transmission scheme is shown in Figure 12-6. A similar terminal would be located at the end of the dc transmission line (in case of a multiterminal scheme, 2 or more terminals would be connected to the dc transmission line). The VSC has a capacitor connected directly across its dc terminals, with no intervening impedance apart from the stray inductance of the connection conductors. The different parts of the terminal

Interface Transformer l U L

Phase reactor U conv ~ lconv =

ld Ud

L AC Filter

Converter

equipment are described in more detail in the following sub-sections.

Figure 12-6 Basic Diagram of one terminal of a VSC Transmission scheme

12.4.1 Two- and Three-Level Converter Structure The Voltage Sourced Converter performs the conversion between ac and dc and vice versa. For short-time transients the dc capacitor at the dc terminals of the converter can be regarded as a constant voltage source. The switching devices interconnect the dc terminals and the ac terminals using a control sequence resulting in an alternating ac waveform at the ac terminals, with a desired fundamental frequency amplitude and phase angle. 12.4.1.1 Two-Level Converter The simplest converter implementation is a 2 level converter as shown in Figure 12-7.

12-17

+
Udc 2 Udc 2

Uac

Udc 2

Uac

Udc 2

t0

t1

t2 t

Figure 12-7 Single Phase 2-level converter with switching output shown

The switching of the VSC valve at time t1 changes the ac output from +U/2 to U/2, and the switching of the valve at time t2 changes the ac output from -U/2 to +U/2. The ac output voltage can only attain two different amplitudes, namely +U/2 or U/2. The diode connected in parallel to the IGBT prevents the direct voltage from changing polarity, since the diode would enter into conduction, if it were to become forward biased, thereby discharging the dc circuit. The current can flow in both directions through the VSC valve, passing through either the IGBT or the diode. This characteristic is one of the main differences between VSC Transmission and LCC HVDC. In case of a short circuit on the dc side the diodes will conduct current into the fault, until the ac voltage at the ac terminals is removed. In order to re-establish power transmission the VSC Transmission scheme would need to be re-started again. In principle it would be possible to switch each valve on and off just once per power frequency cycle to perform the conversion between dc and ac. However, whilst it would be possible to control the phase angle of the fundamental frequency component of the resulting waveform, there would be a fixed ratio between the dc and the ac converter voltage. Thus, in order to control the reactive power, without the use of a tapchanger, it would be necessary to change the dc voltage, which is not desirable on a transmission scheme. Additionally, the ac harmonics would be low frequency and very large. By switching the valves on and off several times each power frequency cycle, additional degrees of freedom are available for the controls. Thus it becomes possible to control both the amplitude and the phase angle of the fundamental frequency component of the ac voltage, independently of the dc voltage, subject to limitation caused by the dc voltage. This method of operation is explained in more detail in section 12.4.2 below. The following paragraphs describe in more detail the basic operation of the 2-level VSC shown in Figure 12-7. Assuming both IGBTs are blocked (i.e. the IGBTs are in the high impedance state) the freewheeling diodes form an uncontrolled rectifier. An external ac voltage source applied as Uac would charge the two dc capacitors via this rectifier to the peak value of the ac voltage Udc/2 = 12-18

Uac peak across the upper and lower dc capacitor, with polarities as shown in Figure 12-7. With the dc capacitors charged and the external source connected, the VSC is ready for operation. The IGBTs can be switched on and off in a desired pattern via the gate signals. During operation, only one of the two IGBTs is in the conductive state and the other is turned off. Turning on both IGBTs at the same time would create a short circuit of the dc capacitors, and must be prevented. In order to avoid such a short circuit, the IGBT in the off-state is turned on shortly after the IGBT in the on-state is turned off. The short instant (about 10 microseconds), during which neither of the IGBTs are on, is called the blanking time, and during this time the current path is provided by the freewheeling diodes. Assuming the upper IGBT is turned on at t = 0, as shown in Figure 12-7, the ac terminal would be connected to the positive terminal of the storage capacitor, resulting in current flow through the upper IGBT. Depending on the pulse pattern applied, the upper IGBT will be turned off after a short time span at t = t1. However, the impedance of the interface transformer and/or converter reactor (Figure 12-6), will maintain the current. Thus, the diode connected in parallel to the lower switch turns on. As a consequence, the voltage Uac changes from plus to minus U/2. The polarity reversal is initiated by turning off the IGBT actively, i.e., while it carries current. The lower IGBT is required to be on whenever the current reverses, to maintain the voltage polarity, and therefore it will get a gate signal following the blanking time. The IGBT will then take over current as soon as the current reverses direction, the timing of which depends on the impedance of the interface transformer and converter reactor, and on the ac system voltage. If current is still flowing in the diode parallel to the lower switch, when at t = t2 it is required to connect the converter ac terminal to the positive terminal of the VSC dc capacitor again, then it is necessary to first ensure that the lower IGBT is turned off. Therefore, the lower IGBT is turned off and the blanking time is allowed to pass, before the upper IGBT is turned on. During the blanking period the current continues to flow through the diode. When the upper IGBT is switched on, a temporary current path from the VSC dc capacitors through the upper IGBT and the lower diode back to the VSC dc capacitors is created. The current flowing in this circuit will extinguish the current in the diode, which will automatically turn off, leaving the upper IGBT as the only current-carrying device, and causing the desired voltage polarity reversal at Uac. The three-phase implementation of a 2-level converter is shown in Figure 12-8.

12-19

U conv-ph

Ud 2 m Ud 2

UL

a b c

~ ~ ~
n

Figure 12-8 Three phase implementation of a 2-level converter

Figure 12-9 shows the terminal voltages on the valve side of the phase reactor for a three phase 2-level VSC in square wave operation.
Uam Ubm Ucm Uab Ubc Uca Uan +2/3Ud +1/3Ud +Ud -Ud +Ud/2 -Ud /2 t

1 fundamental cycle
Figure 12-9 VSC voltages converter sidesquare wave operation

The first three sets of curves are the voltages between phase terminals and the midpoint m of the dc capacitors. The next three are the line-to-line voltages between the ac terminals. The last curve is the phase to neutral (n) voltage in phase a. 12-20

With full wave, square pulse operation, i.e. each valve turning on and off once each cycle of the ac voltage, the waveform is the dual of the converter current for a six pulses LCC converter with zero commutating conductance. 12.4.1.2 Three-Level Converter In the 2-level VSC the ac voltage at the converter ac terminals can attain only the voltage at the two dc terminals. Furthermore, the VSC valves are exposed to the full amplitude of the voltage between the two dc terminals. By subdividing the dc capacitor and the VSC valves it is possible to arrange for the ac voltage at the VSC terminals to move not only to the voltage at the two dc terminals but also to intermediate levels. The number of voltage levels to which the ac terminal voltage can be switched will depend on the number of valves and the number of dc capacitor subdivisions or additional dc capacitors. These arrangements are known as 3-level or multi-level converters, depending on the number of voltage levels that can be achieved. The term multi-level refers to a converter topology where the ac bus can be switched to attain more than three different voltage levels. This section will focus on 3-level converters. Higher level converters can be implemented, but so far only 2-level and 3-level converters have been used for VSC Transmission. In 3-level converters the VSC valves do not normally have to be designed for the full dc terminal-to-terminal voltage. For example, in normal operation each valve in a 3-level converter experiences only 50% of the terminal-to-terminal dc voltage. Figure 12-10 shows a three-phase, 3-level VSC implemented with a Neutral Point Clamped (NPC) topology.

+
U d Ua Ub Uc +Ud -Ud

Neutral mid- point Ud

Figure 12-10 Three-phase, 3-level converter with full wave switching

The converter has three dc terminals to connect to a split or center-tapped dc source. There are twice as many valves as in the 2-level VSC, and additional diodes are required to connect to the 12-21

dc supply center-tap, which is the reference zero potential. With identical valve terminal-toterminal voltage rating, the total dc supply voltage can be doubled so that the output voltage per valve remains the same. The ac waveform shown in Figure 12-10 is the phase-to-neutral voltage, assuming fundamental frequency switching of the valves. The neutral voltage is the voltage at the midpoint of the dc capacitor. The output voltage of the 3-level phase unit can be positive, negative, or zero. Turning on both upper valves in a phase unit produces positive output, while turning on both lower valves produces negative output. Zero voltage is produced when one of the upper and lower middle valves are turned on, and the upper and lower outer valves are turned off, the connection to the center tap of the dc supply being via one of the two diodes. At zero voltage, positive current is conducted by the upper-middle IGBT and the upper NPC diode, and negative current by the lower-middle IGBT and the lower NPC diode. As indicated in Figure 12-10 the relative duration of the positive (and negative) output voltage with respect to the duration of the zero output is a function of control parameter , which defines the conduction interval of the top upper, and the bottom lower valves. The magnitude of the fundamental frequency component of the output voltage produced by the phase unit is a function of . When equals zero degrees it is maximum, while at equals 90 degrees it is zero. Thus, one advantage of the 3-level phase unit is that it has an internal capability to control the magnitude of the output voltage without changing the number of valve switching operations per cycle. The operating advantages of the 3-level phase unit can only be fully realized with some increase in circuit complexity, as well as more rigorous requirements for managing the proper operation of the converter circuit. These requirements are related to executing the current transfers (commutation) between the four (physically large) valves, with well-constrained voltage overshoot, while maintaining the required di/dt and dv/dt for the semiconductors without excessive power loss. An additional requirement is to accommodate the increased ac ripple current with a generally high triplen harmonic content flowing through the mid-point of the dc supply. This may necessitate the use of a larger dc storage capacitor or the employment of other means to minimize the fluctuation of the mid-point voltage. However, once these problems are solved, the 3-level phase unit provides a useful building block to structure high power converters. The conduction periods for the inner and the outer valves is different, and therefore it is possible to use two different designs of VSC valve for the two positions.

12-22

U am U bm U cm U ab U bc U ca U an +4/3Ud

+U d

-Ud

+2Ud

Ud

+2/3Ud

1 fundamental cycle Figure 12-11 Waveshape of full wave switched 3-phase, 3-level NPC converter

Figure 12-11 shows the waveshape of a three-phase, 3-level NPC converter with full wave switching. The waveshapes clearly show that the harmonic performance of the 3-level converter is superior to that of a 2-level converter. 12.4.1.3 Other Converter Arrangements The main incentive for using a converter topology different from the 2-level converter is the achievement of lower power loss and reduction of harmonics. Harmonics can also be reduced by switching the converter valves more than once per power frequency cycle, and this will be discussed below in Section 12.4.2. As mentioned above, the 2-level and the 3-level NPC converter topology are just two of several different topologies that could be used for VSC Transmission. Other converter topologies proposed for VSC Transmission have included floating capacitor solutions, which may be suitable for implementation of 4 or 5-level converters [12]. Another arrangement involves the use of an interface transformer with star and delta connected valve winding voltages, to which the VSCs can be connected, and which would provide harmonic cancellation [13]. However, at the end of 2005, none of these alternative topologies had been implemented on commercial VSC Transmission schemes. As the complexity of the converter increases, the capital cost also increases. Additionally, the space occupied by the VSC Transmission station will increase. However, the power loss for the scheme will decrease, assuming that the basic harmonic performance of the overall scheme is maintained.

12-23

12.4.2 High-Frequency Pulse-Width-Modulation As mentioned above, full-wave switching limits the number of degrees of freedom for a VSC converter. Since the IGBT is capable of repetitive high speed switching, modern motor drives now use VSCs with IGBTs switched at a frequency of several kHz. The advantage of such fast switching of the IGBTs is that the harmonics produced by the VSC are pushed towards higher frequency, which are easier to limit by means of shunt harmonic filters. However, the high voltage IGBTs used in VSC Transmission have higher switching loss than the lower voltage devices used in motor drives. Therefore, the power capability of the power device decreases with increasing switching frequency. Furthermore, the power loss is an important element in the economic assessment of a VSC Transmission scheme, so a careful optimization process has to be carried out to determined the optimum switching frequency for a given application. Typically, the switching frequency used in a VSC Transmission scheme will be less than 2000Hz. Many different control strategies can be used for converters switched more frequently than once per cycle[14][15]. Three different strategies will be described in this section: Carrier-modulated method with pure sinusoidal ac voltage reference Carrier-modulated method with 3rd harmonic modulated ac voltage reference Selective harmonic elimination method (SHEM).

c (a)

+Ud /2 V am (b)

-Ud/2

Figure 12-12 PWM controlled VSC with pure sinewave ac voltage reference and triangular carrier

Figure 12-12 shows the derivation of the switching instants for a control methodology using a triangular carrier wave (Vcarrier) and a pure sine wave ac voltage reference (Vcontrol). ). The figure shows the derivation of the control signals (a), and the resulting waveshape (b). 12-24

The intersection between the carrier triangles and the sine wave determines the switching time for the VSC valves. When Vcontrol is greater than Vcarrier the converter output is positive, and when Vcontrol is smaller than Vcarrier the converter output is negative. The ratio between Vcontrol and Vcarrier is known as the modulation index m.

m=

Vcontrol Vcarrier

The value of the modulation index can be any value between 0 and 1. The fundamental frequency component of the VSC ac voltage varies linearly with the modulation index, with a maximum voltage equal to the dc voltage. In the example shown in Figure 12-12, the frequency of the carrier wave is 9 times the power frequency. If the carrier frequency is an odd integer multiple of the fundamental frequency the ac waveform does not contain any even harmonics. In a three-phase VSC all of the triplen harmonics, i.e. 3rd, 9th, etc are eliminated in the phase to phase voltages. Both of these statements assume that the ac network is balanced and free of background harmonic distortion.

(a)

(b)

(c)

Figure 12-13 PWM switched 2-level VSC with carrier frequency at 21st harmonic.

Figure 12-13 shows a typical voltage harmonic amplitude at the ac terminals of the VSC, when using PWM switching with a carrier frequency of 21 times the fundamental frequency. The 12-25

spectrum changes with operating conditions. The figure shows the stepped ac waveshape, and in dotted line the fundamental frequency component of this waveshape (a), the harmonic voltage for phase to neutral (b), as well as the harmonic voltage phase to phase (c).

