Advanced Angle Stability Controls

CIGRÉ Technical BrochureAdvanced Angle Stability Controls Prepared by Task Force 17 of Advisory Group 02 of Study Committee 38 December 1999 International Conference on Large High Voltage Electric Systems Conférence Internationale des Grands Réseaux Électriques a Haute Tension CIGRÉ TF 38.02.17 Advanced Angle Stability Controls Convenor: Carson Taylor (USA) Members and contributors Florencio Aboytes (Mexico) Tom Anderson(Scotland) Miroslav Begovoc (USA) Henri Bourles (France) Joe Chow (USA) Jeff Dagle (USA) Peter Donalek (USA) Suma Geeves (UK) Paulo Gomes (Brazil) Adel Hammad (Switzerland) Michael Henderson (USA) Takashi Hiyama (Japan) Satoru Ihara (USA) Nelson. Zeni Junior (Brazil) Daniel Karlsson (Sweden) Kwang Lee (USA) Thibault Margotin (France) Jim McCalley (USA) Mojtaba Noroozian (Sweden) Torben Østrup (Denmark) Paulo Paiva (Brazil) Mrinal Pal (USA) Juan Sanchez-Gasca (USA) Dick Schulz (USA) Hisao Taoka (Japan) Ebrahim Vaahedi (USA) Louis Wehenkel (Belgium) Magnus Akke (Sweden) Göran Andersson (Sweden) Bharat Bhargava (USA) Fabio Casamatta (Italy) Sandro Corsi (Italy) Art DeGroff (USA) Carlos Gama (Brazil) Jan Ove Gjerde (Norway) Michael Hadingham (South Africa) John Hauer (USA) David Hill (Australia) Michael Hughes (UK) Lawrence Jones (Sweden) Innocent Kamwa (Canada) Prabha Kundur (Canada) Donald Macdonald (UK) Nelson Martins (Brazil) Dagmar Niebur (USA) Damir Novosel (USA) John Paserba (USA) Mania Pavella (Belgium) Steve Rovnyak (USA) Pierangelo Scarpellini (Italy) Jim Smith (USA) Kjetil Uhlen (Norway) Lei Wang (Canada) Contents 1. Introduction and Survey 1.1 Review of power system synchronous stability basics.............................1-2 1.2 Concepts of power system stability controls ............................................1-6 1.3 Types of power system stability control and possibilities for advanced control .................................................................................1-9 1.4 Dynamic security assessment .................................................................1-18 1.5 Intelligent controls..................................................................................1-19 1.6 Effects of industry restructuring on stability controls ............................1-19 1.7 Experience from recent power failures...................................................1-20 1.8 Coordination with other CIGRÉ and industry work...............................1-20 1.9 Summary ................................................................................................1-21 2. Advanced Linear and Nonlinear Control Design 2.1 Nonlinear control......................................................................................2-2 2.2 Linear control techniques .........................................................................2-4 3. State-of-the-Art in Digital Control 3.1 Review of digital control of dynamic systems .........................................3-3 3.2 Basic structure of digital control systems.................................................3-8 3.3 Evolution of excitation control systems through microprocessor technology.....................................................................3-11 3.4 Application of digital control to SVCs...................................................3-14 3.5 Trends in digital control .........................................................................3-19 4. State-of-the-Art in Intelligent Controls 4.1 Fuzzy systems for power system control..................................................4-2 4.2 ANN for power system control ................................................................4-6 4.3 Decision trees for power system control ................................................4-16 5. Integration of Dynamic Security Assessment and Stability Controls 5.1 On-line dynamic stability assessment design ...........................................5-1 5.2 Other integration of DSA and stability controls.......................................5-9 6. Measurement and Communication Technology 6.1 Introduction to transducers .......................................................................6-3 6.2 The signal environment for power system transducers ............................6-6 6.3 Signal processing in power system transducers .....................................6-10 6.4 Criteria and procedures for evaluating transducer performance.............6-13 6.5 Transducer modeling and simulation .....................................................6-14 6.6 Digital transducers and phasor measurements .......................................6-15 .....1 Brazilian north–south interconnection—application of thyristor controlled series compensation (TCSC) to damp interarea oscillation mode .................8-10 8..................................11 The transducers as an intelligent electronic device .............................................3 The impact of IPP thermal generation on system dynamic performance.............................................9-1 9.I-1 J A new look at damping control .................................................. Applications of Advanced Controls 7...............................F-1 G Field evaluations of power system transducers .......4 Other issues related to power system performance in the new utility environment ................5 Active power modulation of generators and energy storage for oscillatory instability control ..................................................................... Conclusions 9...............6-17 Role of communication channels in wide-area control .......................1 Conclusions ....................1 Some examples of new scenarios..................................................7 6................. E-1 F Laboratory evaluations of power system transducers.3 Wide-area stability control ...................................................................................... positive and negative sequence vectors ..........................6.........................8-14 9.............................................................. C-1 D Sideband production in RMS calculations .................................................................................................. B-1 C Space vector..............7-3 7..............................7-1 7.................................. J-1 ...........................................................................H-1 I Performance of BPA analog communication channels ...............8-9 8.........6-22 7............7-7 7..............................................................8 6......................................................................6-19 Future digital communication for stability control.....................2 Analysis and control of Yimin–Fengtun 500-kV TCSC system ................6-22 Optical sensors ..2 Coordinated planning and operation in a competitive environment ..............9-2 Appendices A Adjustable speed hydro generation..............G-1 H Transducer modeling and simulations....................................9 6..............................6-17 Observed performance of digital communications in the BPA phasor measurement network................................8-1 8.................................. Stability Controls with Industry Restructuring 8..........7-4 7...........4 Active load modulation for stability control .........................................7-8 8.............................................................................................................................2 Areas for future work .A-1 B Dynamic performance changes produced by numerical integration algorithms.........D-1 E Basic phasor calculations ..10 6........... C. S. W. P. Jones Chapter 3: M. Taylor G. F. W. H. W. Hauer (lead editor). W. J. W. Pavella. E. Schulz. L. N. J. W. C. L. M. Vaahedi (lead editor). Andersson and A. Taylor (lead editor). R. Schulz. M. M. P. P. F. E. Ihara. J. B. K. Hammad C. D. Hauer.Chief Editor Assistant Editors/Readers Chapter contributors Chapter 1: C. Pal .. Hammad. K. Akke. A. T. Gama. I. E. P. K. Taoka. M. Akke (lead editor). P. Rovnyak. Gomes (lead editor). F. E. W. Jones. Corsi. DeGroff Chapter 2: J. T. R. Corsi Appendix C M. Donalek Appendix B S. Hiyama. Niebur. Pal. S. de Mello. Wang. Vieira F. Taylor Chapter 5: E. D. F. Henderson. C. Taylor Chapter 8: P. Hauer Appendix J M. Paiva Chapter 4: D. R. Taylor (lead editor) Appendix A P. Østrup. Lee. Noroozian Appendices D–I J. Scarpellini. W. Taylor Chapter 6: J. C. L. C. Taylor Chapter 7: M. Uhlen. Kamwa. S. Martins. C. K. Noroozian (lead editor). A. Kundur. G. M. McCalley. Østrup. P. X. Novosel (lead editor). Cheung. K. Bhargava. Taylor Chapter 9: C. Y. T. Smith (lead editor). Christensen further described such concepts in [1-4]. and industry restructuring pose new challenges.Chapter 1 Introduction and Survey Power system synchronous or angle instability phenomenon limits power transfer. There is great opportunity for synergism in these areas. The effect of the faulted line outage on generator acceleration and stability may be greater than that of the short circuit itself. Fault clearing of severe short circuits can be less than three cycles (50 ms for 60 Hz frequency). thyristor exciters. Effective in an engineering sense means “cost-effective. This report provides guidance on advanced methods to improve stability. These concepts were discussed at a panel session on “More Effective Networks” at the 1996 general meeting in Paris. Much can be gained by technology transfer to the electric power industry from disciplines such as automatic control.1-3]. especially where transmission distances are long. The technologies include control theory and applications. and signal processing.1 reviews the basics of power system stability. greater variation in power schedules. the panel session involved eight study committees. Section 1. The goals are new control strategies that are effective and robust.” Control robustness is the capability to function appropriately for a wide range of power system operating and disturbance conditions.1-2. digital and optical transducers. signal processing. An interesting question arose: • What is the value of direct control of voltage phase angle? Equipment such as powerelectronic controlled series compensation and phase-shifting transformers may directly control the phase angle (and indirectly control generator rotor angles). Recent large-scale power failures in North and South America and in other parts of the world have heightened the concerns. Protection or other engineers responsible for implementation of stability controls may not be entirely familiar with control technology or power system stability phenomena. The synchronous stability problem has been fairly well solved by fast fault clearing. Power system engineers responsible for determining stability-related transfer limits and for developing means for extending transfer limits are always acquainted with state-ofthe-art control technology. and telecommunications. This is well recognized and many methods have been developed to improve stability and increase allowable power transfers [1-1. artificial