(a)

(b)

(c)

Figure 12-14 PWM switched VSC using sinusoidal PWM and 3rd harmonic injection.

Figure 12-14 shows the voltage harmonic spectra using a sinusoidal control voltage with 3rd harmonic injection and triangular carrier. The 3rd harmonic injection enables an increase of the fundamental frequency component of the VSC voltage, for the same dc voltage, giving greater utilization of the converter equipment. The 3 sub-figures are as described for Figure 12-13. As an alternative to the use of the simple triangular carrier wave for the determination of the VSC Valve switching instants, it is possible to use pre-calculated switching instants, which are determined such that specific harmonics are eliminated. This control method is called the Selective Harmonic Elimination Method, SHEM for short, or Optimised Pulse Width Modulation, OPWM for short. Since the switching instant required in order to eliminate specific harmonics depends on the operating conditions, it is either necessary to pre-determine the instants for the full range of normal operating conditions, or they must be determined on-line by the algorithm.

12-26

Half-wave and quarter-wave symmetries have to be preserved in the waveform, to ensure that even harmonics are eliminated. If there are say 4 switching instants in a quarter cycle, then one of these is normally required for the control of the fundamental frequency component of the ac waveform. The other 3 can be used for the elimination of 3 harmonics, say the 5th, 7th and 11th harmonic. Figure 12-15 shows the PWM waveshape and the voltage harmonic spectra for a VSC using SHEM or OPWM. The 3 sub-figures are as described for Figure 12-13.

(a)

(b)

(c)

Figure 12-15 PWM waveforms and harmonics for VSC operated with SHEM modulation

It can be seen that even with the same number of VSC Valve switch operations in a power frequency cycle, the harmonic content in the ac waveshape is considerably reduced, when using the SHEM control method. This means that either smaller ac harmonic filters can be used, reducing the site area required and capital cost, or fewer VSC Valve switch operations can be used, thereby reducing the power loss. The switching strategy can also be optimized to reduce the number of operations at high current, thereby further reducing the power loss. 12.4.3 High Voltage Turn-Off Valves The VSC Valves must be capable of turning on and off in response to control signals, irrespective of the current flowing though the valve, or of the voltage across the valve, at the 12-27

time of the valve operation. This characteristic makes VSC valves fundamentally different from the thyristor valves used in a LCC HVDC scheme. Several different semiconductor devices capable of actively turning off the flow of current are now available [4]. For a VSC Transmission application many devices must be connected in series in order to achieve the required voltage withstand, and this imposes additional requirements on the semiconductor. At present (2005) the Insulated Gate Bipolar Transistor (IGBT) is the device of choice for the following reasons:

It has transistor action, which enables precise control of the device in a manner that is not possible with latching alternatives. For instance:

The converter can be turned off even in short circuit conditions. It is possible to perform active voltage control, which is useful for the sharing of the voltage between a large number of series connected devices.

Control of the device is achieved using low power, since it is a MOS-controlled device. This is advantageous when operating at very high voltage levels, where auxiliary power has to be obtained either from the power circuit itself or from ground level across the insulation barrier. It is capable of high switching speed, thus making high switching frequency feasible.

New semi-conductor devices are being developed all the time, and future schemes may use different devices. At the end of 2005 the maximum voltage withstand capability of an IGBT was about 6.5kV, but such devices have considerably higher power loss than lower voltage devices. The high power loss reduces the current capability, particularly if the devices are switched at high frequency. Therefore, lower voltage devices are currently used for VSC Transmission. The IGBT is designed to have forward voltage blocking capability only, since a parallel diode provides current capability in the reverse direction, providing protection against reverse voltage. High blocking speed diodes, with low recovery charge, are used in order to minimize power losses. Silicon Carbide diodes have been tested in VSC Transmission schemes, but the benefits obtained because of their low recovery charge did not justify the additional costs. Both the IGBT and diode are usually implemented as chips, and are mounted in the same integral housing. In order to increase the reliability of the VSC Valves, the series strings of IGBTs typically include a number of redundant devices, such that operation can continue with satisfactory insulation margins, even when one or more IGBT devices have failed. Presspack mounting, as used for thyristors for LCC HVDC schemes, is also used for the IGBT in a VSC Transmission scheme in order to ensure that a failed device can continue to provide a safe and reliable current path until the next maintenance outage. However, because of the use of numerous chips, the construction is more complex than for a thyristor device, and presspack IGBTs are therefore considerably more expensive than the more usual IGBT modules. The voltage sharing across the series connected devices in a VSC Valve during turn off and in the off state can be controlled by using the IGBTs transistor action. Snubber circuits could also be used, but whilst these would reduce the power loss in the device, they would also create their own power loss, since the energy absorbed during turn off has to be released when the IGBT turns on again. 12-28

To reduce the VSC Valve stress and the power loss, the inductance in the loop formed by the VSC Valve and the dc capacitor must be kept as small as possible. This consideration imposes strict requirements on the arrangement of the VSC Valve, the dc capacitor and the overall circuit. The current capability of the IGBT must be determined taking into account not only the steady state operating conditions, but also the transient conditions encountered during various faults and abnormal operating conditions. Significant power loss is caused in the device during turn-off, since high voltage and high current are present simultaneously, as shown in Figure 12-16. Effectively, the current in the device is driven to zero by the voltage developed across the IGBT.
Voltage & Current Voltage Current

Power Loss Power Loss = U(t)*I(t)

Time
Figure 12-16 Voltage and Current across an IGBT during switching.

Time

During faults in the ac network or close to the converter, the IGBTs must protect themselves by interrupting the fault current before too much heat has been developed in the device. In order to achieve this, protective action must be taken before the current has risen too high. As the current through the device increases, the IGBT de-saturates, thereby increasing the resistance to the fault current. Gate control further enhances this natural response to stop the current quickly and completely. This requirement imposes constraints on the design of the VSC control and protection system. The diodes in parallel with the IGBT form an uncontrolled rectifier in the event of a short ciruit on the dc side of the VSC, and IGBT action cannot stop the current. Therefore, the diodes must be rated to cope with the short circuit current, until the ac side of the VSC has been disconnected from the ac network by circuit breaker action. The power loss caused in the devices during the operation of the VSC must be removed using a cooling system, to ensure that the device temperature is kept in the acceptable design range. The design of the cooling system is very similar to that used in a LCC HVDC scheme. Typically, the valves will be water-cooled and the on-valve pipe work must be designed taking into account the dc stress within the valve structure and between metallic components. To achieve high reliability, similar to that of the VSC Valves, redundant cooling pumps, fans, instrumentation and controls are typically used. The mechanical implementation of the VSC valve must take into account the requirements for low circuit inductance, electrical clearances, space for cooling, access for maintenance etc.

12-29

Figure 12-17 shows VSC valves inside a valve enclosure. A single valve for a 150kVdc, 2level, VSC Transmission scheme, can contain more than 300 series connected IGBTs. The enclosure is made of steel and aluminium, and acts to contain the electromagnetic interference caused by the switching of the VCS Valves. For some applications the enclosures can be implemented as transportable containers, such that a significant part of the checking and testing can be done in the factory, before the valves are sent to site.

Figure 12-17 VSC valves inside Valve Enclosure.

12.4.4 Harmonic Filtering The ac voltage waveshape produced by the VSC is far from sinusoidal, as can be seen in the Figures in section 12.4.2. Therefore, filtering is required to make the ac voltage acceptable to the ac network [16][17][18]. The phase reactor and ac harmonic filters perform the required filtering. The filtering required depends on:

The amplitude and frequency of the harmonics produced by the converter. The characteristics of the ac network. The permissible distortion limits.

It is also necessary to consider the harmonic performance on the dc side, even if the connection between VSC Transmission terminals is by means of cables. This is because harmonic resonance could cause magnification of the harmonics, resulting in localized overloading and premature aging. Furthermore, if the dc cable has limited screening, interference could also be

12-30

injected into adjacent communication circuits. Finally, if an overhead dc line were used between the terminals, interference could be injected in nearby telecommunication circuits. The determination of the harmonics in the ac and dc output is beyond the scope of this chapter. The harmonics depends on many factors, including:

The operating conditions. The converter control strategy. The converter topology. The impedance of the converter components.

The harmonics can be determined by means of Fourier analysis of the wave-shapes, or by mathematical analysis. It should be noted that harmonics are also influenced by the characteristics of the ac and dc circuit to which they are connected, in particular the balance of these circuits, and the presence of any pre-existing harmonics On the ac side the phase reactor plays a significant role in ac harmonic filtering by acting as a low pass filter. The phase reactor and the interface transformer (if used) will change the appearance of the VSC from a voltage source towards that of a current source. One or more ac harmonic filters are connected on the ac system side of the phase reactor, as shown in Figure 12-18.

Interface Transformer l U L

Phase reactor U conv lconv

ld = Ud DC Reactor

L AC Filter

Converter

Figure 12-18 Simplified diagram of one terminal of a VSC Transmission scheme.

The ac harmonic filter at each terminal will typically have a total rating of 10-20% of the converter rating. The design will typically consist of 2 or 3 filter branches, one or more of which being in ungrounded star or delta configuration. Typically the filter branches will include damped high pass filters, to provide damping and attenuation at high frequency. Tuned filters may also be used, e.g. at the PWM carrier frequency. The ac harmonic filters tend to be easier to design than equivalent filters for a LCC HVDC scheme, since the harmonics to be filtered are at higher frequency. However, as for a LCC HVDC scheme, care need to taken to avoid resonance between the filters and the ac network at frequencies where significant pre-existing harmonics may exist in the ac network.

12-31

For a dc cable scheme the dc capacitor and dc smoothing reactor normally provides adequate harmonic filtering. However, as mentioned above, if sensitive telecommunication circuits are located close to the cable route, then it may be necessary to provide additional filtering. This can be provided by additional series reactance, and/or by a shunt dc filter. 12.4.5 Phase Reactor The phase reactor performs the following important role:

It provides constant fundamental frequency impedance for the control of the VSC active and reactive power output. It provides a high frequency blocking filter between the VSC and the ac network. It limits short circuit currents.

Typically, the phase reactor has a short circuit impedance of about 15%, and it is implemented as an air-cored and air-insulated design. The reactor can be very large, and is typically enclosed in a metallic shield, to prevent the escape of high frequency electric and magnetic fields. Forced air circulation is used to cool the reactor. Because of the duty as a high frequency blocking filter the phase reactor is continuously exposed to the voltage steps in the ac output voltage. The reactor design must take this stress into account, to ensure that the phase reactor will perform satisfactorily throughout the life of the VSC Transmission scheme. The phase reactor must also have very low stray capacitance between its terminals, since high frequency current might otherwise bypass the reactor. The need for low stray capacitance makes the air-cored and air-insulated design particularly attractive. In principle, the stepdown transformer could perform the phase reactors role. However, the transformer would then be exposed to repetitive high frequency transients. Additionally, the stray capacitance in an oil immersed transformer is likely to be significantly higher than for an air-cooled, air-insulated reactor, so it would not perform as effectively as a high frequency blocking filter. Therefore, the use of a transformer only solution is likely to be limited to low power and low voltage applications, where a dry type transformer can be considered.

12-32

Figure 12-19 Phase Reactor for a 65MVA, 80kVdc VSC Transmission scheme

12.4.6 DC Capacitor The dc capacitor provides the voltage source for the operation of the VSC. The dc capacitor also plays an important part in the harmonic filtering on the dc side. The dc capacitor is connected directly across the converter dc terminals, and recent VSC Transmission schemes have used dry type dc capacitors, to eliminate the risk of fire. High frequency current flows in the dc capacitor during converter operation. It is important to keep the inductance of the loop formed by the VSC valves and the dc capacitor as small as possible, in order to minimize the VSC Valve stress and the power loss resulting from the stored energy in the current when the valves are switched off. Additionally, the design of the dc capacitor itself must have as low stray inductance as possible. The high frequency current causes electromagnetic noise, which must be contained as it cannot be eliminated at source (the high frequency current is necessary for the operation of the converter). Typically, the dc capacitor is mounted inside the shielded valve enclosure as close as possible to the VSC valves.

12-33

The flow of current causes a change of the voltage across the dc capacitor, and this change is known as the voltage ripple. The amplitude of the voltage ripple depends on:

The capacitance of the dc capacitor - the larger the capacitance the smaller the ripple. The VSC Valve switching strategy - long conduction periods at high current cause larger ripple.

The dc capacitor ripple is also affected by unbalance and pre-existing harmonics in the ac network. The dc capacitor ripple can have an impact on the generation of harmonics on the ac side, and therefore the ripple is typically limited to less than 5% of the rated direct voltage. The dc capacitor must also be rated such that the dc voltage is kept reasonably constant during fast dynamic changes to the VSC Transmission scheme operating conditions. The voltage change during transient conditions, e.g. a short circuit to ground in the ac network, must also be taken into account in the control and protection strategy for the converter. 12.4.6 Interface transformer A transformer may be used to couple the converter ac filter bus and the ac network. The use of a transformer fulfills the following tasks:

It enables the VSC to be designed independently of the ac network voltage constraints. If the transformer has an online tapchanger, it is possible to adjust the voltage on the converter side of the transformer independently of the ac network voltage, to achieve optimum converter performance and reduction of the steady state power loss. It blocks the flow of zero sequence current between the ac network and the converter. It provides additional series impedance on the ac side, which may be beneficial for the harmonic performance of the converter, and may enable a reduction in the rating of the phase reactor.