intelligence. requirements for more intensive use of available generation and transmission. microprocessor controllers. power system stabilizers. and a variety of other stability controls such as generator tripping. more onerous load characteristics. This report on advanced angle stability controls provides industry guidance in solving stability problems with new or relatively new technologies. The initial incentives for this report were advances in synchronized voltage phase angle measurements and in high voltage power electronic equipment to directly or indirectly control transmission voltage and generator rotor angles. Nevertheless. power electronics. . Our emphasis is also on large disturbances and nonlinear aspects of stability control. Figure 1-1 shows the basic “swing equation” block diagram relationship for a generator connected to a power system. dt where J is moment of inertia of the generator and prime mover. and 1-5.A more comprehensive review of advanced technology for stability control is. Power generation is largely obtained by synchronous generators. for example references 1-1. The generator speed determines the generator rotor angle changes relative to other generators. Tm is mechanical prime mover torque. δo Tm + Tacc 1 2H α ∫ • dt ∆ω ω o ∫ • dt δ Te Generator Electrical Equations Power System Disturbances Fig. All generators must operate in synchronism during normal and disturbance conditions. Loss of synchronism of a generator or a group of generators with respect to another group of generators is instability that could result in expensive widespread power blackouts. desirable. Block diagram of generator electromechanical dynamics. which may be interconnected over thousands of kilometers in very large power systems. and Te is electrical torque related to generator electric power output. This introductory chapter surveys the field of power system stability controls. The block diagram representing the internal generator dynamics is explained as follows: 1-2 . 1. ω is speed. 1-2. but there is a close relation between voltage magnitude control and angle stability. and the possibilities for advanced angle stability controls that are described in the following chapters. however. describe the basics—which we briefly review here. Our emphasis in this report is on angle stability.1 Review of Power System Synchronous Stability Basics Many publications. 1-1. The techniques described are applicable to practical large-scale power systems. The essence of synchronous stability is balance of individual generator electrical and mechanical torques as described by Newton’s second law applied to rotation: J dω = Tm − Te . The basic relationship between power and torque is P = Tω . Units are MWseconds/MVA (or seconds). and illustrates that the power system is fundamentally a highly nonlinear system for large disturbances. behind a reactance. loads. is proportional to the moment of inertia and is the kinetic energy at rated speed divided by the generator MVA rating. With some approximations adequate for a time of one second or more following a disturbance. The above equation approximates characteristics of a detailed. Figure 1-4 shows the relation between Pe and δ graphically.• The inertia constant. influenced by speed controls (governors) and prime mover and energy supply system dynamics. Since speed changes are quite small. The pre-disturbance operating point is at the intersection of the load or mechanical power characteristic and the electrical power characteristic. its power factor. The transformer and transmission lines are represented by inductive reactances. and line and generator outages. The transmission network is generally represented by algebraic equations. Using the relation S = E ′I * .) For illustration. • Disturbances include short circuits. • δ o is pre-disturbance rotor angle in radians relative to a reference generator. with accelerating torque Tacc equal to Tm . It is. This causes electric power and torque to be zero. X where V is the large system (infinite bus) voltage and X is the total reactance from the generator internal voltage to the infinite bus. (Although generator current is very high during the short circuit. H. Normal stable operation is at δ o . • The power system block comprises the transmission network. • Tm is mechanical torque in per unit. power electronic devices. and active current and active power are close to zero. For example. a simple conceptual transmission model as shown in Fig. the Figure 1-3 block diagram is realized. E ′ . power is considered equal to torque in per unit. A severe disturbance is a three-phase short circuit near the generator. large-scale model. As a first approximation it’s assumed to be constant. • ω o is rated frequency in radians/second (2πfo. It comprises a remote generator connected to a large power system by two parallel transmission lines with an intermediate switching station. Loads and generators are represented by algebraic and differential equations. and other generators/prime movers/energy supply systems with their controls. a small increase in mechanical power input causes an accelerating power ( Pm − Pe ) that increases δ to increase Pe until accelerating power returns to zero at a slightly different 1-3 . The generator representation is a constant voltage. however. 1-2 is used. where fo is rated frequency in Hz). the generator electrical power has the well-known relation: Pe = E ′V sin δ . Remote power plant to large system. For a remote generator connected to a large system the oscillation frequency is 0. accelerating power/torque again becomes positive resulting in a runaway increase of angle and speed. with increase in generator speed and angle. For small disturbances. and thus instability. Short circuit location is shown. If. 1-2. The opposite is true for the unstable operating point at π − δ o : a small increase in mechanical power will cause a runaway increase in angle. If deceleration reverses the angle swing prior to π − δ o′ . If the acceleration relative to other generators is too large. runaway situation with large variations of voltages and currents that will normally cause protective separation of a generator or a group of generators. the increase in the electrical torque (and power) developed as the angle increases will decelerate the generator. If the angle swing is beyond π − δ o′ . Loss of synchronism is an unstable. the above power-angle equation can be linearized ( sin δ ≅ δ in radians for angles under 30°). mechanical and electrical torques are equal and the generator runs at a constant frequency close to 50 or 60 Hz rated frequency. Simplified block diagram of generator electromechanical dynamics.equilibrium point. The block diagram (Figure 1-3) would then represent a second order differential equation oscillator. Following clearing of the short circuit by line removal. The angle δ o is generally less than 45°. During normal operation. δo Dm Pm - + Pe 1 • dt 2H ∫ + ∆ω ω o ∫ • dt δ De + E ′V sin(•) X Fig. stability can be maintained at a new operating point δ o′ (Figure 14). 1-3. a short circuit occurs on a transmission line the electric power output will be momentarily partially blocked from reaching loads and the generator (or group of generators) will accelerate. however. E’ ∠δ V ∠0 ~ Fig.1 Hz. 1-4 .8–1. synchronism will be lost. and other devices. or by decreasing the mechanical power input. 1-4. or by both. Stability controls help maintain stability by decreasing the accelerating area or increasing the decelerating area. Black shading for deceleration energy. P Pre-disturbance electrical power Pm Post-disturbance electrical power Fault on electrical power δo π δ o′ δ (a) ∆ω Unstable δo δ δ o′ Stable (b) Fig. stability can be maintained. (a) Power angle curve and equal area criterion. Dotted trajectory is for unstable case. Light shading for additional acceleration energy because of line outage. This may be achieved during the forward angle swing by increasing the electrical power output. loads. If the decelerating area (energy) above the mechanical power load line is greater the accelerating area below the load line. damping power in phase with speed) that represent oscillation damping mechanisms respectively in the prime-mover and generator. (b) Angle–speed phase plane. Dark shading for acceleration energy during fault. For positive ∆ω the mechanical 1-5 . Figure 1-3 also shows mechanical and electrical damping paths (dashed.Figure 1-4 illustrates the equal area stability criterion for “first swing” stability. reduces the mechanical input torque whereas the electrical damping enhances the electrical output torque.1–0. many oscillation modes are present.8 Hz) associated with interarea oscillations between large portions of a power system are the most difficult to damp. External stability controls may also be added to improve damping. Improperly designed or tuned controls may contribute to 1-6 . For first swing stability. and for development of power system stability controls. Stability problems typically involve disturbances such as short circuits.2 Concepts of Power System Stability Controls Figure 1-5 shows a general structure for analysis of power system stability.damping. These disturbances stimulate power system electromechanical dynamics. The low frequency modes (0. can introduce negative electrical damping at some oscillation frequencies.) For stability. General power system structure showing stability controls [1-8]. Generation or load may be lost. high gain combined with time delays can cause positive feedback and instability. resulting in generation–load imbalance and frequency excursions. notably generator automatic voltage regulators with high gains. including friction and windage losses. the net damping must be positive for both normal conditions and for large disturbances with outages. (In any feedback control system. 1. ∆ gen Power System Disturbances ∆ load structural changes direct detection Power System Dynamics ∆P + - structural changes ∆y System Variables switched gen Feedforward Controls u Feedback Controls switched load response detection Fig. The above analysis can be generalized to large interconnected systems. with subsequent removal of faulted elements. For damping. 