Since the transformer is not exposed to direct voltage, and the harmonic filtering is performed on the converter side of the transformer, a conventional station transformer can be used. A tertiary winding can be provided for the connection of the auxiliary supply system. In some applications a transformer may not be necessary, e.g. if the ac network voltage level is such that no voltage transformation is necessary. 12.4.7 Other VSC Transmission sub-station equipment A typical VSC Transmission substation includes a number of items in addition to the main equipment described above. These include:

12-34

AC circuit breakers. Typically a circuit breaker is only used at the connection to the ac network, since the ac harmonic filter does not need to be switched during the operation or starting of the station. The ac circuit breaker may include a pre-insertion resistor, in order to limit the dc capacitor voltage overshoot during first energisation. In some projects the energy absorption requirements for the inrush limiting resistor is so large that a separate resistor with a parallel circuit breaker has been connected in series with the normal breaker to fulfill this role. RFI and PLC filters. A small high frequency blocking reactor, and a shunt capacitor is used to limit the injection of electromagnetic disturbances into the ac network to an acceptable level. Voltage and Current measuring transducers are required at a number of locations in order to provide information for the control and protection systems. The transducers may be duplicated, unless backup control and protection use different inputs. Surge arresters are used for over-voltage protection. Arresters are used on either side of the converter. Disconnectors and earth switches are used at the ac and dc terminals to enable the equipment to be isolated and made safe for maintenance work. Auxiliary power supplies. Typically, these will be taken from two independent sources, one of which may be a tertiary winding on the interphase transformer. The auxiliary supply system will typically provide duplication of the supply to all critical equipment, with automatic changeover in the event of a failure of the primary supply. Fire Protection. The requirements are similar to those for a LCC HVDC scheme, but the equipment can be smaller, since the VSC Transmission station and the equipment within it is smaller and more compact. Civil works. The design of the civil works must take into account the requirements for housing and environmental control for the power equipment, protection and control, as well as of any operator and maintenance facilities. Additionally, the civil works must be designed to limit the audible and electromagnetic noise from the operating equipment.

Figure 12-20 shows the layout of the equipment for one terminal of the Gotland VSC Transmission scheme, which has a rating of 50MW and operates at 80kVdc.

12-35

Phase A, B and C valve compartments Phase Reactors

DC Yard Equipment

Auxiliary Power System & Cooling Control

Cooling towers Building 45 x 18 m

Figure 12-20 Layout of 50MW, 80kVdc VSC Transmission sub-station

AC Yard & Harmonic filters

12.4.9 DC Cables The transmission of energy between the terminals in a VSC Transmission scheme is achieved by dc cables. In principle, when the transmission is over land it could also be achieved using an overhead dc line. However, as discussed above, a fault on the dc side impose additional stress on the diodes and converter equipment, and requires tripping of the ac circuit breakers at all terminals, to clear the fault. Therefore, at present (end 2005) all commercial VSC Transmission schemes use dc cables. The use of dc cables has a number of advantages compared to the use of dc overhead lines:

Cables often have less environmental impact than an overhead line. The right of way required for a cable installation is considerably less than that for an overhead line, and the land above the installed cable can often be returned to its previous use. Cables can be run next to roads and through existing tunnels. Cables are much less prone to faults than overhead lines. In particular, lightning strikes or pollution do not affect a cable.

The voltage polarity on a dc cable in a VSC Transmission scheme is fixed and independent of the power direction. This enables the use of XLPE type polymeric cables for VSC Transmission. Polymeric type cables cannot be used for LCC HVDC schemes, because the high resistivity and low mobility of trapped charges in the material would cause unacceptable and potentially damaging voltage distribution within the dc cable immediately after a voltage reversal[19]. 12-36

Polymeric cables can be lighter and have smaller bending radii than the conventional oilimpregnated or oil-filled cables, which are used with LCC HVDC schemes. Therefore, polymeric land cables are easier to handle during installation, and special direct ploughing and laying techniques can be used where the ground conditions are favourable, making installation less time consuming. An additional advantage of the use of polymeric cables is that the environmental risks are lower than with an oil-impregnated or oil-filled cable. Transport logistics limits the length of each section of land cables to 1 2 km. This means that land cables have many field joints. The joint of a cable tends to be less reliable that the cable itself, and it is important to ensure that adequate type and life testing has been performed on the joints before the cable is installed. Furthermore, the field joints must be installed under clean and controlled conditions. Figure 12-21 shows a mobile jointing container in which the work can be performed. Generally, the performance record of cables for VSC Transmission schemes has been very good, one exception being the Direct Link scheme in Australia, where many joint failures occurred in the early years after the scheme entered service, due to a design defect.

Figure 12-21 Jointing Container used for the Murray Link scheme in Australia

Submarine cables can also be provided as polymeric cables. Submarine cables require armoring, to enable the cable to withstand the mechanical stresses imposed during the laying operation. Submarine cables are normally installed from specially equipped ships with the cable coiled on a large turntable. Continuous cable lengths of up to 100km can be laid, minimizing the number of cable joints required.

12-37

12.5 Basic Operational Limits

A simplified steady state operating range of one of the terminals in a VSC Transmission scheme, expressed as the VA capability at its ac connection point, is illustrated in Figure 12-22. The figure has been drawn for 3 different ac network voltages, to illustrate the dependence of the capability on the ac voltage. The use of a tapchanger on the interface transformer can remove this dependence for steady state operation, thereby enabling the capability of the converter to be utilized more fully.
Pconv Desired Reactive Power Desired Active Power Qconv Inductive Capacitive Uac = Max Uac = Nom Inverter Mode
Figure 12-22 Simplified PQ characteristic of a VSC Transmission terminal

Rectifier Mode DC cable Thermal limit

Uac = Min

The maximum current capability of the VSC Valve dictates the MVA capability at a given ac voltage. Assuming that the IGBT and the diode have the same current capability, the MVA capability at a given ac voltage can be assumed to be described by a circle. The current capability of the IGBT and the diode do not necessarily have to be identical. The thermal duty on the IGBT consists of the switching loss and the conduction loss. The switching loss exceeds the conduction loss when the IGBT is operated at a relatively high switching frequency. Therefore the difference in the thermal duty is not dramatically different, whether the converter operates as a rectifier or as an inverter. For the diode the switching loss tends to be much smaller than the conduction loss. Since the conduction duty on the diode is greater when the converter acts as a rectifier than when it operates as an inverter, it would in principle be possible to design the scheme such that it has a higher transfer capacity in one direction than in the other. However, the actual saving has to be considered in the context of the overall solution, and must take into account the benefits resulting from the use of standard building blocks. Therefore, the converters at the two ends of a VSC Transmission scheme are normally identical, and provide the same power flow capability in both directions. In the example shown in Figure 12-22 the active power capability of the VSC Transmission scheme exceeds the desired capability at all ac voltages within the range of Umin to Umax. 12-38

In addition to the limit on operation caused by the thermal duty on the semi-conductors, another number of issues must also be considered:

The voltage on the dc capacitors and dc cables must be kept low enough to provide safe operation for both the VSC valve and the dc capacitor and cable themselves. The maximum ac voltage produced by the converter depends on the direct voltage on the dc capacitor. This then results in a limitation on the capacitive reactive power, which the converter can produce. Since the generation of reactive power requires the converter voltage amplitude to be higher than the ac network voltage, the capacitive power capability falls with increasing ac voltage. In the example shown in Figure 12-22 the capacitive power capability is lower than the desired output at the maximum ac network voltage. In practice this is unlikely to be of any consequence, since generation of reactive power is unlikely to be required when the ac network voltage is already higher than the nominal value. When specifying the required reactive power capability, it is necessary to state explicitly the ac voltage and active power exchange at which the reactive power capability is required. The dc cable may impose a restriction on the maximum dc current allowed to flow between the VSC Transmission terminals. This limitation will typically be just above the desired power transfer capability, and the VSC control system will ensure that this limit is not exceeded. For clarity the dc cable limit in Figure 12-22 is shown significantly higher than the desired power transfer capability. In addition to the limitation on the capacitive power capability at high ac system voltage, there is also a minimum dc voltage limit for the steady state operation of the VSC Transmission converter. This limits the reactive power that the converter can absorb.

12.6 Converter Rating and Protection Considerations

The rating of the VSC Transmission substation must take into account a number of dynamic and transient conditions in addition to the steady state design considerations as outlined in section 12.5 above. The voltage withstand requirement for components and insulation must be determined based on the over-voltages that may be experienced due to any credible event. Similarly, the equipment must be capable of withstanding the transient and dynamic current surges resulting from any credible event outside and within the VSC Transmission station. 12.6.1 Over-voltage Stresses and Protection When the first VSC Transmission substation is energized by the closing of its ac circuit breaker the converter diodes acts as an uncontrolled rectifier charging the local and remote dc capacitors and the dc cable. Under some conditions, the combined impedance of the ac network and interface transformer and phase reactor will react with the total dc system capacitance such that an over-voltage is created. The over-voltage can be reduced by control action as soon as the IGBTs can be switched. If the initial over-voltage is unacceptable, then it may be necessary to use a pre-insertion resistor on the ac circuit breaker, or to use other techniques to limit the overvoltage. When the interface transformer has a tapchanger, it is normally placed in the position giving the lowest converter side ac voltage, prior to energisation of the transformer.

12-39

In some multi-level converter circuits, uncontrolled rectifier charging of the dc circuit will result in unequal voltage distribution on the dc capacitors. The rate of charging can be slowed down by the use of a pre-insertion resistor, such that the IGBTs have time to control the dc capacitor voltages. During a sudden load rejection at the inverter a significant over-voltage may be experienced on the dc side. The over-voltage will be reduced very quickly by the converter control, as can be seen in Figure 12-23, which has been taken from an ABB site measurement and digital simulation. The figure shows a load rejection of 330MW at 0.47seconds, which results in an increase of the dc voltage from about 147kVdc to a peak value of 205kV within 10ms. The voltage is reduced to less than 160kV within a further 20ms.

Figure 12-23 DC over-voltage on rejection of 330MW

Just like any ac substation, a VSC Transmission substation will be subjected to transient voltages caused by lightning strikes on ac overhead lines, or even within the substation itself. Overvoltages caused by switching events, e.g. the clearance of line faults or equipment failures in the ac network, and within the VSC Transmission substation itself, may also be experienced. In the VSC Transmission substation surge arresters are typically connected close to the interface transformer, to limit the over-voltage amplitude and to protect the transformer and other power equipment. Surge arresters are typically also provided on the converter side of the interface transformer in order limit any transferred over-voltages. There is no significant difference between the application of surge arresters to protect against lightning, and switching surge overvoltages and the insulation co-ordination process for a conventional ac substation and a VSC Transmission substation in this respect. The interface transformer will act as a barrier to the transfer of over-voltages from the HV bus to the converter ac harmonic filter bus. However, care must be taken to determine the transfer of over-voltages through the transformer, taking into account the stray capacitance across and between windings, and the impedance of the components on the converter side of the interface transformer.

12-40

It should be noted that if pre-insertion resistors were used to limit the over-voltage on first energisation, these pre-insertion resistors would not be in the circuit when the ac voltage suddenly returns after the clearing of a local three-phase fault to ground. Typically the control and protection system would prevent the dc circuit from fully discharging in the event of a major fault in the ac network. Therefore, energisation of a discharged dc circuit without pre-insertion resistors can be considered to be a double contingency failure. Whilst the ac harmonic filters are smaller than those used on a LCC HVDC scheme, it is still important to check that there is no resonance between the filter capacitance and the combined impedance of the interface transformer and ac network. If a resonance is possible, then it may cause magnification of the switching surge type over-voltages appearing during clearance of faults in the ac network. The surge arresters may limit the over-voltages to acceptable levels, but in case of a resonance the energy absorbed may be large, and need to be taken into account in the rating of the arrester. In a 2-level converter the VSC valves are rated to withstand the full voltage between the dc terminals, therefore major over-voltages across components will not be caused by faults within the converter. In multi-level converters internal faults in the converter or a control malfunction could potentially cause the build up of significant unbalance of the dc capacitor voltages. Surge arresters will prevent these over-voltages from causing damage to other converter equipment. However, if the condition were allowed to continue the surge arresters would be overloaded and suffer a catastrophic thermal run-away. Therefore, voltage monitoring and protection is necessary to detect the onset of such a situation. A breakdown to ground internally in a multi-level converter can result in an instant and large over-voltage across VSC valves and other components. Surge arresters will limit the overvoltage and prevent equipment damage but, to prevent the over-voltage from continuing and overload the surge arrester, it is necessary to trip the converters at both ends of the VSC Transmission scheme. When available, communication between the terminals can be used to speed up the protective action. However, the protection should be designed such that equipment is safe even without inter-station communication, and such that it can detect that a fault has occurred at the remote terminal. 12.6.2 Over-current stresses and protection During start up of the VSC Transmission substation the diodes in the first terminal to be energized will act as an uncontrolled rectifier and charge the local and remote dc capacitors and the dc cable. The diodes must be rated for the current duty associated with this event. The interface transformer and phase reactor will block lightning strikes in the ac network, and the lightning current would be diverted to ground by the surge arresters. The design of the VSC Transmission substation in respect of over-current stresses due to lightning strikes is therefore no different from the design of a conventional ac substation.

12-41

In the event of a short circuit to ground in the ac network close to the VSC Transmission terminal operating as an inverter, there would be a rapid increase in the converter ac current. The combined impedance of the phase reactor and the interface transformer will limit the rate of rise of the current. As the current increases, the IGBTs will de-saturate, increasing the voltage drop across the device, and the protection will take action to actively reduce the over-current, by turning off the IGBT. The design of the converter and its protection must be such that the IGBT stress during this event remains within the IGBTs safe operating area, and such that the converter can immediately resume operation, as soon as the fault in the ac network has been cleared. If the short circuit to ground appears within the VSC Transmission terminal, the rate of rise of the fault current will be higher than when the fault appears in the ac network, since the impedance in the fault current path will be lower. The protection system will still limit and stop the fault current, but the ac circuit breaker must also be tripped, to prevent further damage. The operation of the VSC includes a blanking period (see section 12.4.1.1) to prevent the IGBTs in the upper and the lower valve groups from conducting at the same time. If both IGBTs were to conduct, a so-called fire-through would be caused, creating a short circuit of the dc capacitor. Since the inductance in the VSC Valve and dc capacitor loop is kept very small by design, the rate of rise of the fault current would be extremely high, and the risk of IGBT failure would be substantial. The risk of a control failure causing such a short circuit to occur across the dc capacitor must be made very small by design. During a fault on the dc side of the VSC, the diodes at both VSC Transmission terminals will act as uncontrolled rectifiers, resulting in large current flowing into the fault. The current will continue to flow until the ac circuit breakers at the VSC Transmission terminals have been opened. The fault will be detected by the converter protection, and immediate tripping of the circuit breaker is initiated, in order to limit the thermal stress on the diodes, and to remove the voltage stress. Whilst faults are rare on a cable transmission system, a short circuit on the dc side must be taken into account when rating the diodes.