1-5. synchronous stability between two critical groups of generators is usually of concern. all of which require positive damping. Controls. with microprocessors facilitating implementation. controls for reactive power compensation such as static var systems. The most important feedback (closed-loop) controls are the generator excitation controls (automatic voltage regulator often including power system stabilizer). In complex power systems. and duplicated or 1-7 . and based on local measurements. Other feedback controls include prime mover controls. Generally speaking. This is analogous to the very effective biological systems that operate on the basis of excitatory stimuli [1-9]. however. feedforward controls can be very powerful. Feedforward controls. Short circuit or outage events can be directly detected to initiate pre-planned actions such as generator or load tripping. and take no action until certain monitored variables are out-of-range. pattern recognition). special stability controls [1-3]. There are. “Response-based” feedforward controls are also possible.e. Feedforward controls have been termed discrete supplementary controls [1-5]. however. continuously active. with rules developed from simulations (i. Although the reliability of special stability controls is often an issue [1-12]. continuously-controlled equipment may cause adverse modal interactions [1-7. and special controls for HVDC links. Duplicated or multiple sensors. Continuous feedback controls are potentially unstable.stability problems. adequate reliability can be obtained by careful design. remedial action schemes. as mentioned. one example is negative damping torques caused by generator automatic voltage regulators. interesting possibilities for very effective discontinuous feedback controls. These “event-based” controls are very effective since rapid control action prevents electromechanical dynamics from becoming stability threatening. Also shown on Figure 1-5 are specialized feedforward (openloop) controls that are a powerful stabilizing force for severe disturbances and for highly stressed operating conditions. Controls are typically required to be as reliable as primary protective relaying. can be discontinuous. Because of power system synchronizing and damping forces (including the feedback controls shown on Figure 1-5). redundant communications. These controls initiate stabilizing actions for arbitrary disturbances that cause significant “swing” of measured variables. Feedback controls. and emergency control systems [1-11]. Discontinuous controls have certain advantages over continuous controls. providing time for linear continuous controls to become effective.. Bang-bang discontinuous control can operate several times to control large amplitude oscillations.8]. special protection systems [1-10]. Modern digital controls. Feedforward controls such as generator or load tripping can ensure a post-disturbance equilibrium with sufficient region of attraction. These controls are usually linear. With fast control action the region of attraction can be small compared to requirements with only feedback controls. stability is maintained for most disturbances and operating conditions. These controls are rule-based. or reactive power compensation switching. thyristor-switched capacitor banks. infrequent misoperation or unnecessary operation of HVDC fast power change. Power electronic phase control or switching using thyristors has been widely used in generator exciters.g. There are recent advances in robust control theory. however. loss of synchronism (instability) occurs on the first swing within about one second. or controlled separation. load tripping. 1-8 . Power systems have many electromechanical oscillation modes. Circuit breaker technology and reliability have improved in recent years [1-14. especially gate-turnoff thyristors. two-cycle opening time. There are tradeoffs between cost and performance.1-15]. unrestricted switching frequency. and are usually sufficiently fast for electromechanical stability (e. Lower frequency interarea modes are the most difficult to stabilize. Bang-bang control (up to perhaps five operations) for interarea oscillations with periods of two seconds or longer is feasible [1-16]. Misoperation of generator tripping (especially of steam-turbine generators). Online simulation using actual operating conditions reduces uncertainty. emphasis should be on knowing uncertainty bounds and on sensitivity analysis using detailed nonlinear. Undesired operation by some feedforward controls are relatively benign. temporary fast valving of fossil units. Mechanical actuators (circuit breakers) are lower cost. and minimal transients and maintenance. Power system electromechanical stability means that synchronous generators and motors must remain in synchronism following disturbances — with positive damping of rotor angle oscillations (“swings”). reactive power compensation switching. instability may occur on the second or subsequent swings because of a combination of insufficient synchronizing and damping torques at synchronous machines. For example. For real nonlinear systems.voting logic are common [1-13]. For very severe disturbances and operating conditions. five-cycle closing time). For less severe disturbances and operating conditions. but with advanced technologies and intelligent controls [1-17]. Advantages of power electronic actuators are very fast control. and each mode can potentially become unstable. fast valving. They have restricted operating frequency and are generally used for feedforward controls. and can be used for control adaptation. HVDC links. and transient excitation boosting may not be very disruptive. Synchronizing and damping torques. Response-based controls are often less expensive than event-based controls because fewer sensors and communication paths are needed. Actuators may be mechanical or power electronic. large-scale simulation. it can approach or even exceed the sophistication of controls of. Newer devices. Effectiveness and robustness. and static var compensators. the sensitivity of controls to different operating conditions and load characteristics should be studied. Controls must be designed to be effective for one or more modes and must not cause adverse interaction for other modes. Mechanical switching has traditionally used simple relays. are disruptive and costly. now have voltage and current ratings sufficient for high power transmission applications (other semiconductor devices with current turnoff capabilities are available at lower power ratings). Actuators. for example.” For example. and controls can be “trigger happy.. especially for linear systems. For example. should be used to the extent possible. Simulations are usually off line. Experience shows that instability incidents are usually not caused by three-phase faults near large generating plants that are typically specified in deterministic reliability criteria. UPFC. Reliability criteria also provide a performance margin to account for the many uncertainties in simulation analysis. Of main concern is multiple related (common-mode) failures involving lines on the same right-of-way or with common terminations. Excessive investment to obtain high performance such as rapid damping of oscillations is not desirable. and differences between the simulated and the actual operating conditions. The purpose of stability controls is to remove stability as a limit on power transfers. infrequent generator tripping may be cost-effective compared to new power electronic actuated equipment. Reliability criteria. for example. and circuit breakers. HVDC transmission equipment.For economy. perhaps supplemented with intelligent controls. Uncertainties can include modeling and data errors.3 Types of Power System Stability Controls and Possibilities for Advanced Controls Stability controls are of many types including: • Generator excitation controls • Prime mover controls including fast valving • Generator tripping • Fast fault clearing • High speed reclosing. existing actuators. SMES. These include generator excitation and prime mover equipment. and single-pole switching • Dynamic braking • Load tripping and modulation • Reactive power compensation switching or modulation (series and shunt) • Current and voltage injections by voltage source inverter devices (STATCOM. The three-phase fault reliability criterion is often considered an umbrella criterion providing a sufficient stability margin for less predictable disturbances involving multiple failures such as single-phase short circuits with “sympathetic” tripping of unfaulted lines. a power margin on allowable transfer (typically 5%). or a voltage dip of no more than 20–30% during swings. battery storage) 1-9 . Reliability criteria margins can be. Online. and are often performed several months before actual operation. Rather they are the result of a combination of unusual failures and unforeseen circumstances. near real-time simulations reduces operating condition uncertainty. 1. Purpose of stability controls. the prime mover and prime mover controls. The control described in reference 1-26 is a feedforward control that injects a decaying pulse into the voltage regulators at a large power plant following direct detection of a large disturbance.1-26]. often retrofitted on existing equipment. In the AEP application at Rockport [1-27]. Several forms of discontinuous control have been applied to keep excitation field voltage near ceiling levels during the first forward interarea swing [1-2. References 1-2 and 1-27 describe investigations and recent implementations of fast valving. and the energy supply system (boiler). however. Excitation control. Recalling the proposed use of angle measurement for stability control. Modern automatic voltage regulators and power system stabilizers are digital. because of the coordination required between characteristics of the electrical power system. Fast valving has been found to be lower cost than tripping of turbo-generators.1-25. Generator excitation controls are a basic stability control. Therefore full effectiveness may not be obtained for interarea stability problems where local measurements are not sufficient.• Fast voltage phase angle control • HVDC link supplementary controls • Adjustable-speed (doubly-fed) generation • Controlled separation and underfrequency load shedding We will summarize these controls. Reference 1-18 describes use of many of these controls in Japan. Digital prime mover controls facilitate addition of special features for stability enhancement. which include additional transmission circuits. the control described in references 1-2 and 1-25 computes change in rotor angle locally from the power system stabilizer (PSS) speed change signal. Thyristor exciters with high ceiling voltage provide powerful and economical means to ensure stability for large disturbances. AEP and several other utilities make continual use of this means of improving rotor angle stability. turbine power can be modulated by prime movers controls to improve damping of interarea oscillations. 1-10 . Digital boiler controls. Chapter 17 of reference 1-2 provides considerable additional information. temporary fast valving has been found to be attractive. and to improve coordination with static var compensators that normally control transmission voltage with small droops. Line drop compensation [1-23–24] is one method to increase the effectiveness (sensitivity) of excitation control. Figure 1-6 shows simulation results using this Transient Excitation Boosting TEB. may improve the feasibility of fast valving. Although not common. facilitating additional capabilities such as adaptive control and special logic [1-19–22]. Excitation control is usually based on local measurements. Fast mechanical power reduction (fast valving) at generators is an effective means of stability improvement. Prime mover control including fast valving. since both the first cost and operating costs of these fast valving schemes are less than the best alternative. Use has been limited. 250 Relative angle - degrees w/o TEB 200 150 w/ TEB 100 50 0 2 4 6 Time - seconds 8 10 Fig. 1-6. Rotor angle swing of Grand Coulee Unit 19 in Pacific Northwest relative to the San Onofre nuclear plant in Southern California. The effect of transient excitation boosting (TEB) at the Grand Coulee Third Power Plant following bipolar outage of the Pacific HVDC Intertie (3100 MW) is shown [1-26]. although few of these applications are documented in the literature. Sustained fast valving (sustained power reduction) may be necessary for a stable post-disturbance equilibrium. AEP routinely reexamines the stability of the Rockport generation–transmission complex and the effectiveness of temporary fast valving. The Rockport Operating Guide is updated to reflect changes in operating conditions, changes in controls or operating practices, and changes in the regional transmission network. Figure 1-7 illustrates the effectiveness of the fast valving. The simulated operating conditions and event include a single prior outage and a single phase fault, unsuccessfully cleared by single-phase switching at +50 milliseconds, with successful backup three phase clearing 0.55 seconds after the fault. The plots are of the consequent changes in speed and rotor angle position. The upper plots of Figure 1-7 are with temporary fast valving, and the lower plots are without fast valving. Generator tripping. Generator tripping is an effective and economic control especially if hydro units are used. Tripping of fossil units, especially gas- or oil-fired units, may be attractive if tripping to house load is possible and reliable. Gas turbine and combinedcycle plants constitute a large percentage of new generation. Occasional tripping of these units is feasible and can become an attractive stability control in the future. Most generator tripping controls are event-based (based on outage of generating plant out-going lines or outage of tie lines). Several advanced response-based generator tripping controls, however, have been implemented. 1-11 Fig. 1-7. Simulation of effect of temporary fast valving at Rockport for prior circuit outage and single-phase fault with unsuccessful single-pole switching. Top plots are with fast valving and bottom plots are without fast valving. The Acceleration Trend Relay (ATR) is implemented at the Colstrip generating plant in eastern Montana [1-28]. The plant consists of two 330 MW units and two 700 MW units. The microprocessor-based controller measures rotor speed and generator power, and computes acceleration and angle. Tripping of 16–100% of plant generation is based on eleven trip algorithms involving acceleration, speed and angle changes. Because of the long distance to Pacific Northwest load centers, the ATR has operated many times, both desirably and undesirably. There are proposals to use voltage angle measurement information (Colstrip 500-kV voltage angle relative to Grand Coulee and other Northwest locations) to adaptively adjust ATR settings, or as additional information for trip 1-12 algorithms. Another possibility is to provide speed or frequency measurements from Grand Coulee and other locations to base algorithms on speed difference rather than only the local Colstrip speed [1-29]. A Tokyo Electric Power Company stabilizing control predicts generator angle changes and decides the minimum number of generators to trip [1-30]. Local generator electric power, voltage and current measurements are used to estimate angles. The control has worked correctly for several actual disturbances. The Tokyo Electric Power Company is also developing an emergency control system which uses a predictive prevention method for step-out of pumped storage generators [131,1-32]. In the new method, the generators in TEPCO’s network which swing against their local pumped storage generators after serious fault are treated as an external power system. The parameters in the external system, such as angle and inertia, are estimated by using local on-line information. The behavior of a local pumped storage generator is predicted based on equations of motion. Control actions (the number of generators to be tripped) are determined based on the prediction. Reference 1-33 describes response-based generator tripping using a phase-plane controller. The controller is based on the apparent resistance/rate of change of apparent resistance (R–Rdot) phase plane, which is closely related to an angle difference/speed difference phase plane between two areas. The primary use of the controller is for controlled separation of the Pacific AC Intertie. Figure 1-8 shows simulation results where 600 MW of generator tripping reduces the likelihood of controlled separation. Fig. 1-8. R–Rdot phase plane for loss of Pacific HVDC Intertie (2000 MW). Solid trajectory is without additional generator tripping. Dashed trajectory is with additional 600 MW of generator tripping initiated by the R–Rdot controller generator trip switching line [1-33]. 1-13 high speed reclosing keeps the maximum number of lines in service. One attractive method not 1-14 . Nevertheless. the four-reactor scheme [1-39. For long lines. Single-pole switching is a practical means to improve stability and reliability in EHV networks where most circuit breakers have independent pole operation [1-36. A hybrid reclosing method used by Bonneville Power Administration employs single-pole tripping. and reclosing from the weak end with hot-line checking prior to reclosing at the generator end. the removed transmission lines or other elements may be the major contributor to generator acceleration. but with three-pole tripping on the backswing followed by rapid three-pole reclosure. High magnitude short circuits may be detected as fast as one-fourth cycle by non-directional overcurrent relays. but special breakers are seldom justified. Reclosing is before the maximum of the first forward angular swing. Typical EHV circuit breakers have two-cycle opening time. but after 30–40 cycle time for arc extinction. Shunt dynamic brakes using mechanical switching have been used infrequently [1-2]. Clearing time of close-in faults can be less than three cycles using conventional protective relays and circuit breakers. high-speed reclosing.1-37]. the threepole tripping ensures secondary arc extinction [1-36]. This is especially true if non-faulted equipment is removed by “sympathetic” relaying. Single-pole switching may necessitate positive sequence filtering in stability control input signals. signal processing and pattern recognition techniques may be developed to detect secondary arc extinction [1-42. no special methods are needed.?????? Fast fault clearing. and can also compound the torsional duty imposed on turbine-generator shafts. For advanced stability control. Solutions include reclosing only for single-phase faults. Unsuccessful high-speed reclosing into a permanent fault can cause instability.1-43]. With such short clearing times. High-speed grounding switches may be used [1-41]. The probability of power failures because of multiple line outages is greatly reduced. Normally the insertion time is fixed. One-cycle breakers have been developed [1-34]. Communication signals from the weak end indicating successful reclosing can also be used to enable reclosing at the generator end [1-38]. and single-pole switching. High-speed reclosing is effective when unfaulted lines trip because of relay misoperations. High-speed three-pole reclosing is an effective method of improving stability and reliability. fast reclosing provides “defense-in-depth” for frequently occurring singlephase temporary faults and false operation of protective relays. High-speed reclosing or single-pole switching may not allow increased power transfers because deterministic reliability criteria generally specifies permanent faults. Ultra-high-speed traveling wave relays are also available [1-35]. For short lines. Dynamic braking. and considering that most EHV faults are single-phase.1-40] is most commonly used. During a lightning storm. Several methods are used to ensure secondary arc extinction. Reclosing into a fault is avoided and single-pole reclosing success is improved. is a better solution. and the development of fast communications and actuators. or air conditioners. Thyristor switching or phase control minimizes generator torsional duty [1-44]. high-speed series capacitor switching has been used effectively on the North American Pacific AC intertie for over 25 years [1-52]. Mechanical switching has the advantage of lower cost. This is described in Chapter 7. For continuous modulation. Although unlikely because of economics. Braking automatically results for ground faults — which are most common. Interruptible industrial load is commonly used. This is less disruptive and the consumer may not even notice brief interruptions. The feasibility of this control depends on implementation of direct load control as part of demand side management. The operating times of circuit breakers are usually adequate. Here the communication and actuator speeds are generally not as critical. Load tripping is also used for voltage stability. For switched compensation. with series compensation generally being the most cost effective [1-86]. In addition to previously mentioned advantages. operators bypass the series capacitors some minutes after the 1-15 . thyristor phase control of a reactor (TCR) is used. Clearly load tripping or modulation of small loads will depend on the economics. power electronic control has advantages in subsynchronous resonance performance [1-51]. appliances such as heaters could be designed to provide frequency sensitivity by local measurements. which helps ensure a post-disturbance equilibrium. reference 1-46 describes tripping of up to 3000 MW of industrial load following outages during power import conditions. Thyristor switching of dynamic brakes has been proposed. Load tripping and modulation. Controlled series or shunt compensation improves stability. The main application is for full or partial outages of the parallel Pacific HVDC intertie (eventdriven control using transfer trip over microwave radio). Reactive power compensation switching or modulation. and can be a subsynchronous resonance countermeasure [1-45]. Load tripping is similar in concept to generator tripping but is at the receiving end to reduce deceleration of receiving-end generation. For example. It’s also possible to modulate loads such as heaters to damp oscillations [1-47–50].requiring switching is neutral-to-ground resistors in generator step-up transformers. Mechanical switching is generally single insertion of compensation for synchronizing support. and on the installation of high-speed communication links to consumers with high-speed actuators at load devices. For synchronizing support. Rather than tripping large blocks of industrial load. either mechanical or power electronic switches may be used. especially for interarea oscillations. Series capacitors are inserted by circuit breaker opening. it may be possible to trip low priority