12.7 Practical Operating Characteristics

12.7.3 Terminal Voltage and Current relationships When the VSC Transmission substation is connected to an ac network which has other sources of generation and loads, the relationship between the VSC Transmission ac voltage and current depends on:

the amplitude and phase angle of the converter voltage (or more precisely, the ac filter bus voltage) relative to that of the ac system EMF, and the impedance between the converter EMF and the ac system EMF.

This relationship was discussed in Section 12.3 for normal operating conditions, and it was assumed that the impedance between the two EMFs was largely inductive. In this case the active power flow into the ac network depends on the phase angle between the two EMFs, and the 12-42

reactive power flow depends on the relative amplitude of the two EMFs. The VSC Transmission terminal therefore operates electrically similarly to a generator, and the active and the reactive power can be controlled independently of each other. In this operating mode the frequency used for the control of the VSC is determined from an electrical measurement of the ac network voltage. If the VSC Transmission terminal is connected to an ac network which does not have its own generation, then the absolute phase angle of the converter voltage is of no significance. The frequency used for the control of the VSC is determined purely by the control system, in accordance with settings from the network control center. The current flowing into the ac network is determined purely by the amplitude of the converter voltage, and the combined impedance as seen from the VSC Transmission ac terminal plus the impedance of the terminal itself. As the load in the ac network increases the converter EMF must be increased to maintain the ac voltage at the connection point. The load current on the ac network transmission lines may change the reactive power flow in the ac network, and therefore the ac voltage profile in the ac network. The VSC Transmission scheme is able to adjust/control the voltage amplitude at its connection point, but in doing so it will also change the active power delivered to the loads. Thus, when feeding a passive load there is not true independence between the active and the reactive power at the VSC Transmission ac terminals. The maximum amplitude of ac current that the VSC Transmission scheme can inject into a short circuit ac network will be close to the maximum current capability of the scheme. Therefore, special consideration needs to be given to the ac network protection, when a VSC Transmission scheme is the sole supply of power (see section 12.8.3). When the ac network is extremely weak at the point of connection of a VSC Transmission terminal it becomes more difficult to control the VSC. The frequency used for the control of the VSC has to be derived from the ac network voltage, in order to prevent unacceptable active power flows in the ac network. It is also desirable to control the ac voltage at the connection point to a constant value, the constant value being changed in accordance with a pre-set strategy e.g. a slope setting, as the active load changes. In order to avoid power swings in the ac network, it may be desirable to control the amplitude of the active current flowing into the network to a constant value. 12.7.2 Active Power vs. Reactive Power Generation/Absorption Capability The PQ capability of a VSC as described in section 12.5 did not take into account the presence of the interface transformer impedance or the ac harmonic filter impedance. These impedances modify the capability of the VSC, shifting the characteristics towards more reactive power absorption or generation capability, depending on the relative impedance of the two elements. When required the capability of the VSC Transmission terminal can be deliberately shifted further towards absorption or generation, by the connection of a shunt reactor or a shunt capacitor, either on the VSC side or the network side of the interface transformer. When the reactive power elements are connected on the VSC side the rating of the interface transformer may have to be increased. By connecting any additional reactive power elements by means of circuit breakers, the capability could be shifted as necessary to suit the particular transmission 12-43

and loading conditions, thereby optimizing the converter operating conditions and reducing the power loss. 12.7.3 Operating Losses vs. Transmitted Power Significant reductions of the power loss, relative to the first generation of VSC Transmission schemes, have been achieved by the development of the technology. The first generation of VSC Transmission used 2-level converters with triangular carrier wave PWM control. Subsequently 3-level converters were used, and the latest large scheme (2005) will use 2-level converters with SHEM or OPWM control. Additionally, improvements have also been made in the design of VSC Valves and other equipment. As a result of these developments the power loss at rated power has been reduced from approximately 3% to approximately 1.9% per terminal. The power loss is still considerably larger than the power loss for a LCC HVDC converter station, which would typically be about 0.8% at rated power. Whilst future development and optimization is likely to result in a further fall in the power loss, it is unlikely that it will get as low as for a LCC HVDC scheme, since the conduction loss of the silicon based IGBT is significantly larger than that for a thyristor. Furthermore, the VSC Transmission scheme will have higher power loss because of the IGBT switching loss and the higher frequency switching of the VSC. A formal methodology has not yet been established for the determination of the power loss of a VSC Transmission scheme. The major contributions to the power loss of a VSC Transmission scheme are as follows:

The VSC Valves The interface transformer The phase reactor The ac harmonic filter Auxiliary/service loads The dc smoothing reactor

Simplistically, the power loss in the VSC Valve is caused partly by the conduction loss and partly by the switching loss. The switching loss is caused primarily by the simultaneous presence of high voltage and high current during the turn-off process. Figure 12-16 in section 12.4.3 illustrated the turn-off process for an IGBT, and the power loss caused during this process. Whilst the IGBT power loss due to conduction varies almost linearly with the converter current, the switching power loss has a substantial value, even at virtually zero converter current. The conduction and the switching power loss in the IGBTs and the diodes have different values. Therefore, the VSC Valve power loss depends on the direction of power flow, the terminal acting as rectifier tending to have slightly lower power loss than the inverter terminal.

12-44

The VSC Valve power loss also varies with the reactive power absorbed or generated by the converter. The total power loss for any given conditions must be determined based on the detailed current flowing through the different valve components during a power frequency cycle. The power loss in the interface transformer is determined as for a conventional transformer, since the harmonic current flowing through the transformer is negligible, compared with the fundamental frequency component. The transformer power loss includes a no-load loss and a load loss. The power loss in the phase reactor is largely due to the flow of fundamental frequency current. As for the interface transformer, the harmonic current flowing through the reactor is negligible, compared with the fundamental frequency component. The power loss in the ac harmonic filter is dominated by the harmonic contributions. The auxiliary power required for the operation of the VSC Transmission substation tends to have a large fixed component (no-load loss), and a smaller component that varies with the operating condition of the converter, e.g. cooling plant losses etc. Figure 12-24 shows the approximate variation of the total power loss for the two terminals of a VSC Transmission scheme. The power loss shown does not include the power loss in the dc cable, which will naturally depend on the length and design of the cable.
Power Loss (% of rated power)

4 3.5 3 2.5 2 1.5 1 0.5 0 0 20 40 60 80 100

VSC Converter Load (%)

Figure 12-24 Variation of power loss with load for the two substations in a VSC Transmission scheme.

In the absence of a standard way of determining the power loss for a VSC Transmission scheme, consideration may be given to the measurement of the power loss. Such an approach would not be easy, and would require very accurate measuring transducers at the ac terminals of both VSC Transmission stations, assuming that the total power loss including the dc cable is to be determined. The measurements would have to be performed at all operating conditions at which performance has been guaranteed. Unless the auxiliary power for the operation of the scheme has been derived from tertiary windings on the interface transformers, it would be necessary also 12-45

to measure the auxiliary power consumed at each station. It may also be necessary to correct the power loss for ambient temperature.

12.8 Control Principles and Structures

The control and protection system is an extremely important element of a VSC Transmission scheme. An ideal control and protection system enables the full use of the capability of the equipment in accordance with operator commands and pre-set strategies for transient and dynamic conditions, and it will take action to prevent damage to the equipment in case of failures or unforeseen circumstances. Modern control and protection equipment is close to providing an ideal system. Many elements of the control and protection system for a VSC Transmission scheme are similar to those used for a LCC HVDC scheme, and their implementation use many identical components. Naturally, some control algorithms are different, e.g. the determination of the turnon and turn-off instants for the IGBTs and for the thyristors. This section will describe a typical control and protection structure used for a VSC Transmission scheme, and will describe how some of the control objectives are achieved. The description has been based on an ABB diagram, but other implementations are of course possible[20]. The section also provides a description of a typical protection system for a VSC Transmission scheme. It should be noted that the control and protection systems are not completely separate, as the control system plays a significant part in keeping component stresses within acceptable limits, even during dynamic and transient conditions. 12.8.1 Control Structure Figure 12-25 shows a simplified single line diagram of one terminal of a VSC Transmission scheme and a typical control structure. The control system includes the following parts:

I/O Interface PQU Order Control Current Order Control Current Control Valve Control

12-46

Figure 12-25 Typical control structure of a VSC Transmission scheme (courtesy of ABB).

The structure of a VSC Transmission system can be different, and the functions may be implemented differently than shown in Figure 12-25.. However, all of the functions outlined below must be implemented in the control system. The following descriptions relate to the implementation shown in Figure 12-25. 12.8.1.1 I/O Interface The I/O Interface takes measurements from the VSC Transmission substation and processes these for use in Current Control. The processing includes Analogue to Digital conversion, filtering and normalization. The data is passed directly to Current Control. The input data includes:

AC voltage and current at the point of common coupling, AC voltage at the filter bus, Positive and negative direct voltage on the dc capacitors, 12-47

Direct current at the positive and the negative dc terminals.

12.8.1.2 PQU Order Control The PQU Order Control receives setting information from the operator, internal and external system status information, and data from Current Control describing the operation of the system. PQU Order Control acts directly on the interface transformer tapchanger, to optimize the operating conditions for the VSC. PQU Order Control also provides reference values for the use of Current Order Control, the reference values having been derived to achieve the settings from the system operator, subject to operational limits, e.g. temperature and equipment limitations. PQU Order Control also takes into account any high-level automatic control algorithms, e.g. power system frequency control, damping control, runback control etc, when determining the reference values passed to Current Order Control. The setting references are the conditions to be achieved at the VSC terminals, and are dc or rms quantities. The data communicated to Current Order Control include:

Active Power reference Reactive Power reference Direct Voltage reference Block (digital signal to turn off IGBTs) Runback (digital signal) DC Voltage Control enable (digital signal)

The last 3 signals over-ride the setting references. 12.8.1.3 Current Order Control Current Order Control receives the reference values from PQU Order Control and data from Current Control, which gives the following data for the VSC:

Positive Sequence dq voltage at the ac terminals Positive and negative direct voltage at the dc terminals Direct current at the positive and negative dc terminals

The signals provided to Current Control are instantaneous values and include:

d axis ac current reference q axis ac current reference Block signal

These values are determined by the following control loops: 12-48

PQ2IDQ, which calculates the ac d and q current orders with respect to the provided positive sequence voltage. DCVC, which calculates the current order required to keep the direct voltage at the desired level.

Typically, one VSC Transmission terminal is assigned to steady state control of the active power flow between stations, and the other is assigned to steady state control of the direct voltage. At the station operating in active power control, the direct voltage reference depends on the active power exchange, but during transient or dynamic conditions dc voltage limits may come into force in order to avoid over or under-voltage on the dc side. If the ac voltage falls below the normal operating range for the ac network the control will change from power control to current control. Current Order Control also includes a number of pre-set characteristics with the objectives of keeping the operation of the VSC within safe limits during dynamic and transient conditions. The objectives are:

Keeping the dc voltage within safe limits (UDCCOL dc voltage dependent current order limit). Keeping the ac voltage at the filter bus within safe limits (UACCOL ac voltage dependent current order limit). Keeping the device current within the safe operating area.

12.8.1.4 Current Control Current Control processes the data from the I/O Interface and determines the control instants for switching VSC valves on and off, in order to satisfy the Id and Iq reference values sent by Current Order Control. The de-composition of the ac voltage measurement into positive and negative voltage sequence components must be done very quickly, in order to provide accurate and independent control of active and reactive power. In order to achieve de-coupling between the active and reactive power, a quasi-positive sequence voltage at the point of common coupling can be derived and used together with the true negative sequence voltage for the ac current control, effectively providing a feed forward loop. A Phase Locked Loop (PLL) is used to achieve synchronization between the converter voltage and the ac voltage at the point of common coupling. An additional Phase Locked Loop (DQPLL) operates on the positive sequence voltage derived from the ac filter bus voltage, and provides the sampling instants. The objective of the AC Current Control is to control the current through the converter reactors. Symmetrical three-phase currents are provided by the converter irrespective of whether the ac network voltage is symmetric or not. The current order is calculated from the power order. 12-49

The PWM function, which may be a SHEM PWM or OPWM strategy, is used in conjunction with the DQPLL, and determines the switching instants for the VSC Valves. 12.8.1.5 Firing Control Firing control checks that it is safe for the VSC Valve to turn on and off, before issuing the control signal to the valve level. The blanking time interval is implemented and checked in the Firing Control. Firing Control also checks that it is safe to turn on the valve, e.g. that the valve redundancy has not been exceeded by IGBT or diode failures. The Firing Control communicates with the valve level by means of a dedicated fiber optic cable. Firing Control also receives status information from the valve level, e.g. healthy/failed status and other information. 12.8.2 Control Strategies for Different Applications Figure 12-25 shows a number of reference signals as input to PQU Order Control. For some applications some of these signals may in fact be derived automatically based on other local information. The processing of the local information, and the derivation of the reference signals may be done in a separate control block, but can also be derived in the PQU Order Control block. The following sub-sections describes the higher level controls for a number of different applications. The descriptions are for the steady state and dynamic control of the VSC Transmission scheme. In all cases the control limits within Current Order Control will also be available, and will act as required during abnormal conditions in the ac network in order to:

Limit the stresses on the converter equipment to within acceptable limits. Ensure that normal operation can be resumed as quickly as possible when normal ac conditions are resumed.