commercial and residential load such as space and water heaters. Often generator tripping. As described in Chapter 7. thyristor-controlled series compensation was chosen for the 1020 km. Thyristor-controlled series compensation (TCSC) allows significant time-current dependent increase in series capacitive reactance over the nominal reactance. and other nonlinear and adaptive strategies. Minnesota USA includes two 300 MVAr 500-kV shunt capacitor banks [1-59]. Also described in Chapter 7 is a TCSC application in China for integration of a remote power plant using two parallel 500-kV transmission lines (1300 km).1-61] showed line current magnitude to be the most effective input signal. Current injection by voltage source converters. One study [1-1. static 1-16 .1-53. With appropriate controls.1-56]. four 200 MVAr shunt banks are switched for HVDC and 500-kV ac line outages [1-16]. For series or shunt power electronic devices. synchronizing versus damping control. Generally it’s cost-effective to augment power electronic controlled compensation with fixed or mechanically-switched compensation. Static var compensators are applied along interconnections to improve synchronizing and damping support. Advanced power electronic controlled equipment employ gate turn-off thyristors or other devices with current turnoff capability. this increase in reactance can be a powerful stabilizing force [1-55.1-54]. Thyristor-based series compensation switching or modulation has been developed with several installations in service or planned [1-10. Voltage support at intermediate points allow operation at angles above 90°. Transient stability simulations indicate that 25% thyristor controlled compensation is more effective than 45% fixed compensation. but nowadays are usually not competitive with power electronic equipment. High speed mechanical switching of shunt banks as part of a static var system is common. and new response-based controls are being investigated. Reactive power injection devices include static compensator (STATCOM). response-based controls based on voltage are installed. For synchronizing support. Available SVCs in load areas may be used to indirectly modulate load to provide synchronizing or damping forces. SVCs are modulated to improve oscillation damping. Synchronous condensers can provide similar benefits. Reference 1-60 provides an example using five SVCs with only voltage control to improve stability for a proposed interconnection of the Scandinavian (Nordel) and main European (UCPTE) power systems. For example. Gain supervision and optimization adaptive control is common. high speed switching of shunt capacitor banks is also effective.event. Response-based control using an impedance relay was also used for some years. Again on the Pacific AC intertie. 500-kV intertie between the Brazilian north/northeast networks and the south/southeast networks [1-57]. Several advanced TCSC control techniques are promising [1-58]. the Forbes static var system near Duluth. Digital controls allow many new control strategies. control mode selection allows bang-bang control. and unified power flow controller (UPFC). In contrast to the above power electronic devices. 1-69. Bonneville Power Administration developed a novel method for transient stability by high speed 120° phase rotation of transmission lines between networks losing synchronism [1-54].1-7].synchronous series compensator (SSSC). this can be visualized as keeping high synchronizing coefficient (slope of power– angle curve) during disturbances. and one installation (not a transient stability application) is in service [1-53]. Reference 1-1 describes use of these devices for oscillation damping. The unified power flow controller incorporates GTO-thyristor phase shifting and series compensation control. On a power–angle curve. This provides powerful stability control. Figure 1-9 shows commissioning test results. high cost has presumably prevented installations. One concept employs power electronic series or phase shifting equipment to directly control angles across an interconnection within a small range [1-64]. Voltage phase angles and thereby rotor angles can be rapidly controlled by power electronic controlled series compensation (discussed above) or phase shifting transformers. For long distance HVDC links within a synchronous network. References 1-1. SMES can be of smaller MVA size and possibly lower cost than a STATCOM. For angle stability control. Although one type of thyristor-controlled phase shifting transformer was developed almost twenty years ago [1-62]. injection of real power is more effective than reactive power. Fast voltage phase angle control. 1-68. HVDC link supplementary controls. it’s often effective for voltage source inverter control to also coordinate mechanical switching. the available HVDC converters provide the actuators so that stability control is inexpensive. Control robustness. For transient stability improvement. HVDC dc links are installed for power transfer reasons. Voltage source inverters may also be used for real power series or shunt injection. It has not been implemented. SMES or battery storage provides both active and reactive power control. however. HVDC modulation can provide powerful stabilization. with active and reactive power injections at each converter. implemented in 1976. The Pacific HVDC Intertie modulation control. is unique in that a remote input signal from the parallel Pacific AC Intertie was used. 1-65–67 and 1-87 describe HVDC link stability controls. SMES may be less location dependent than a STATCOM. Adjustable-speed (doubly-fed) generation. Superconducting magnetic energy storage (SMES) or battery storage is the most common. Reference 1-63 describes an application study. and Appendix A describe stability benefits of adjustable speed synchronous machines that have been 1-17 . As with conventional thyristor-based equipment. References 1-1. is a concern [1-1. This technique is very powerful (perhaps too powerful!) and raises reliability and robustness issues especially in the usual case where several lines form the interconnection. Undesirable generation tripping during voltage and frequency swings must be minimized through adequate control and protection design and settings. on-line dynamic (or transient) stability/security assessment software has been developed. maintaining synchronism may not be possible or cost-effective. 1-70. References 1-33. Recent proposals advocate use of voltage phase angle measurements for controlled separation. are usually based on off-line simulation (time and frequency domain). Simulation of potential disturbances is then based on actual operating conditions. Underfrequency load shedding may be required in islands that were importing power. With today’s computer capabilities. developed for pumped storage applications. Security assessment is made efficient by techniques such as fast screening and contingency 1-18 . however. System response to Pacific AC Intertie series capacitor bypass with and without dc modulation [1-66]. Controls must then operate appropriately for a variety of operating conditions and disturbances. State estimation and on-line power flow monitoring provide the base operating conditions. 1-9. along with transfer limits.Fig. the cost may be low enough to be competitive with alternatives. Controlled separation (islanding) based on out-of-step detection or parallel path outages mitigates the effects of instability. Dynamic security assessment is presently used to determine arming levels for generator tripping controls [172.1-73]. Recently. The generation/load imbalances in the islands that are formed should be small enough that the islands stabilize. hundreds or thousands of large-scale simulations may be run each day to provide an organized database of system stability properties. Control of excitation frequency enables direct control of rotor angle. Since the frequency converter only supplies power to the rotor.4 Dynamic Security Assessment Control design and settings. reducing uncertainty of the control environment. and 1-71 describe advanced controlled separation schemes. Controlled separation and underfrequency load shedding. 1. For very severe disturbances and failures. Reference 1-88 describes doubly-fed turbo-generators. and on field tests. control tends to be faster and both final states are zero (using angle. transmission and distribution makes necessary power system engineering more difficult. fast valving. frequently changing power transfer patterns cause new stability problems. 1-19 . rather than by new transmission lines [1-75]. The controllers are located at generator plants. etc. and smart termination of strongly stable or unstable cases. We further describe intelligent controls in Chapter 4. Pattern recognition may be considered data compression of security assessment results. New power industry standards along with ancillary services mechanisms are being developed.selection. The SONFC could be expanded to incorporate remote measurements. and power system stabilizers may be ancillary services. Dynamic security assessment provides the database for pattern recognition techniques. the techniques developed for power system stabilizers. common initiation may be used for the different contingencies In the future. Compared to the angle–speed phase plane. for example. Industry restructuring requiring near real-time power transfer capability determination may accelerate the implementation of dynamic security assessment. New. Fuzzy logic may be used for rule-based control. facilitating advanced stability controls. 1. Another possibility is stability control based on neural network or decision tree pattern recognition.5 Intelligent Controls Mention has already been made of rule-based controls and pattern recognition based controls. The SONFC is suitable for generator tripping. dynamic security assessment may be used for control adaptation to current operating conditions.6 Effect of Industry Restructuring on Stability Controls Industry restructuring will have many impacts on power system stability. Dynamic security assessment simulations could be used for updating/retraining of the neural network fuzzy controller. Controls such as generator or load tripping. the post-disturbance equilibrium angle is not known in advance). Parallel computation is straightforward using multiple workstations for different simulation cases. Most stability and transfer capability problems must be solved by new controls and new substation equipment. independent grid operators or security coordination centers may facilitate dynamic security assessment and centralized stability controls. HVDC control. In large interconnections. As a possibility. series or shunt capacitor switching. We further describe on-line security assessment in Chapter 5. Different ownership of generation. 1. higher than standard exciter ceilings. Acceleration and speed can be easily measured/computed using. reference 1-74 describes a sophisticated self-organizing neural fuzzy controller (SONFC) based on the speed–acceleration phase plane. • Power system stabilizer design and tuning. the need for “defense-in-depth” or “multiple lines of defense.02.20 Advanced Power System Controls Using Intelligent Systems. Figure 1-10 shows the development of the August 10 breakup Other blackouts have occurred recently in the North American Upper Midwest [1-80]. Impact of Interactions among Power System Controls.We further describe the effect of industry restructuring on stability controls in Chapter 8.19.g. best practice local stability controls (e. CIGRÉ TF 38. and underfrequency or undervoltage load shedding. providing comprehensive description of many aspects of stability... controlled separation. and in Brazil. and CIGRÉ TF 38. are currently underway.” and the need for advanced stability controls. 1.16. • Power system modeling and data validation for control design. 1-81. 1996 power failures [1-76–80] in western North America showed need for improvements and innovations in stability control areas such as: • Fast insertion of reactive power compensation for voltage support.02. reduced power transfers during storm conditions). and may include generator/load tripping. and fast generator tripping using response-based controls. and 1-82 document recent CIGRÉ and IEEE work related to angle stability control. System Protection in the Power System: modeling and analysis. Defense-in-depth/multiple line of defense for system reliability includes risk management in system operation (e. In Brazil. 1-20 . The July 2. Our intent is to complement rather than duplicate other industry work.7 Experience from Recent Power Failures Recent cascading power outages demonstrated the impact of control and protection failures. fast and reliable protective relaying. thyristor exciters with PSS). 1. These works are valuable. 1996 and August 10.g. • HVDC. high-speed three or single pole reclosing. TCSC.02. CIGRÉ TF 38. • Control adaptation to actual operating conditions. The final lines of defense mitigate the effects of extreme disturbances. • Controlled separation. new emergency controls for generator/load tripping and controlled separation are being added. and SVC control for stability.8 Coordination with other CIGRÉ and Industry Work References 1-1. Vol. 794-800. 1. This chapter provides a broad survey of available stability control techniques with emphasis on new and emerging technology. uncontrolled separations broke the system into four islands with loss of 30. December 1996. Analysis and Control of Power System Oscillations.008 Malin -Round Mountain #1 MW caseID=Aug10E5loadPF casetime =04/16/98_14:41:48 15:48:51 Out-of-Step separation 1500 15:42:03 Keeler-Allston line trips 15:47:36 Ross-Lexington line trips/ McNary generation drops off 1400 1300 0. 1-21 . 1996 breakup. Power System Stability and Control. pp. August 1996. 3. Power flow on Oregon–California 500-kV line during initial portion of August 10. No.489 MW of load. Some methods have been used effectively for many years.01. “Annotated Bibliography on Power System Stability Controls: 1986-1994.9 Summary Power system angle stability can be improved by a wide variety of controls. References 1-1 CIGRÉ TF 38. both at generating plants and in transmission networks.264 Hz. 1-2 P. 1-3 IEEE Special Stability Controls Working Group. 2. New control techniques and actuating equipment are promising. 1-10.46% damping 1200 Reference time = 15:35:30 PDT [bitmap version 1100 200 300 400 500 Time in Seconds 600 700 800 Fig. The following chapters provide in-depth evaluation of the many issues in the selection and design of stability controls. 1994.252 Hz 0. 11. Brochure 111.” IEEE Transactions on Power Systems. McGraw-Hill. Kundur.07. Following separation of the Pacific AC intertie.276 Hz 0. 163. Kim. 44.” CIGRÉ. “An Outline of Power System Technologies in Japan.” Electra. B. I.” 32nd IEEE Conference on Decision and Control. 1993. S. and D. H. 11. F. 33rd Power Instrumentation Symposium. 14. Djakov. May 21–23. Vol. NERC Planning Standards. M. F. Berdnikov. G. October 1998. 1-12 IEEE/ CIGRÉ Committee Report (P. August 1997. 1-17 K. LeReverend). 1998. E. Doel. and S. paper 13-114. Studt. December 15–17.” Electric Power Systems Research. Brunke. 7. “Design and Installation of 500-kV Back-toBack Shunt Capacitor Banks. “Robust Damping Controls for Large Power Systems. 1990. and V. 1998. 1-7 J.” CIGRÉ. 1-5 IEEE Discrete Supplementary Control Task Force. No. “Stability and Control of HVDC and AC Transmission in Parallel.” IEEE Transactions on Power Apparatus and Systems. pp. 1994. 1-9 T. pp. H. 3. Dodge. January/February 1978. M. Texas. pp. W. “Robustness Issues in Stability Control of Large Electric Power Systems.1. Torizuka and H. 1166–1179. 173. F. A. Z. V. Esztergalyos.” IEEE Transactions on Power Delivery. Gluskin. Portnoi. A. B. 1998. “Development of an Intelligent Autoreclosing Concept using Neuro-Fuzzy Techniques — An Optimal Controlled Switching for System Operation. H. “A Description of Discrete Supplementary Controls for Stability. Tanaka. “Computer Scientists Search for Ties to Biological Intelligence. pp. and J. Hammad. 4. September 1997 (available at www.com). paper 90-1323.” ISA Instrumentation in Power Industry. 149–165. 1-10 North American Electric Reliability Council. No. “Industry Experience with Special Protection Schemes. 1–5. Stroev. D. B. pp. August 1996. W.” IEEE Transactions on Power Delivery. “New Control Strategies for Utilizing Power System Network More Effectively. 5–16.1-4 J. Vol. 1-11 A. Anderson and B. V. 1-14 CIGRÉ Task Force 13. Furumasu and R. 1-15 J. I. M. December 1995. Hauer.” Electra. Vol. No. A. Cho. October 1999.” CIGRÉ. PAS-97. Park. Johnson. Hauer.00. V. J. 1-16 B. Vol. 1-22 . Bondarenko. pp. 1-8 J. 1-18 T. “Benefits of Microprocessor-Based Circuit Breaker Control. April 1992. A. K. 1-6 A. Khan. pp. 539–545. C. Kovalev. Christensen. No. paper 39-109. San Antonio. J.” R&D Magazine. “Controlled Switching — A State-of-the-Art Survey. Vol. 1-13 D. “Power System Stability Control Using Fault Tolerant Technology. 2. paper 23/13-10.nerc. No.” IEEE Transactions on Power Systems. Shim. No. F. “The Operation of Integrated Power Systems Close to Operating Limits with the Help of Emergency Control Systems. Semenov. Smith. 65–97. 77–78. Hasibar. 33. January 1989.” IEEE Control Systems Magazine. Woods. E. Matthews. Rubenstein and W. No. 292–299. Vol. 3. Taylor. Stigers. 3. Salyzyn. February 1998. 1-20 K. R. paper 38-01. Vol. 1-24 A. 1-25 D. C. and R. M. Vol. “Digital Excitation Technology — A Review of Features. R. Takeuchi. Vol. 10. Raffaelli. No. T. R. September 1998. Nakamura. “The Acceleration Trend Relay for Generator Stabilization at Colstrip. pp.” IEEE Transactions on Energy Conversion. 11. 1-21 L. No. 1-23 A. and A. 1986. Mechenbier. No. 1-28 C. 3. Yanagihashi. No. N. L. 1-30 K. “Dynamic Voltage Control by Remote Voltage Regulation for Pumped Storage Plants. 13. “Experience with Digital Power System Stabilizers at Steam and Hydro Generating Plants. A. 1. No. Rosa. 3. 1291–1298. “Advanced Excitation Controls for Power System Stability Enhancement. February 1996. Baroffio. G. December 1957. Lee and P.” IEEE Transactions on Power Systems.” IEEE Transactions on Power Systems. Tsukita. A. Dehdashti. 1-22 V. Esztergalyos. 2. pp. Vol. August 1993. G. and E. pp. “Stabilizing Control System Preventing Loss of Synchronism from Extension and Its Actual Operating Experience. Ottaviani.” IEEE Transactions on Energy Conversion.” IEEE Transactions on Power Systems. August 1995. 1074–1081. 12. M.” CIGRÉ. Functions and Benefits. F. pp. J. September 1997.” IEEE Transactions on Power Systems. 1-26 C. “Feasibility Study of Using Synchronized Phasor Measurements for Generator Dropping Controls in the Colstrip System. Stigers. Vol. “Field Experience with Momentary Fast Turbine Valving and Other Special Stability Controls Employed at AEP’s Rockport Plant. 13. “Utility Experience with Digital Excitation Systems. pp. Sato. 3. 3. July 1997. K. Togno. J. 13.” AIEE Transactions.” IEEE Transactions on Power Systems. Corsi. pp. J. Kundur. 165–170. E. 961–970. S. Vol. “Control of Reactive KVA with Modern Amplidyne Voltage Regulators. C. Walkley. Greenwood-Madsen. 1-27 N. 3. No. 1. and C. Smith. Vol. 8. Gerube. 12. Arcidiancone. and C. Setterstrom. W. S. S. August 1998. Hajagos. 8. W. J. August 1988. Peng. Vol. 3. T. Luini.” IEEE Transactions on Power Delivery. Vol. “The ENEL’s Experience on the Evolution of Excitation Control Systems through Microprocessor Technology. and M.” IEEE Transactions on Power Systems. S. No. C. No. pp. J. pp. 1606–1613. Matsuzawa. Nettleton. D. “Transient Excitation Boosting at Grand Coulee Third Power Plant. Bhatt. and G. No. B. 1-29 D. 155–161. S.1-19 IEEE Digital Excitation Applications Task Force. Kosterev. 755– 762. 1-23 . pp. and Z. 1188–1192.” IEEE Transactions on Energy Conversion. June 1993. Bollinger. September/October 1974. Lieberman. M. Mittelstadt. O. H. “Relay-to-Relay Digital Logic Communication for Line Protection. V. Kojima. 285–290. Vol. G. PWRD-1. Y. 1978. 1-35 J. C. Spokane. PAS-83. M. V. Belotelov. Dyakov. “Design and Implementation of an Adaptive Single Pole Autoreclosure Technique for Transmission Lines using Artificial Neural Networks. Oshida. April 1996.” IEEE Transactions on Power Apparatus and Systems. D. Johns. paper 34-203. “Single-Pole Switching for Stability and Reliability. M.9. W. H. T. Ilynichnin. PAS-100. Kimbark. and T. 118-B. T. Kojima. Strelkov. G. Vol. and Control. “On-line Modeling for Emergency Control Systems. 1-37 A. Legate. Skreiner. M. and A. Syoji. “Development of Predictive Prevention Method for Mid-term Stability Problem using Only Local Information. 25–36. 1240–1251. Oshida. Kowada. No. Shelton. and C. No. H. 1962. 1-33 J. K. and S.” CIGRÉ. PAS-93. S. Fitton. C. G. M. Vol. H. F. H. pp. 627–632. 748–756. and T. Washington USA. pp. Behrendt. Vol. “Experience with the R-Rdot Outof-Step Relay. Dewey. Leviush. 1-41 R. Monitoring. 1998. 1997.” Proceedings of the 23rd Annual Western Protective Relay Conference. Hasibar. T. March 1964. W. Peterson. pp. 1-24 . pp. Knutsen. No. Bennett.” CIGRÉ. and V. PWRS-1. Barkan. A. K. “Single-Phase Switching of Transmission lines Using Reactors for Extinction of the Secondary Arc. and W. P. K. Imai.” Transactions of IEE Japan. 2. 1-38 K. Esztergalyos. No. Y. T. May 1986. Dunn. A. L. “Suppression of Ground-Fault Arcs on Single-Pole Switched Lines by Shunt Reactors. A. M. April 1981. Berglund. pp.” IEEE Transactions on Power Delivery. “Application of Automatic Reclosing in High Voltage Networks of the UPG of Russia Under New Conditions. R.” IEEE Transactions on Power Systems. Goda. “The Development and Operation of an Ultra High Speed Relaying System for EHV Transmission Lines. 4. Vol. October 1996. 3. “The Application of High-Speed Grounding Switches for Single-Pole Reclosing on 500-kV Power Systems. 1512–1515. “One-Cycle Fault Interruption at 500 kV: System Benefits and Breaker Designs. April 1986. IEEE Transactions on Power Apparatus and Systems. Yanagihashi.1-31 Y. Brunke. 11. paper 310. paper 34-04. 2. Taylor. Vol. 1-36 IEEE Committee Report. W. Goda. R. Haner. 1-40 E. W. C. Chamia. G.” CIGRÉ. 