12.8.2.1 Power Transmission between weak ac networks One station of the VSC Transmission scheme will normally be assigned to dc voltage control, whilst the other will be assigned to active power control. If the dc voltage at the inverter terminal is controlled and the transmission distance is large, it may be desirable to have a slope setting on the reference voltage. This would enable the mean dc voltage to be increased at part load, thereby decreasing the power loss in the cable. When the station in dc voltage control needs to increase the dc voltage, e.g. because it has dropped below the reference value, the station has to increase the active power injected if it is a rectifier, or decrease the power extracted if it is operated as an inverter. If one of the stations is connected to a relatively weaker ac network than the other, then it may be desirable for that station to be in power control. Similarly, if one ac network has a requirement for frequency and/or damping control, then it may be preferable for that station to be in power control. 12-50

If both ac networks are weak it is normally desirable for both stations to be in ac voltage control. The ac voltage setting may have a slope with active power, such that an appropriate ac voltage profile is established in the ac network. If one or both terminals are in a strong ac network, the system operators may prefer to operate that terminal or those terminals in reactive power control mode. Again the reactive power reference may have a slope with active power, to suit the particular ac network conditions. 12.8.2.2 Feeding of Passive Network When a VSC Transmission scheme feeds a passive load, the rectifier (i.e. the terminal in the main ac network) will typically be in dc voltage control and either ac voltage control or reactive power control, dependent on the system operators preference. The PLL will lock to the ac system voltage. The terminal in the passive network will be in frequency and ac voltage control mode. As the ac system does not directly create its own ac voltage EMF, the PLL operates from its own internal oscillator. If synchronous generator(s) may occasionally operate in the passive ac network, then the control system must be capable of switching between an external ac system voltage and its own internal oscillator. In this case, the VSC Transmission scheme will be in active power control mode. The active power will then typically be determined by a frequency control loop. If the generators supply only a small fraction of the load, it may be difficult to detect their presence, and signalling between the generators and the VSC Transmission scheme may be necessary. 12.8.2.3 Connection of a Wind Farm The connection of a large, remote wind farm is a special condition of the example described in 12.8.2.2 above. The VSC Transmission scheme may both have to operate as the single power supply to a passive network, e.g. during conditions or either no wind or extreme wind, and as an exporter of power, when wind conditions are suitable for generation. Typically, the mainland terminal will be in dc voltage control and either ac voltage or reactive power control, as desired by the system operator or appropriate to the ac system conditions. The wind farm terminal will typically be in frequency control (with internal oscillator) when the scheme is importing power, and in active power or frequency control (using system frequency or internal oscillator, depending on generator type) when exporting power. The control of the wind generators and the VSC Transmission scheme should be co-ordinated to achieve optimum operation of the overall remote system. 12.8.2.4 Multi-terminal schemes If the dc capacitors of three or more VSC Transmission terminals are connected in parallel by dc cables, then the scheme will operate as a multi-terminal scheme. Typically, one terminal would be assigned to dc voltage control, with the other terminals in active power control mode. If 12-51

required, the dc voltage can be controlled with a slope, to optimize the voltage profile on the dc network. The power input and output must be balanced in order to maintain a stable dc voltage. The terminal operating in dc voltage control will provide or absorb active power, as required to maintain the dc voltage at the desired level. When a rating limit is exceeded, the terminal will no longer be able to maintain dc voltage control. Therefore, the other terminals must also have a dc voltage control mode, which can over-ride the active power control, when the dc voltage deviates outside set limits. These separate dc voltage control modes must be co-ordinated, e.g. by the use of different dead-bands, such that a stable and robust control system is achieved.[21]. 12.8.3 Performance During and After System Faults Aspects of the performance of the VSC Transmission scheme during faults in the ac or dc system have already been discussed above in section 12.6. This section focuses on the transient and dynamic responses. During a fault within the converter or on the dc side, the IGBTs are immediately turned off and a trip signal is sent to the ac circuit breakers at all VSC Transmission terminals. To minimize the stress caused by the over-current, the circuit breakers normally have fast operation and clearance time of less than 3 power frequency cycles. Due to the severity of the fault, an inspection of the converter and the dc system is typically undertaken before the VSC Transmission system is energized again. During a fault in the ac system the IGBTs may be temporarily turned off, if the ac voltage is very low (typically below 0.2pu), to prevent excessive over-current. The IGBT is turned on again when the current has been controlled down to an acceptable level. The performance of a VSC Transmission scheme during a remote single phase fault in the ac grid is illustrated in Figure 12-26During the recovery from a fault in the ac network, the VSC Transmission scheme would in principle be able to increase the power back to the pre-fault level within 1 power frequency cycle. However, depending on the ac network characteristics, such quick resumption of active power may not be the optimum action, particularly not if the ac network is weak at the point of connection. In such a case, a better strategy may be to give preference to the control of the ac system voltage, with a slower return of the active power to the pre-fault level. The optimum post fault strategy is best determined by system studies. . The traces have been provided by ABB and are for the Cross Sound Cable project (see section 12.12.5 for further detail of this project). The upper trace shows the ac voltage at the point of common coupling (PCC). The fault appears just before time 0 and results in a significant reduction in ac voltage, which lasts until fault clearance at about 75ms. The lower trace shows the transmitted power, which can be seen to reduce during the fault (because of the current limit for the IGBTs), but recovers very quickly after the clearance of the fault. Full power transmission has been restored within 80ms of fault clearance.

12-52

Figure 12-26 Dynamic Response to a remote ac fault.

During the recovery from a fault in the ac network, the VSC Transmission scheme would in principle be able to increase the power back to the pre-fault level within 1 power frequency cycle. However, depending on the ac network characteristics, such quick resumption of active power may not be the optimum action, particularly not if the ac network is weak at the point of connection. In such a case, a better strategy may be to give preference to the control of the ac system voltage, with a slower return of the active power to the pre-fault level. The optimum post fault strategy is best determined by system studies. If the VSC Transmission scheme is the sole power supply to the ac network, or the ac link to other parts of the ac network is extremely weak, then the protection in the ac network needs special review. The fault current from the VSC Transmission scheme is typically only 20% of that of a comparable generator. Whilst this reduces the system stress during a fault, and therefore gives more time for fault clearance, some protection systems may not function correctly at these low currents. For example, conventional over-current relays may not be suitable for the protection of ac lines and major equipment, since the fault current would be much smaller than in an ac network powered by conventional generators. The VSC Transmission scheme will normally be arranged to feed as much current into the network as possible, to facilitate the operation of protection in the ac network. However, prolonged operation of the VSC Transmission scheme may not be possible, if the fault is very close such that the fault impedance as seen from the VSC Transmission substation is very low. 12.8.4 VSC Transmission protection The purpose of the VSC Transmission protection system is to detect equipment failures and maloperation of the converter control. Figure 12-27 shows a typical protection block diagram for a VSC substation. 12-53

Figure 12-27 Typical Protection diagram for VSC Transmission substation (Courtesy of ABB)

In the event of a fault the following actions may be taken:

Temporary blocking (turning off) of the converter. Single phase transient current limitation. Permanent blocking of the converter. Trip of ac circuit breakers.

The action taken depends on the severity and type of the fault. Note that for a VSC blocking action means that an off pulse is sent to the IGBT, thereby stopping the conduction of the IGBT. Since a control system failure could be responsible for protection actions, those protection indications that do not require immediate action may first initiate a changeover between the active control channel and the stand-by channel. If the problem disappears after the changeover, the previous control channel is locked out, an alarm is raised and operation continues with the new control channel. 12-54

The protection of the interface transformer is basically that of a conventional sub-station transformer:

Differential protection Over-current protection Restricted earth fault protection Transformer oil & winding temperature, gas detector and oil level

The main ac bus typically uses differential protection and ac circuit breaker failure protection. The converter bus, i.e. the zone between the converter and the interface transformer, typically has differential and abnormal ac voltage protection. Over-current protection is not typically provided for the phase reactor, since the reactor has considerably higher over current capability than the semi-conductor devices. The ac harmonic filters have capacitor unbalance and resistor/reactor over-load protection similar to that used on a LCC HVDC scheme. The protection has to take into account the higher frequencies flowing in the filter components, compared with those in a LCC HVDC scheme. The converter valve protection typically includes:

Converter over-current protection. This protection uses temporary blocking to limit transient currents, and uses blocking and tripping in the event of persistent over-current AC terminal short circuit current protection VSC Valve short circuit protection High dI/dt IGBT current. This is also used for short circuit current detection IGBT monitoring. To ensure that the converter is tripped in the event of more IGBT failures than the number of redundant levels.

The converter dc protection includes abnormal dc voltage protection, which protects the converter from dc over-voltage and unbalanced dc voltage. An under-voltage protection is also included, to protect the VSC valves from operating when the dc voltage is inadequate for the gate unit power supply. The VSC Transmission scheme typically also has a number of other protections and indications, including VSC Valve cooling protection and alarms, dc cable fault location, auxiliary power supply protection and alarms and fire protection.

12.9 Installation Considerations

Since a VSC Transmission scheme does not require breaker switched ac harmonic filters, the land areas required for its substations are considerably less than those required for a LCC HVDC scheme. The actual area required depends on the ac voltage level as well as on the power 12-55

capability and the dc voltage. Typically, a VSC Transmission substation will occupy 40% or less of the space required for a LCC HVDC installation of the same power rating and ac connection voltage. Because the VSC Transmission substation is relatively compact, it becomes reasonable to enclose all the equipment between the secondary side of the interface transformer and the dc cable connection in a warehouse type building. Enclosing the electrical equipment in a building provides a number of advantages:

It provides containment for the electromagnetic noise generated by the high frequency switching of the converter. The floor, ceiling and walls include an electric screen, which functions similar to a faraday cage. This is a similar requirement to that for a LCC HVDC scheme, which requires the thyristor valves to be located in an electrically screened valve hall. It provides protection against adverse weather conditions. Depending on the voltage used for the dc transmission and the pollution level at the place of installation, the building can be either fully enclosed and pressurized, with a closed cooling system for the valve hall environment, or can be a relatively open building, which allows air to enter and escape freely. It provides attenuation for the audible noise generated by the converter operation. The audible noise from the electrical equipment tends to be at higher frequency than for a LCC HVDC scheme, which makes it somewhat easier to attenuate the noise. It makes it unnecessary to mount equipment on steel structures, as access can be restricted to the equipment to times when the scheme is out of service and safely earthed down. This means that the overall building height can be kept to a minimum.

The compactness of the VSC Transmission substation is of considerable benefit if it is intended to install one of the terminals off shore (e.g. for a wind farm application, or for the feeding of power to an off shore oil or gas production platform) or in a downtown city area. The VSC Valves may be mounted in special enclosures at the factory, and shipped to site after extensive testing. The control and protection equipment is similarly housed in a special enclosure, and extensively tested prior to shipment to site. The enclosures are placed on relatively simple foundations within the warehouse type building. The interface transformer, ac circuit breaker, ac isolator and PLC/RFI are typically located outdoors, as are the cooling fans required for the VSC Valves and other equipment. In order to satisfy local requirements in terms of audible noise, the overall design of the VSC Transmission substation must take into account all sources of audible noise. The main sources include the interface transformer and the VSC Valve cooling equipment, but attention must also be paid to the noise from the ac harmonic filters, HF filters and DC capacitors. As the latter components may be inside a warehouse style building, incorporation of noise attenuation in the roof and walls of the building can reduce the noise from these components to an acceptable level. Additionally, consideration may be given to the orientation of the substation, such that maximum 12-56

distances are developed between sensitive locations external to the station and the noise producing components. The use of underground dc cables, which are much less visually intrusive than overhead lines, for the connection of the VSC Transmission substations makes it easier to find suitable locations for the substations close to the desired point of connection.

12.10 Modularity
Whilst a large number of LCC HVDC installations are in operation worldwide, each installation tends to be tailor made for the application. Standard ratings and dc voltage levels tend to be brought about because of technical limitations, e.g. 500kVdc is considered a proven technology, which provides reliable and economic transmission at high power. Similarly, 3000MW is considered the maximum economic rating for a bipolar LCC HVDC scheme operating at 500kVdc, partly because the associated direct current is close to the maximum available current capability for high voltage thyristors, and partly because the impact that a failure of a larger rated scheme would have on the ac network performance. However, for smaller LCC HVDC schemes a large range of direct voltages and converter ratings are in use. Other issues which makes it more natural to tailor-make a LCC HVDC scheme are the reactive power absorption by the converter, and the generation of relatively large amplitudes of harmonic current at low harmonic orders. The harmonic filters supplied to limit the harmonic interaction with the ac network also serve the dual purpose of providing compensation for the converters reactive power absorption, and need to be specially designed for the particular ac network. Additionally, the reactive power banks are a potential source of large ac over-voltages, during fault recovery and malfunction of the converter. Therefore, insulation co-ordination for a LCC HVDC station has to take into account the specific ac network conditions, and the reactive power compensation equipment installed. Naturally, within the tailor made LCC HVDC solution, each manufacturer uses a number of standard building blocks from which the converter valves, control and protection system, circuit breakers, and other switchgear etc, are constructed. Nevertheless, the engineering of an LCC HVDC scheme can account for up to 15% of the total cost of converter stations with a rating of 500MW, the percentage decreasing as the rating increases. A VSC Transmission substation does not usually have large switchable reactive power bank, but has the ability to control its reactive power by converter action. Therefore, ac overvoltage issues are less onerous than with a LCC HVDC scheme. Similarly, as the harmonics generated are at relatively high harmonic orders, the interface transformer typically dominates the harmonic impedance as seen from the converter, making the design of the ac harmonic filters easier. The application of a VSC Transmission scheme in an ac network becomes much more like the installation of a generator (or large machine), than the installation of a complex power electronic system. Therefore, the likelihood of being able to standardize on solutions is greater for a VSC Transmission, than for a LCC HVDC.

12-57

Nevertheless, there are a number of engineering issues that need to be addressed for specific VSC Transmission substation sites, just as they have to be addressed for a LCC HVDC scheme. In addition to determination of the basic design parameters for the converter, converter reactor, interface transformer and ac switchgear, the following issues need to be addressed:

Harmonics Audible Noise Radio Frequency Interference Power Line Carrier Noise Cooling plant design

In many cases the action may be limited to confirmation that the performance is met with the manufacturers standard solution, thereby minimizing the design work. At the end of 2005 ABB was the only manufacturer offering commercial VSC Transmission solutions. ABB had completed 5 commercial VSC Transmission schemes, with a further scheme in progress, in addition to two demonstrators (Helsjon, 3MW and Tjreborg, 7MW) and a back to back scheme (Eagle Pass, 36MVA). The rating of the 6 commercial schemes were:

Gotland, 50MW, 80kVdc Direct Link, 3 x 60MW, 80kVdc Murray Link, 200MW, 150kVdc Cross Sound, 330MW, 150kVdc Troll, 2 x 40MW, 60kVdc Estlink, 350MW, 150kVdc

More information about these schemes will be provided in section 12.12. ABB documentation describing the HVDC Light technology [23] sets out 9 standard solutions for VSC Transmission. These use dc voltages of 80kVdc, 150kVdc or 300kVdc and direct current of 627, 1233 or 1881Adc, providing sending end power levels between 101MW and 1140MW. The interface transformer and ac switchgear is designed according to the ac connection voltage. The advantages of standard solutions include:

Cost reduction because of reduced engineering Lower project risks because of use of proven solutions Shorter project delivery times because of reduced engineering and setup times Immediate support from manufacturer in case of problems because a larger knowledge base will be available

12-58

For a potential customer the perceived disadvantage of using a standard solution may be that the standard rating is higher than that actually required. However, the cost differential, between the larger standard solution and a tailor made solution may be minimal. Additionally, the larger rating may provide better future proofing of the transmission solution.

12.11 Reliability, Availability and Maintainability

Reliable performance and good availability is achieved by good design practice that takes into account the steady state, dynamic and transient stresses likely to occur during the service life of the equipment, and which incorporate equipment margins where necessary. A protection system, which quickly detects the onset of a fault and takes the correct action to minimize any damage caused, is another essential part of good design. When the VSC Transmission substation is unmanned, as is typically the case, the clear communication and presentation of indications, alarms and protection signals to the system operator is another essential requirement for good availability performance. The reliability of equipment or sub-systems, which may be subject to random failures of components, can be improved significantly by the use of on-line redundancy. In a VSC Transmission scheme, redundancy is typically provided in the following equipment and subsystems:

VSC Valves additional series connected IGBT levels Cooling plant additional pumps, cooling units and fans Control system complete duplication, one system in hot stand-by AC and DC capacitors additional series and parallel connected elements Auxiliary power supplies two independent supplies and two fully rated systems, each one capable of supplying all station loads, with automatic changeover between the two.

The Availability of a VSC Transmission system can be increased by:

The provision of adequate spares such that failed equipment can be replaced and the system returned to operation as quickly as possible. Some expensive spare components could be shared between the substations, provided that suitable transport infrastructure exists to quickly move the spare component to the other station, in case of need. Availability of well-trained maintenance and repair personnel, who can quickly attend the VSC Transmission substation, thereby cutting down the outage time and reducing unavailability. The ability to access the information from Digital Transient Recorders and Sequence of Events Recorder from a remote location, where specialists can interpret the data in order to determine the cause of failure and the appropriate actions to be taken.

Prior to operation of the VSC Transmission scheme, the only way to determine the Reliability and Availability performance is through statistical calculations. The liquidated damages should reflect the potential loss of revenue resulting from poorer than guaranteed performance, as well 12-59

as the manner in which the performance is to be measured. Typically, manufacturers will wish to limit their financial risks, e.g. by placing an overall limit on liquidated damages. In addition to any financial damages, consideration could be given to specifying that the manufacturer shall rectify all problems during the Reliability and Availability period, shall report on each problem and, where relevant, shall make design changes to prevent future occurrences. The maintenance activities for a VSC Transmission scheme are very similar to those required for a LCC HVDC scheme, but can be completed in a shorter time since less equipment is provided, particularly on the ac side of the converter (ac filters and ac switchgear). Certain on-line redundant equipment in a typical VSC Transmission substation can be maintained with the system live, but the scheme would be at higher risk of trip during such maintenance work. The equipment may include:

Cooling pumps and fans Control and protection system Auxiliary power supply equipment (e.g. batteries and battery chargers, and ac switchboards)

Maintenance of other equipment requires a complete shut down of the VCS Transmission substation. Typically, all substations in a VSC Transmission system are maintained simultaneously during the shutdown, to minimize the overall shutdown time. This approach requires a larger maintenance team, than if the substations were maintained sequentially. During the shutdown of the VSC Transmission system for maintenance, power supply to any island loads must be provided by alternative means. If two parallel VSC Transmission schemes are used then scheduled maintenance work is typically done during periods of low load in the ac network. As the VSC Transmission technology has now been in service in commercial applications for several years, the early learning experiences gained have been incorporated in the design of later schemes, resulting in continuing reliability and availabilityperformance improvements. Commercial projects are now being quoted with guaranteed Reliability and Availability performance similar to that of LCC HVDC schemes, i.e. in the range of 98 to 99%. Formal records of the Reliability and Availability performance of have been collected by CIGRE for LCC HVDC schemes for many years [24]. Similar records have not been collected for VSC Transmission schemes. The only published information describing the actual Reliability and Availability performance for VSC Transmission projects identified during the writing of this chapter can be summarized as follows:

The IGBT failure rate on the Cross Sound Cables scheme during the period 1st July 2004 and 30th June 2005 was 0.38%, compared with the guaranteed rate of 0.5% per annum (Presentation at 8th EPRI FACTS Meeting, Stamford, CT, 17-19 August 2005.) Long Island Power Authority has expressed satisfaction with the availability and reliability of the Cross Sound scheme in a press release issued on 14th July 2005.

12-60

During the first year of operation the scheme transmitted 1,195,895 MWh of electricity to Long Island, a utilization of 41% (LIPA Press statement mentioned above). Apparently, the scheme has seen higher utilization since this report, with the power flow regularly being 330MW in the period between 9am and 10pm. The daily power flow on the scheme can be seen on the New England ISO website. There were no emergency or Forced Outage Applications for the Cross Sound project during the first 6 months of 2005 (presentation by New England ISO). The Annual Report provided by the Technical Regulator of South Australia [25] shows that the total outage time for the Murray Link scheme during the period 1st July 2003 to 30th June 2004 was 131hrs, equivalent to an availability of 98.5%, which is better than the target availability of 98% for the project.

12.12 VSC Transmission Application Examples

This section provides information about the 7 VSC Transmission systems in service or under construction at the end of 2005. Since ABB is the only supplier to have delivered commercial VSC Transmission schemes, all illustrations in this section have been obtained from ABB, and their kind assistance is appreciated. 12.12.1 Gotland, Sweden The purpose of the Gotland VSC Transmission scheme [26] was to provide a 70km long transmission link for wind generation at the south of Gotland to the more populated North, from where energy can also be exported to mainland Sweden when the generation exceeds the load requirement on Gotland. Transmission of the power by underground cable was more acceptable to the population than an additional 130kVac overhead line. The VSC Transmission scheme has also resulted in power quality improvements on the island, in particular voltage flicker at the terminals of a large industrial processing plant on the East Coast has been significantly reduced. The power quality of the power from the wind generators is also significantly better than with the use of an ac overhead or cable line connection between the North and the South. The scheme operates at 80kVdc and has a rating of 50MW. A simplified single line diagram of the transmission scheme is shown in Figure 12-28. The VSC Transmission scheme operates in parallel with an existing 70kVac line, not shown in the diagram.

12-61

Figure 12-28 Single Line diagram of the Gotland VSC Transmission scheme.

The scheme uses a 2-level converter with triangular carrier PWM switching at 1950Hz. Both terminals use ac voltage control, and are capable of either dc voltage control or active power control. Normally, dc voltage control is at the inverter end. The scheme operates unmanned and is controlled from the control room on Gotland. The scheme does not require telecommunication between the terminals for its operation. Figure 12-29 shows the VSC Transmission substation at Bcks, near Visby in the North of the island. The VSC Valve cooling plant can be seen on the side of the building.

Figure 12-29 The VSC Transmission substation at Bcks.

12-62

The Gotland scheme entered service in 1999. The scheme is reported to have met all the specified requirements in terms of power transmission capability, and to have provided significant improvement of power quality in the Northern part of the ac network. 12.12.2 Tjreborg, Denmark The Tjreborg VSC Transmission scheme is a wind power transmission demonstration project, built by ELTRA, Denmark in anticipation of the construction of several large offshore wind farms [27]. The scheme has a relatively small rating of 7MW, and operates at 9kVdc with a dc cable length of only 4.5km. However, the low rating was considered sufficient for the purpose of investigating the benefits which could be derived from the use of a variable frequency supply to test a number of different wind generators.. A simplified single line diagram of the transmission scheme is shown in Figure 12-30. The VSC Transmission scheme operates in parallel with an existing 10.5kVac line as shown in the diagram.

Figure 12-30 Single Line diagram of the Tjreborg VSC Transmission scheme.

The scheme uses a 2-level converter with triangular carrier PWM switching at 1950Hz. Both terminals use ac voltage control, and are capable of either dc voltage control or active power control. Normally, dc voltage control is at the inverter end. The interface transformers are dry type transformers. The scheme operates unmanned and is controlled from the control room at Tjreborg Enge. The scheme does not require telecommunication between the terminals for its operation.

12-63

Figure 12-31shows the VSC Transmission substation at Tjreborg Enge.

Figure 12-31 The VSC Transmission substation at Tjreborg Enge.

The Tjreborg scheme entered service in 2000. The scheme has been used as intended, and a number of papers have been published to present the results of the studies[28]. There have been no reports of any major problems. 12.12.3 Direct Link, Australia The Direct Link VSC Transmission scheme was built to provide a non-regulated power link for power trading between New South Wales and Queensland Australia [29]. The use of underground cables enabled planning permission to be obtained much earlier than a competing regulated ac overhead line, thereby enabling opportunity to be taken from a significant energy price differential between the two states. The scheme consists of 3 identical parallel VSC Transmission schemes, providing a total transmission capability of 180MW, 195MVA. The ac connection points in both states are relatively weak, making the controllability of reactive power particularly useful in this application. Revenues can be earned not only when transmitting active power, but also by providing a reactive power or ac voltage control service. The use of virtually the same converter rating as for the Gotland scheme meant that the equipment could be manufactured very quickly. The three cable systems operate at 80kVdc and have a dc cable length of 65km. A simplified single line diagram of the transmission scheme is shown in Figure 12-32.

12-64

Figure 12-32 Single Line diagram of the Direct Link VSC Transmission scheme.

The scheme uses 2-level converters with triangular carrier PWM switching at 1950Hz. Both terminals can use ac voltage or reactive power control, and are capable of either dc voltage control or active power control. Normally, dc voltage control is at the inverter end. The scheme operates unmanned and is controlled from the grid control center. The scheme does not require telecommunication between the terminals for its operation. Figure 12-33 shows the VSC Transmission substation at Mullumbimby.

Figure 12-33 The VSC Transmission substation at Mullumbimby

12-65

The Direct Link scheme entered service in 2000. Soon after entering service the scheme suffered numerous failures of the joints between cable segments. The faults were traced to a manufacturing error, which necessitated gradual replacement of a large number of cable joints. As a non-regulated trading link the Direct Link VSC Transmission scheme can be used in many different ways, including bulk power transmission and power balancing between networks. Figure 12-34 shows the variation of active power flow during one summer week, where rapid variation of the power between import and export can be seen. For a LCC HVDC scheme such rapid and frequent changes in power flow would have necessitated numerous switching operations of the ac harmonic filters, reducing the time interval between maintenance and adjustment of the breakers substantially.

Figure 12-34 Power flow on the Direct Link scheme during one week in summer

Apart from the cable joint failures, the scheme has been reported to meet all requirements and there have been no reports of any major problems. 12.12.4 Murray Link The Murray Link VSC Transmission scheme was the second non-regulated power link for power trading in Australia, and provides a link between the power networks in the states of South Australia and Victoria [30]. A simplified single line diagram of the transmission scheme is shown in Figure 12-35.

12-66

Figure 12-35 Single Line diagram of the Murray Link VSC Transmission scheme

As for the Direct Link project, the use of underground cables enabled planning permission relatively quickly. The scheme is rated at 200MW, 150kVdc and the cable length is 180km, making this the longest overland cable scheme in the world. The ac network connection points at both ends of the scheme are relatively weak. As with the Direct Link scheme, revenue can be earned not only when transmitting active power, but also by providing a reactive power or ac voltage control service. This scheme marked the second generation of VSC Transmission schemes using a transmission voltage of 150kVdc as well as 3-level Neutral Point Clamped converters with triangular carrier PWM switching at 1350Hz. The lower PWM frequency and the use of 3-level converters resulted in a significant decrease in the power loss, compared with the earlier VSC Transmission schemes. Figure 12-36 shows the VSC Transmission substation at Berri.

Figure 12-36 The VSC Transmission substation at Berri.

12-67

Both terminals can use ac voltage or reactive power control, and are capable of either dc voltage control or active power control. Normally, dc voltage control is at the inverter end. The scheme operates unmanned and is controlled from the grid control center. The scheme does not require telecommunication between the terminals for its operation. The Murray Link scheme entered service in 2002. There have been no reports of any major problems. 12.12.5 Cross Sound, USA The Cross Sound VSC Transmission scheme provides a non-regulated energy trading link between New Haven, Connecticut and Shoreham on Long Island, Massachusetts. A 40km submarine cable across the Long Island Sound connects the two terminals. The scheme was designed and built almost at the same time as the Murray Link, and uses the same technology as that scheme, operating at 150kVdc, but with a higher power rating of 330MW, 346MVA [31]. A simplified single line diagram of the transmission scheme is shown in Figure 12-37.