35–39. A. I. Aggarwal. J. No. 1-42 D. Yee.” IEEE Transactions on Power Apparatus and Systems. pp. 1-34 R. M. pp. Laughlin. A.” IEEE Transactions on Power Delivery. Fokin. 1-39 N. 1-32 S. Taoka. and K. 1998. Vol.” IFAC/CIGRE Symposium on Control of Power Systems and Power Plant. “Transient Stability Enhancement by 120-Degree Phase Rotation. “Distributed-FACTS: End-Use Load Control for Power System Dynamic Stability Enhancement.. “Northwest Power Pool Transient Stability and Load Shedding Controls for Generation–Load Imbalances. Nassief. C. 1-54 R. November 1998. Johnson. 12. Donnelly. and J. Taylor. pp. R. W. D. Vol. D. 1996. R. 1-51 A. J. G. pp. October 1995. PAS-85. La Jolla. J.” IEEE Transactions on Power Systems. Kamwa. A. No. 67– 73. Hammad and M. Kreipe. 3486– 3495. Trudel. R. PAS-100. 4. Vol. 1-25 . “Damping of Electro-Mechanical Oscillations in a Multimachine System by Direct Load Control. Cresap. R. M. 1. PAS-103. November 1997. Nelson. 1. 2. Bian. Samuelsson and B. February 1993. 8 No. D. K. Povh. and G. 1604–1609. Asber. Dagle. 1-53 M. F. Rahman. July 1981. paper 14-306. Brush. 1-47 O. J. and G.” IEEE Transactions on Power Delivery. California. 2. February 1966. 1-44 W. L. B. The Future of Power Delivery in the 21st Century. 18–20 November 1997. Trudel. P. 4.” EPRI Conference. pp. “Long Distance Transmission with Parallel AC/DC Link From Cahora Bassa (Mozambique) to South Africa and Zimbabwe. 1-50 I. O’Keefe. Vol. pp.” IEEE Transactions on Power Systems. Gingras. Hauer. F. W.” IEEE Transactions on Power Systems.” IEEE Transactions on Power Systems. No. R. J. pp. Jacobson.” CIGRÉ. Adapa. Guedes. W. and R. V. 1-46 C. Asber. J. May 1999. A. 1-49 J. Vol. 10. Terzija. “UPFC Application on the AEP System: Planning Considerations. “Control of a Dynamic Brake to Reduce Turbine-Generator Shaft Transient Torques.1-43 M.” IEEE Transactions on Power Apparatus and Systems. No. and M. W. 4. pp.” IEEE Transactions on Power Systems. Grondin. PAS-100. No. No. Vol. R. M. pp. February 1981. Kimbark. November 1997. Marshall. 1793–1798. El-Sadek. Bayer. Gutman. Taylor. J. L. Cresap. No. Vol. “Large-Scale Active-Load Modulation for Angle Stability Improvement.” IEEE Transactions on Power Apparatus and Systems. “Active Power Stabilizers for Multimachine Power Systems: Challenges and Prospects. Ahmed. “Improvement of System Stability by Switched Series Capacitors. 1-45 M. D. R. Gingras. 1-48 I. 14. Vol. No. pp.” IEEE Transactions on Power Apparatus and Systems. P. No. 13. Kamwa. 1695–1701. 12. Eliasson. 7. pp. R. “Application of a Thyristor Controlled VAr Compensator for Damping Sub-synchronous Oscillations in Power Systems. M. January 1984. “A New Approach to the Arcing Faults Detection for Fast Autoreclosure in Transmission Systems. 4. Smith. 582–590. Vol. J. Vol. R. Vol. J. 1-52 E. pp 198–212. Grondin. and D. Djuric and V.” IEEE Trans. 180–188. Habur. 745–753. and R. K. 1352–1358. 1. W. Chow. W. R.” IEEE Computer Applications in Power. 1-26 . “HVDC Controls for System Dynamic Performance. No. Christl. No. Scott. W. B. July 1996. paper 14-101. Taylor. 1987. L. Vol. Eiras. Hedin. Cresap.” CIGRÉ. 36–53. 1-57 C. Vol. C. P. Güth. paper 14-104. Sweezy. M. and R. Furumasu. 4147. pp. Johnson.” IEEE Transactions on Power Systems. J. pp. IEE Proc. 132.” IEEE Transactions on Power Delivery. paper 14/37/38-05. S. C. M.—Gener. D. pp.. Gama. 157–168. N. “Brazilian North–South Interconnection — Application of Thyristor Controlled Series Compensation (TCSC) to Damp InterArea Oscillation Mode. Mittelstadt. 145. pp. Piwko. Larsen and J. Dellwo. Sadek. 1998. Østrup. 743–752. L. H.” IEEE Transactions on Power Apparatus and Systems. Lützelberger. 1507–1514. L. Cresap. 1992. 1-63 Y. pp. G. and C. No. Vol.” CIGRE 14-05. 11.” Brown Boveri Review. 1-67 R. Leoni. Mazur. R.” Application of Static Var Systems for System Dynamic Performance. July/August 1978. Christensen and T. 1978. A. Vol. March 1982. W. “Analysis and Control of Yimin–Fentun 500 kV TCSC System. Installation. W.” Electra. A. 73–78. W. 1-65 IEEE Committee Report. 1-58 X.J. Giroux. May 1991. Litzenberger. 3. R. C. and J. pp. J. 1. Krause. “Damping of Pacific AC Intertie Oscillations via Modulation of the Parallel Pacific HVDC Intertie. Stemmler and G. Scott. Fang and D. “Simulator and Field Testing of Forbes SVC. 1-59 G. Taylor. 2. P. D. pp. “SVC Control Design Concepts for System Dynamic Performance. 1-56 R. Torgerson. Damsky. “Keeping an Eye on Power System Dynamics. Transm. 3. No. Sybille. Macdonald. Mittelstadt. 1-60 J. and J. and C. R. “The Slatt Thyristor-Controlled Series Capacitor Project — Design. H. B. Trudnowski.” Electric Power Systems Research 46 (1998). McKenna. Zhou et al.. Ricardo. D. W. Tenório. V.” CIGRÉ. Fraga. CIGRÉ. S. F. No. 1-61 E. “Operating Experience with Modulation of the Pacific HVDC Intertie. J. Distrib. Cavalcanti. PAS-98. “Advanced Series Compensation (ASC) with Thyristor Controlled Impedance. Eden. Rogers. 26–30. January 1998. “Dynamic Quadrature Booster as an Aid to System Stability. “The Thyristor-Controlled Static Phase Shifter—A New Tool for Power Flow Control in AC Transmission Systems. Hauer. A. D. Gribel. E. IEEE Special Publication 87TH1087-5-PWR. P. 69. 6. Mittelstadt. A. Ping. Vol. pp. J. 1053–1059. 1-62 H. January 1997. 10. A. Montoya. 1994. N. L. 1-64 J. No. Wegner. 1-66 R. K. Commissioning and System Testing.1-55 N. and G. and D. Vol. “ Example of Reinforcing an AC Interconnection Between Two Large Networks by Use of SVC. October 1990. Chang and M. 241–253. No. 1-79 R. K. Lochmatov. Omata. “Neural Network-Based Self-Organizing Fuzzy Controller for Transient Stability of Multimachine Power Systems. Milano. August 1996. No. K. Phadke. 10. Yanagihashi. Athay. Vol. 2. 1-71 V. M. Vol. 1-72 Y. 1-76 WSCC reports on July 2. A. 61–71. S. Dubois. Michel. 11 No. 1998. 26–30. paper 11/37-04. W. Yamaura. 5. 12. Taylor and D. E. K. Vol. No. 1-77 C. A. 1-69 Y. 5. Schurz. and Y. 1-74 H. Mortensen.” IEEE Transactions on Power Delivery. and G. Deuse. Corns. 1996 outages — available at www. 1. Bunch and D. H. M. “Design and Implementation of AC Voltage Dependent Current Order Limiter at Pacific HVDC Intertie. 1-27 . J. 1-78 D. 1281–1284.1-68 W. 1-70 Y. “An Adaptive Out of Step Relay. 10. N. A. 3. “A Predictive Out-of-Step Protection System Based on Observation of the Phase Difference Between Substations. Mamilo-Niants. 1. “Transmission System Open Access Versus System Reliability: A Case History—the MAPP Disturbance of June 25. August 1999. and W. Centeno. 1996 WSCC System Outage. 1-73 H. N. L.” IEEE Transactions on Power Systems. A. H.” IEEE Transactions on Energy Conversion. C. Kitayama. Analysis.2171–2176. November 1990. W. K. J. pp. Mittelstadt. pp. No. “Model Validation for the August 10. K. J. 100. W. Vol.com. pp. Chang. K. Vaahedi. and F. February 1995. Taylor. Y. “B. Ito. N. M. Kosterev. Ohura. No. Vinitsky. pp. Omata. 1463–1472. Ota. and H. Shakarian.” IEEE Transactions on Power Systems. See also discussion of Joint Session 11/37 in Electra. Kokai. H. 1-75 K. C. Suzuki. Vol.” IEEE Transactions on Power Apparatus and Systems. “An Adjustable Speed Synchronous Machine For Hydroelectric Power Applications. 14.wscc. “Development of Transient Stability Control System (TSC System) Based on OnLine Stability Calculation. 967–979. Fukushima. Garrett. Erickson. 1998. Wang. B. Watanabe. pp. and K. C.” to appear in IEEE Transactions on Power Systems. Schleif. June 1995. pp.” IEEE Computer Applications in Power.” IEEE Transactions on Power Systems.” CIGRÉ. R. G. February 1999. pp. Kauffmann. “Recording and Analyzing the July 2 Cascading Outage. No. Mansour. No. Mitamura. Y. May 1981. B. 1. T. and J. Morita. Kosterev. Mangel. Vol. Vol. 339–347. 10. J. “Role of Two-Axis Excitation Generators in Power Systems. Gaudi. 182. 4. Demaree. Edris. Y. January 1997.” Proceedings of IEEE/PES 1999 Winter Meeting. B. pp. Benton. Gish. No. and Post-processing. 1996 and August 10. January 1997. 3. A. M. C. pp. Karoui. Vol. N.” IEEE Transactions on Power Delivery. 1695–1704. T. Hydro’s On-line Transient Stability Assessment (TSA) Model Development. Cheung. Nakamura. Kuwabara. Vinitsky.” IEEE Transactions on Power Delivery. “Utilising HVDC to Damp Power Oscillations. 1-83 T. pp. Vol. 1-82 IEEE FACTS Working Group. paper 11/37-01. 2. June 1996. Andersson. Schäfer. 40–45.06 Load Flow Control in High Voltage Systems Using FACTS Controllers. 376–384. J. Shakaryan. and K. 1-85 D. No. “Adjustable Speed Asynchronous Machine in Hydro Power Plants and its Advantages for Electric Grid Stability. No.” IEEE Spectrum. Z. Lochmatov. Vol. 1-28 . Hvoschinskaya. 8.1-80 C. 1998. 1-86 M. April 1993.” IEEE Transactions on Power Delivery. No. and K. 2. Taylor.01. June 1999. February 1999. October 1995. Kauffmann. No. Kita. “Design and Dynamic Response Characteristics of 400 MW Adjustable Speed Pumped Storage Unit for Ohkawachi Power Station. A. Karoui. B. 620–627. 1998. 1-87 T. Simond. pp. Andersson. J. Vol. October 1994. A.” CIGRÉ. 2. J. H. Willy. 2046–2054. No. See also discussion of Joint Session 11/37 in Electra. pp. 1-84 K. Smed and G. Deuse. pp. 1-81 CIGRÉ TF 38. “Japanese Experience With a Converter-fed Variable Speed Pumped Storage System. 1.” IEEE Transactions on Energy Conversion. 11. E.” CIGRÉ. paper 11/37-04. Mange. 182. 9. 4. Furuta. FACTS Applications. H. 1-88 Y.” The International Journal on Hydropower & Dams. J. Mamikonyants. W. Noroozian and G. M. “Improving Grid Behavior. L. Kudo. March 1994. Dubois. Vol. “Damping of Power System Oscillations by Use of Controllable Components. IEEE/PES 96TP116-0. “Role of TwoAxis Excitation Turbo-Generators in Power Systems. Y. Mitsuhashi. Shibuya. and J. power system rotor angle stability is inherently a nonlinear control problem. Despite all of these difficulties. the distribution of loads in the network and the topology of the electrical network. • The system often contains significant levels of noise due partly to the constant changing of many loads. including saturation of generators. weekly and seasonal cycles. The changes in electrical torque within a generator can be resolved into two components. as introduced in Chapter 1. which may make communication and monitoring of the system difficult and expensive. As is the case with essentially all physical phenomena. or linearizable. nonlinear power transfer characteristics and nonlinear load characteristics. The electrical torque will often change more rapidly than the mechanical torque input because it is dependent upon the electrical network variables which can change rapidly. These concepts can be generalized in terms of state space modeling [2-68]. These variables include the power transmission capacity of the network and the state of all other machine rotors in the system. These include: • An accurate mathematical representation of an interconnected power system is usually of very high order. many aspects of the control problem are addressable in terms of the vast amount of theoretical information applicable to linear. • An interconnected power system covers a large geographic area. with daily and seasonal cycles as well sudden short term changes. one in phase with rotor angle and