Figure 12-37 Single Line diagram of the Cross Sound VSC Transmission scheme

The submarine cables and fiber-optic cables were bundled together to minimize the impact on protected shellfish. The cable bundle was buried 6 feet below the seabed, to achieve protection against damage from fishing vessels and anchors, and to minimize the long-term impact on shellfish living on the seabed. Hard rock below the seabed made it impossible to achieve a burial depth of 6 feet in some areas, and environmentalists objected to the potential impact of the cables on creatures on the seabed. Therefore, it was not possible to enter commercial service on conclusion of commissioning in 2002. The dispute over the environmental issue resulting from the lower burial depth of the cable continued until the Big Black Out in North East USA on the 14th August 2003, when the Cross Sound scheme played a major role in restoring power supply to Long Island. The scheme continued in service until April 2004, when transmission again had to be stopped due to the environmental protests. A settlement was reached in June 2004, resulting in immediate re-energisation and operation of the link. This scheme use second generation VSC Transmission technology with 3-level neutral Point Clamped converters with triangular carrier PWM switching at 1260Hz. The lower PWM 12-68

frequency and the use of 3-level converters resulted in a significant decrease in the power loss, compared with the first generation VSC Transmission schemes. Figure 12-38 shows the VSC Transmission substation at Shoreham. The interface transformers can be seen in the lower left corner, RFI and PLC filters can be seen between the transformers and the cooling units for the VSC Valves. The building behind the coolers houses the ac harmonic filters, and the larger building in the lower right hand corner houses the phase reactors, VSC Valves, dc smoothing reactors etc.

Figure 12-38 Cross Sound VSC Transmission substation at Shoreham.

Both terminals can use ac voltage or reactive power control, and are capable of either dc voltage control or active power control. Normally, dc voltage control is at the inverter end. The scheme operates unmanned and is controlled from the grid control center. The scheme does not require telecommunication between the terminals for its operation. There have been no reports of any major problems. 12.12.6 Troll, Norway The Troll VSC Transmission scheme provides the first HVDC power connection to an offshore gas production platform[9]. The Troll platform has 2 large compressors to enable more gas to be extracted from the sub-sea reservoirs, and to transport the gas through the pipeline to the processing station on mainland Norway. As the gas reservoir is being emptied, more and more power is needed for the extraction, and it was decided to import the power from the mainland grid, rather than to provide large on-platform generators. The main reasons for the choice were financial, e.g. because of lower taxes due to reduced CO2 emissions and lower maintenance and operation costs. 12-69

Two parallel systems are being provided, each rated at 40MW and operating at 60kVdc. A simplified single line diagram of one of the transmission schemes is shown in Figure 12-39. The cable distance is 70km, and each scheme uses two cables. The VSC Transmission substation at Kolness uses an interface transformer for the connection to the converter to the mainland 132kV ac network. The converter on the platform connects directly through the phase reactor to a high voltage motor, which is also supplied by ABB. The high voltage motor drives the precompressor.

Figure 12-39 Single Line diagram of one of the Troll VSC Transmission schemes

This scheme uses first generation VSC Transmission technology with 2-level converters with triangular carrier PWM switching at 1950Hz. The main reason for this choice was the smaller space and lower weight occupied when compared with a 3-level NPC converter, space and weight being an extremely significant consideration for an offshore installation. The safety requirements placed on equipment in a petrochemical environment meant that many of the normal protection and control principles had to be revised. In particular, continuity of compression was less important than safety, resulting in the normal temporary blocking actions being redesigned to provide block and trip actions. The offshore environment also placed stringent requirement on the design of equipment. Virtually all the equipment is housed within an enclosure, which was constructed on the mainland and shipped to the platform, where it was raised into position on the reconstructed platform using a large crane. Figure 12-40 shows the enclosure in position on the Troll A platform. The enclosure houses the two separate VSC Transmission substations.

12-70

Figure 12-40 Troll A VSC Transmission enclosure.

The mainland terminal is in dc voltage control and can use ac voltage or reactive power control, as required by the network operator. The offshore converter is controlled to function as a varible speed drive, capable of providing a frequency between 0 and 63Hz and an ac voltage between 0 and 56kVac. The scheme operates unmanned and is normally controlled from the onshore control center. The scheme does not require telecommunication between the terminals for its operation. The enclosure was installed in 2004, and the first scheme was commissioned late 2004/early 2005, the second following during the spring and summer of 2005. The scheme entered service in the autumn of 2005, after extensive testing. There have been no reports of any major problems. 12.12.7 Estlink, Estonia-Finland The Estlink project was ordered early 2005, and will provide an interconnection between Estonia and Finland, crossing the Gulf of Finland. The primary objective of the link will be to provide 12-71

the Nordic electricity market with electricity generated in the Baltic States. The European Union has approved the link as a non-regulated trading link. The rating of the interconnector is 350MW, 150kVdc. The VSC Transmission substation will be located at Harku, near Tallinn in Estonia and Espoo near Helsinki in Finland. The cable between the two VSC Transmission substations will include 74km submarine cable and 31km land cable. No information is yet available describing the implementation of the VSC Transmission substations.

12.13 Acknowledgement
The author acknowledges with thanks ABBs kind permission to use some illustrations from their book Its time to Connect and other publications.

12.14 Selected References

12.14.1 General and Applications [1] [2] VSC Transmission, CIGRE Working Group B4.37, CIGRE Brochure No 269, April 2005. Feasibility of Gate Turnoff Thyristor in a High Voltage Direct Current Transmission System. EPRI Research Project 2443-5, EL5332, Electric Power Research Institute (EPRI), Palo Alto, California , August 1987. Power Electronics: Converters, Applications and Design. Mohan, N., Undeland, T.M., Robbins, W.P., John Wiley & Sons, Hoboken, NJ, 2003. Power Semiconductors in Transmission and Distribution Applications. Chokhawala, R., Danielsson, B., ngquist, L. Proceedings of the 2001 International Symposium on Power Semiconductor Devices & ICs (ISPSD), Osaka, Japan, pp. 3-10. Power system stability benefits with VSC DC-transmission systems. Johansson SG, Asplund G, Jansson E & Rudervall R, CIGRE 2004 Session, Paper B4-204. "Co-ordination of Parallel AC-DC Systems for Optimum Performance." Castro, A.D., Ellstrm, R., Hffner, Y.J., Liljegren, C., Power Delivery Conference, Madrid, September 1999. Enhancing of power quality and availability in distribution systems by means of Voltage Source Converters. Grunbaum, R., IEE 16th International Conference and Exhibition on Electricity Distribution, CIRED2001, Conference Publication No. 482, Amsterdam, June 2001.

[3] [4]

[5] [6]

[7]

12-72

[8]

Cost-efficient XLPE cable system solutions. Karlstrand, J., Bergman, G., Jnsson, H., IEE 7th International Conference on AC-DC Power Transmission, Conference Publication No. 485, pp. 33-38, November 2001. New Application of Voltage source converter (VSC) HVDC to be installed on the gas platform Troll A, Hyttinen, M., Lamell, J.O., Nestli, T.F., CIGRE session paper B4-210, Paris, 2004. The challenge of integrating large-scale offshore wind farms into power systems. Belhomme, R., Joncquel, E., Sbrink, K., Abildgaard, H., Woodford, D., CIGRE session paper B4-204, Paris, 2002. HVDC Light experiences for power transmission from offshore wind power parks. Eriksson, K., Liljegren, C., K. Sbrink, K., ASME/AIAA, Wind Energy Symposium, Reno NV, USA, January 2004.

[9]

[10]

[11]

12.14.2 Implementation of VSC Transmission schemes [12] Topologies for VSC Transmission. Andersen B.R., Xu L., Wong K.T.G., 7th International Conference on AC-DC Power Transmission, IEE Conf. Publ. No.485, pp. 298-304, IEE 2001. HVDC Transmission Systems using Voltage Sourced Converters - Design and Applications. Schettler F., Huang H., Christl N., IEEE PES Summer Meeting 2000, Seattle, Washington, USA, PWM and Control of Two and Three-level High Power Voltage Source Converters. Lindberg, A., ISSN-1100.1616. TRITA-EHE 9501, Royal Institute of Technology, Stockholm, Sweden. 1995. Programmed PWM Techniques to eliminate harmonics: a critical evaluation. Enjeti, P.N., Ziogas, P.D., Lindsay, J.F., IEEE Transactions on Industry Applications. Vol. 26, No. 2, pp. 302-316, March/April 1990. Recommended practices and requirements for harmonic control in electrical power systems, IEEE Standard 519-1992 (replacing IEEE Standard 519-1981). Connection of harmonic producing installations in AC high voltage networks with particular reference to HVDC. Guide for limiting interference caused by harmonic currents with special attention for telecommunication systems. JTF 02 of CIGR WG 14.03 and CCO2 (CIGR 36.05/CIRED 2), Electra reference No. 159, 1995. Guide to the Specification and Design Evaluation of AC Filters for HVDC Schemes. CIGRE WG 14.30, CIGRE Publication No. 139.

[13]

[14]

[15]

[16] [17]

[18]

12-73

[19]

Cost-efficient XLPE cable system solutions. Karlstrand, J., Bergman, G., Jnsson, H., IEE 7th International Conference on AC-DC Power Transmission, Conference Publication No. 485, pp. 33-38, November 2001. A control system for HVDC transmission by voltage sourced converters. , Nakajima, T., Irokawa, S., Proceedings, IEEE Power Engineering Society Summer Meeting 1999, Vol. 2, pp. 1113-1119, 1999. DC Overvoltage control during loss of converter in Multi-terminal Voltage Sourced Converter Based HVDC (M-VSC-HVDC), Lu W, Ooi B.T., IEEE Transactions on Power Delivery. VSC transmission operating under unbalanced AC conditions - analysis and control design. Xu L., Andersen B.R., Cartwright P.J., IEEE Transactions on Power Delivery, Volume 20, Issue 1, Jan 2005. Its Time to Connect, Technical description of HVDC Light technology, ABB, available from www.abb.com/hvdc. A Survey of the Reliability of HVDC Systems throughout the World during 2001-2002, Vancers I., Christofersen D.J., Leirbukt A., Bennett M.G., CIGRE WG B4.04, CIGRE 2004 Session, paper B4-201. Annual Report by the Technical Regulator 2003/04, Electricity, Government of South Australia, available from www.technicalregulator.sa.gov.au

[20]

[21]

[22]

[23] [24]

[25]

12.14.3 Description of VSC Transmission schemes [26] Gotland HVDC Light Transmission - worlds first commercial small scale dc transmission. Axelsson U., Holm A., Liljegren C., Eriksson K., Weimers L., CIRED, Nice, France, 1999. DC Feeder for Connection of a Wind Farm, Sbrink K., Srensen P.L., Christensen P., Andersen N., Eriksson K., CIGR Symposium Paper 500-06, Kuala Lumpur, Malaysia, 1999. Development and testing of ride-through capability solutions for a wind turbine with doubly fed induction generator using VSC transmission. Eek, J., Pedersen, K.O.H., Sbrink, K., CIGRE session paper No. B4-302, Paris, 2004. The Direct Link VSC-based HVDC project and its commissioning, Railing B.D., Moreau G., Wasborg J., Stanley D., Miller J.J., Jiang-Hafner Y., CIGR session paper No B4-108, Paris, 2002.

[27]

[28]

[29]

12-74

[30]

Murray Link, The Longest Underground HVDC Cable in the World. Mattsson W.I, Ericson A., Railing B.D., Miller J.J., Williams B., Moreau G., Clarke C.D., CIGR session paper No B4-103, Paris, 2004. Cross Sound Cable Project, second Generation VSC Technology or HVDC, Railing, B.D., Miller, J.J., Steckley, P., Moreau, G., Bard, P., Ronstrom, L., Lindberg, J., CIGR session paper No B4-102 , 2004.

[31]

12-75

SAMPLE AUTHORS GUIDE

Authors Guide
EPRI Power Electronics-Based Transmission Controllers Reference Book (The Gold Book)
November 2005

DRAFT

D-1

Contents

Section

Page Number

Introduction Production Process How to Set Up Files Format: Headings Format: Tables and Figures Format: Equations SI Units and English Measures Writing Principles Language Guidelines References Index Copyright Issues Terms

3 4 8 9 10 11 12 14 15 17 19 20 21

Figures Figure 1. Production Process Figure 2. Sample Copyedit Page Figure 3. Sample Layout Page

Page Number 5 6 7

D-2

Introduction This Authors Guide is designed to provide guidelines for authors to follow in the writing of chapters for the Gold Book. These guidelines will help to ensure that the completed book has a consistency and uniformity of presentation and style. Sources The guidelines are based on several sourcesincluding the EPRI Editorial Style Guide, The Authoritative Dictionary of IEEE Standards Terms (7th edition), and the EPRI AC Transmission Line Reference Book, 200 kV and Above (Red Book). Editorial Committee The Editorial Committee for the Gold Book will review each chapter to ensure the accuracy and consistency of the book. The Editorial Committee includes: (*to be determined*).

D-3

Production Process The production process is designed to be easy for authors to use, and to require no additional purchase of software for writing or review. The production will take place in five steps (see Figure 1): 1. First Drafts First drafts are written by authors or author-teams. First drafts should not be submitted for production until the entire chapter is complete and the draft has been reviewed and approved by all members of the chapter team. 2. Technical Review Once the first draft is complete, it should be submitted for Technical Review. The reviewer(s) may draft a detailed review memo. Alternately, reviewers may also choose to comment on and suggest revisions using the Track Changes mode in Microsoft Word (Tools, Track Changes). Authors will review these suggestions and accept or reject these changes using Track Changes to produce a clean manuscript. 3. Copyedit After changes from the Technical Review have been resolved, the chapter will be copyedited. Copyediting will revise the text for spelling, punctuation, grammar, and adherence to the style guidelines. The copyedit will be made in the Track Changes mode of Microsoft Word (see Figure 2). The copyedited chapter will sent back for review and approval by the authors before proceeding. 4. Layout Once changes from copyediting have been resolved, the chapter will be laid out. The production team will lay out chapters in a two-column page format (see Figure 3). Pages will be modeled on the format and look of EPRIs AC Transmission Line Reference Book, 200 kV and Above (Red Book). Important: Substantive revisions to chapters should be made BEFORE layoutas much as possible. Once the book is in layout, changes are more time consuming. (Compression ratio: The laid-out pages reduce the chapter page counts by about half.) 5. Publication Once the document has been laid out and approved by authors, the book will be published.

D-4

1. First Draft

2. Technical Review

Author Review

3. Copyedit

Author Review

4. Layout

Author Review

5. Publication

Figure 1 Production process.

D-5

Figure 2 Sample copyedit page.

D-6

Figure 3 Sample page layout.

D-7

How to Set Up Files When writing first drafts, please use the following guidelines for creating your files: Text Write files in Microsoft Word. Drafts should be in 12 point type, Times New Roman, single-spaced, in one column, with 1inch margins. For new paragraphs, do not indent first line. Just do a double return between paragraphs. Include all text, tables, equations, and figures in the file.

Equations Write equations in Microsoft Equation 3.0.

FiguresGraphs and Photographs Important: If graphs or photos are created in any other program than Word, PowerPoint or Excel, please go to File, Save As. Then save files as .tif, .jpg, or .bmp format and insert that figure in the document. (.tif is preferable.) If you include figures in Excel, supply the original spreadsheet. Annotate graphs. Label parts of the graph. Do not use legends. If necessary, hand-drawn figures may be submitted to the editor to be drawn. Use the font Helvetica for wording in figures, if possible. Do not use the font Arial; it may lose or change characters during production. For photos, use a version as close to the original as possible.

Index Make an alphabetical list of key words and their section location for an Index. See instructions on page 19.

Completed Files Post completed files on the FTP site in the folder for your chapter. If your completed chapter is large (more than a few MB), please post it in separate section files. This will create files that are more manageable for others to handle.

D-8

Format: Headings For consistency, all chapters should have the same style of main headings and subheadings. There are four levels of headings, as shown below. Main Heading One-Decimal Number, All Capital Letters, Boldface Example: 1.2 FUNDAMENTAL RELATIONSHIPS OF AC POWER TRANSMISSION First Subheading Two-Decimal Number, Capital and Lowercase Letters, Boldface Example: 1.2.1 Analytical Characterization Second Subheading Capital and Lowercase Letters, Boldface and Italic Example: Ideal (Lossless) Line Third Subheading Capital and Lowercase Letters, Italic Example: Conclusions

D-9

Format: Tables and Figures Tables and figures should be numbered with the number of the subsection in which they appear. That is, the first table in Subsection 2.2 would be Table 2.2-1. The first figure in Subsection 2.2 would be Figure 2.2-1. Tables Table titles should be initial capped. Example: Table 2.5-1 Steady-State Operating Characteristics Table titles appear above tables. Table footnotes are superscript numbers. Figures Figure captions should be first-word-only capped and with a period at the end. Example: Figure 3.2-23 Simplified diagram of VSC transmission scheme. Figure captions appear below figures. Photographs are considered figures.

D-10

Format: Equations Equations should be written using Microsoft Equation 3.0, which is available free with Microsoft Word. (To access it, go to Insert, Object, and select Equation 3.0.) Equations should be numbered to the right of the equation, using the same system as for tables and figures. Example, the third equation in Section 3.4 would be 3.4-3. If equations are long, they may need to be broken in the layout. Please indicate where they can broken. When identifying symbols in an equation, put them in a vertical list, not run-in on one line. Put the Where: at the top of the list. If equations are jpg files, rewrite them in Equation 3.0. Or, submit them to the editor to be generated in Equation 3.0.

D-11

SI Units and English Measures Wherever reasonable, use the International System of Units (SI), with the English Units following in parentheses. Please note: If existing tables or figures use English units, leave as is. If conversion to SI units is difficult, leave as is. Imperial units may be used, but SI units are preferred, as the Gold Book is migrating to them for the future. SI Base Units Quantity length mass time electric current thermodynamic temperature Name meter kilogram second ampere kelvin SI Derived Units
Derived Quantity area volume speed, velocity acceleration wave number density, mass density specific volume current density magnetic field strength Name square meter cubic meter meter per second meter per second squared reciprocal meter kilogram per cubic meter cubic meter per kilogram ampere per square meter ampere per meter Symbol m2 m3 m/s m/s2 m-1 kg/m3 m3/kg A/m2 A/m

Symbol m kg s A K

SI Derived Units with Special Names and Symbols Derived Quantity frequency force pressure, stress energy, work, quantity of heat power electric charge electric potential difference capacitance Name hertz newton pascal joule watt coulomb volt farad
D-12

Symbol Hz N Pa J W C V F

electric resistance electric conductance magnetic flux magnetic flux density inductance Celsius temperature

ohm siemens weber tesla henry Degree Celsius

S Wb T H C

Exceptions Use gauss, milligauss. Put tesla in parentheses.

Usage Periods are not ordinarily used with abbreviations for units of measure (e.g., m, s, kg). One exception is the abbreviation for inch (in.). When it stands alone (e.g., 6 in.), it should have a period so that it is not confused with the preposition in. When used with an exponent (in2) or as part of a compound unit of measure (in/s), the period is omitted. For more information, see the 2001 edition of The International Systems of Units (SI), NIST Special Publication 330, which is available on the web at http://physics.nist.gov/Pubs/SP330/sp330.pdf.

D-13

Writing Principles Authors should keep in mind a number of general principles, so that all chapters are consistent in their presentation and style. Introduction Each chapter should begin with a brief (5-10 paragraphs) introduction. The introduction should describe the scope of the chapter and its context within the book. It may describe the topics significance and changing attitudes/approaches toward it. In addition, it should provide a quick roadmap to the main sections included within the chapter. Flow When writing your chapter, be sure the information is developed logically from section to section and within each section. Be aware of the number of sections: Is it within the average of other chapters? Avoid overly long or overly short sections. If you list three topics at the start of a section, discuss them in that order. Depth Present information in an appropriate level of detail. Avoid drilling down too deeply in one area and covering another too shallowly. Be aware of total page count, which will be discussed prior to the start of writing: Are you within the expected range? Tone Emulate the tone of other chapters. Use clear, declarative sentences. Appendices Present data, figures, and other detailed information, which would clutter the main text, in appendices.

D-14

Language Guidelines Guidelines for language should follow the EPRI Editorial Style Guide, which is at: http://www.epri.com/corporate/discover_epri/epri_facts/reportspecs/styleguide.html. Specific guidelines are as follows: Numbers Use numerals: For all numbers 10 and above. In tables and figures. With units of measure (5 Hz, 12C) With percentages (67%) Use words to express: All numbers zero through nine in text (e.g., A determination of conductor tension depends on whether we select three, four, or five beats as being indicative of the traveling wave return time.) A zero is ordinarily used before a decimal point: 0.01% Percentages Use %, not percent. Hyphens Hyphenate compound adjectives containing units of measure and time. a 7.3-m (24-ft) room a 2.54-cm (1-in.) diameter pipe a 3-m pole In almost all cases, hyphenate high- and low- adjectival compounds. high-voltage transmission Parallelism Be aware of parallelism. E.g., if you create a bulleted list in which most lines begin with a verb, start all lines with a verb. Length Avoid overly long sentences and paragraphs. Clarity is usually improved by shortening.

D-15

Periods/Spaces Use a single spacenot a double spaceafter the period at the end of a sentence. Units Spacing Follow IEEE style. Put one space between number and unit. E.g., 21.1 Hz. Abbreviations Follow IEEE spelling. Page Breaks In first drafts, do not pay attention to page breaks. Do not worry if tables break over two pages. These breaks will be fixed in page layout.

D-16

References References should be cited using the author-date method. The advantage of this system is that it avoids the need to track and revise footnote numbers. In text, to cite a source, write the authors last name and the date of publication in parentheses. For example: (Gyugyi 1988). Two or more authors would be written as: (Hingorani and Gyugyi 2000). If there are more than two authors, use et al. (Edris et al. 1950). If the author has more than one work in one year, use letters to distinguish (Grunbaum 1950a). If the author is an organization, include enough of the name to enable a reader to locate it in the end-of-chapter list. For example: (IEEE Working Group 1978). If there is no author, include the first few words of the title. For example: (EHV Transmission Line 1968). If you wish to specify a page number, put that after the date. For example: (Dalziel 1950, p. 1163). Citations should be inserted within the sentence as close as possible to the point being referenced. At the end of the chapter, provide a list of references alphabetized by the authors last name. If there is no author, alphabetize by the citations title. See examples below for the correct format. Only reference publicly available documents. Bibliography If desired, include a Bibliography separate from the References. It should include all major seminal works. Website References Website references may be used. In text, include a recognizable short form of the URL. At the end of the chapter, include the full URL.

D-17

Reference List Style Scientific Journal, One Author Gyugyi, L. 1994. Dynamic Compensation of AC Transmission Lines by Solid-State Synchronous Voltage Sources. IEEE Transactions on Power Delivery. Vol. 9. No. 2. April. Scientific Journal, Multiple Authors Enjeti, P. N., P. D. Ziogas, and J. F. Lindsay. 1990. Programmed PWM Techniques to Eliminate Harmonics: A Critical Evaluation. IEEE Transactions on Industry Applications. Vol. 26. No. 2. pp. 302-316. March/April. Conference Paper Schettler F., H. Huang, and N. Christl. 2000. HVDC Transmission Systems Using Voltage Sourced ConvertersDesign and Applications. IEEE PES Summer Meeting, Seattle, Washington.

D-18

Index When submitting your chapter, create an alphabetical list of key words and section numbers. This will be merged with lists from other chapters. Keywords in the index will be spelled in adherence with the spelling of the IEEE Dictionary of Standards Terms.

D-19

Copyright Issues If you plan to reproduce figures, photos, or tables from another publication, EPRI will need to seek copyright permissions from the publisher and/or author of the original publication. The Legal Department of EPRI has a specific form and process for obtaining copyright permissions for reproduction of material. Following completion of the first draft, the production team will work with EPRI Legal to obtain the required permissions. As an author, your responsibility is to maintain a clear list of references, showing the source of all reproduced material. Required information for books includes authors name, title, edition, publisher, year, and page number. Required information for journal articles includes authors name, article title, journal name, volume and issue numbers, and page numbers. Please note that page numbers are required in both forms, and are required in most publishers copyright documentation. Note: Government Sources. Much material published by government agencies is in the public domain, and does not require copyright permissions. However, please document all reproduced material, so that the production team can verify each case. Authors Publications. If you are reproducing material from your own earlier publication, we may still need to obtain permission to publish, because, in some cases, the publisher, not the author, holds the copyright. Denied Permissions. In some cases, the publisher may deny permission or require an onerous fee or pre-condition. In this event, we may need to create a new figure or table, or paraphrase the information. Revised Figure. If a figure is substantially changed, you do not need permission from the original source. Product Photo. A photo from a utility of a product on a line requires permission from the utility, not the product manufacturer.

D-20

Terms Bibliography. List of documents (books, articles, and websites) related to the subject. (See References.) Copyediting. Revision of text to correct spelling and grammar, and to ensure consistency of style, format, and usage. English Units. Units of measure commonly used in North America (e.g., feet, pounds, miles). Page Layout. Placement of the text, tables, and figures on pages in a consistent design, with specifications for type size, line spacing, margins, etc. References. List of documents (books, articles, and websites) cited in the text. (See Bibliography.) SI Units. (Le Systme International dUnits, or International System of Units). Metric measurement system, dominant measurement system used in international commerce. Technical Review. Review of draft documents by experts in the content field.

D-21

SAMPLE ONE-PAGE LAYOUT

E-1

Export Control Restrictions Access to and use of EPRI Intellectual Property is granted with the specific understanding and requirement that responsibility for ensuring full compliance with all applicable U.S. and foreign export laws and regulations is being undertaken by you and your company. This includes an obligation to ensure that any individual receiving access hereunder who is not a U.S. citizen or permanent U.S. resident is permitted access under applicable U.S. and foreign export laws and regulations. In the event you are uncertain whether you or your company may lawfully obtain access to this EPRI Intellectual Property, you acknowledge that it is your obligation to consult with your companys legal counsel to determine whether this access is lawful. Although EPRI may make available on a case-by-case basis an informal assessment of the applicable U.S. export classification for specific EPRI Intellectual Property, you and your company acknowledge that this assessment is solely for informational purposes and not for reliance purposes. You and your company acknowledge that it is still the obligation of you and your company to make your own assessment of the applicable U.S. export classification and ensure compliance accordingly. You and your company understand and acknowledge your obligations to make a prompt report to EPRI and the appropriate authorities regarding any access to or use of EPRI Intellectual Property hereunder that may be in violation of applicable U.S. or foreign export laws or regulations.

The Electric Power Research Institute (EPRI)

The Electric Power Research Institute (EPRI), with major locations in Palo Alto, California, and Charlotte, North Carolina, was established in 1973 as an independent, nonprofit center for public interest energy and environmental research. EPRI brings together members, participants, the Institutes scientists and engineers, and other leading experts to work collaboratively on solutions to the challenges of electric power. These solutions span nearly every area of electricity generation, delivery, and use, including health, safety, and environment. EPRIs members represent over 90% of the electricity generated in the United States. International participation represents nearly 15% of EPRIs total research, development, and demonstration program. TogetherShaping the Future of Electricity

2005 Electric Power Research Institute (EPRI), Inc. All rights reserved. Electric Power Research Institute and EPRI are registered service marks of the Electric Power Research Institute, Inc. Printed on recycled paper in the United States of America 1010635

ELECTRIC POWER RESEARCH INSTITUTE 3420 Hillview Avenue, Palo Alto, California 94304-1395 PO Box 10412, Palo Alto, California 94303-0813 USA 800.313.3774 650.855.2121 askepri@epri.com www.epri.com