the other in phase with rotor speed. • The system is multivariable. systems. and oscillations may arise from slowly evolving operating conditions. The control problem in power system angle stability has several additional complicating factors. These concepts illustrate two separate aspects of the rotor angle stability problem. • The system contains numerous nonlinearities. When a disturbance impacts a generator’s mechanical and electrical torque balance. often containing numerous generators each with their own controllers. Short-term transients will occur following a disturbance to the power system. exciters.Chapter 2 Advanced Linear and Nonlinear Control Design Rotor angle stability of a power system. The rotor angles between generators normally change very slowly as the system changes operating point through daily. often containing several thousand state variables.2-2]. The rotors of all connected AC generators must operate at the same synchronous speed. concerns the electromechanical dynamics of generator rotors [2-1. These components are often referred to respectively as synchronizing and damping torques [23]. A lack of . the rotor of that machine must either speed up or slow down. Changes in the rotor angle relationship between generators are a function of generator loading. • The system is continuously time varying. Small oscillations between generator rotors occur frequently. References 2-11 and 2-67 discuss aspects of the nonlinear nature of power systems. Knowledge of the stability conditions of an interconnected power system is vital for reliable operation. power electronic devices.2-13]. poles and zeros. and HVDC links. resistive breaking. The main impetus is to obtain more effective controllers by having the design account for the system nonlinearities. a model of the system 2-2 . control designs to enhance damping torque usually rely on applications of theory from linear feedback control design. The nonlinear dynamics of the system are transformed into a linear (or partially linear) system so that linear control techniques can be used.1 Nonlinear Control Although power systems are inherently nonlinear most of the control design used in practice is based on linear control theory. Some recent works however have incorporated feedback control for fast valving [2-8]. The availability and proper design of stability controls can significantly extend the safe operating limits of interconnected power systems. Reference 2-54 presents the application of feedback linearization to excitation control for angle stability of a multi-machine system. and large-disturbance stability (synchronizing torques) should exist for most severe disturbances. As the system changes its operating point. 2. Reference 2-53 provides an overview of rotor angle stability related to PSS. there have been several advances in the application of nonlinear control theory. These control designs deal primarily with small disturbance stability described in terms of linear control concepts such as eigenvalues. The control actions include fault clearing. In contrast. This approach has been applied to power systems to control generator power [2-8. network reconfigurations. The result is a transformation or an input signal that contains a nonlinear as well as a linear component. Some or all of the nonlinearities are treated in terms of time varying changes in the system. Small-disturbance stability (damping torques) must always exist. One approach involves feedback linearization. In any disturbance both the synchronizing torque and damping torque aspects of the rotor angle stability problem exist. The problem is referred to as the transient stability problem. Both papers illustrate a significant improvement in damping and transient stability of the power system when the mechanical power input to the generator can be effectively controlled. or generator tripping [2-5–7] and they often do not utilize feedback. Adaptive control. generator fast valving [2-4]. bode plots. In recent years.2-10]. and occasionally nonlinear feedback control design. Synchronizing torque is restored by fast acting control actions. modulated loads.synchronizing torque often leads to rotor angle instability in the first swing of the generator rotor. Reference 2-66 describes feedback linearization applied to a small parallel AC/DC test system for the enhancement of transient stability. and damping. Some common control actuators for small signal stability are generator excitation systems including power system stabilizers (PSS). References 2-12 and 2-42 discuss theoretical aspects. Feedback linearization. Some recent works have discussed the plausibility of using feedback control to modulate the system loads to improve damping [2-9. however. thyristor controlled series capacitors (TCSCs) and power system stabilizers. This approach is only feasible when the number of design parameters to be determined is relatively small. Energy (Lyapunov) function methods. The basic idea is that a passive system always consumes energy. a simple quadratic cost function is used to evaluate controller design parameters for a TCSC. Dissipativity. The basic problem in most of these strategies is to determine the appropriate level of control action and the correct timing for the switching actions. Discontinuous controls or “bang-bang” controls are the most commonly used emergency measures for maintaining transient stability when large disturbances occur in a power network. References 2-6. static var compensators (SVCs). All higher order terms are neglected in linear analysis.can be determined and the control applied according to information about the model or the deviation of the system from the model. Cost function. generator excitation boosting. There are also adaptive control approaches involving fuzzy systems and or neural nets [2-17–19]. Normal forms include the effects of some higher order terms in the Taylor series expansion and can provide insight into the modal interactions exhibited by power systems. Both the linear approach and the normal forms approach use approximations to the full nonlinear system. Some conventional adaptive controllers have been applied to power system problems [2-14–16]. and dynamic braking. In reference 2-20. These approaches are very effective in mitigating disturbances and maintaining rotor angle stability during the first swing of the rotor angles. The standard approach for linearizing a nonlinear system involves using only the first or linear term in the Taylor series expansion of the nonlinear system. Many of these approaches rely on detailed and extensive simulation studies and they do not utilize feedback. Normal forms. References 2-25 and 2-26 discuss the basic theory behind normal forms. This approach to nonlinear control design involves the use of a cost function or penalty function to evaluate the effectiveness of controller parameters for a given control structure. Reference 2-52 proposes a unifying framework for analysis and synthesis of controllers to damp low frequency oscillation in power systems. Discontinuous control. The controllers can be HVDC links. The method involves a large number of simulation studies to determine the best set of design parameters for a set of operating conditions and expected disturbances. Examples include generator tripping.22-3 . In some cases the problem may reduce to being able to detect the appropriate conditions and begin the control sequence. Reference 2-27 is concerned with including second order terms to affect nonlinear tuning of controller gains. 2-21–23. and 2-80 describe these types of control. In general this is a nonlinear control problem. Some approaches do utilize feedback [2-24]. The application of energy (Lyapunov) function methods in transient stability analysis of electric power systems is well known [2-71. Recent work on nonlinear control using normal forms indicates that stressed power systems exhibit characteristics that can be addressed by including additional terms in the Taylor series expansion of the nonlinear system. There are many approaches to adaptive control. series capacitor switching. but the normal forms approach is able to include more of the system nonlinearities. to large investments in new power electronic devices. In recent years. These can range from engineering work retuning existing controllers such as PSS.. Reference 2-70 describes an integrated fuzzy controller for voltage regulation.2 Linear Control Techniques Power system linear control design is a process that can be divided into distinct steps. • Add control equipment to existing devices. Then the problem is to find the most cost-efficient way to solve the angle stability problem. [2-75] study the effects of applying controls to FACTS devices based on energy function methods for lossless system models. Another design situation occurs if the control principle is not yet decided. This results in modelling assumptions that are rather restrictive. 2-4 . • Upgrade control equipment for existing primary controllers such as HVDC.72]. and phase shifting transformers. and they have large regions of validity as they are based on the nonlinear system. Field tests of a fuzzy PSS are also briefly described. One situation arises if the control principle is already decided. controlled series compensation. the number depends on the situation. A main limitation is that the derivation requires that an energy function of the system model be found. power system stabilizer and governor control of a generator. structural uncertainty is not a main issue. Grönquist et al. load modulation control of electrical heaters used in district heating. As described in Chapter 4. fuzzy system and neural network applications to rotor angle stability problems is a research area. An important problem to overcome in power system angle stability applications is that an expert may not be available to provide guidance in forming the fuzzy rules due to the complexity and variability of the dynamic processes. Reference 2-74 investigates and evaluates transient stability enhancement of large-scale power systems by control strategies for unified power flow controller. Reference 2-30 proposes a three-step design procedure for end-use load control. and HVDC link and SVC controllers. and 3) select the compensating parameters. Nonlinear fuzzy and neural net control.e. The advantage of fuzzy controllers is their ability to incorporate nonlinear effects into the resulting control surfaces. Reference 2-73 describes control strategy for HVDC converter controls based on energy function methods. 2. The controls are applied to a CIGRÉ test system that has dynamic properties similar to the Swedish and interconnected Nordic power system. Advantages of energy function control strategies are that the form is independent of the structure. use of energy function principles to derive control strategies for large-scale power systems has received increased research attention [2-73. AVR. i.2-74. 2) choose feedback signals. The steps are: 1) select a location for control actuation. References 2-28 and 2-29 recent work in this area. Some alternatives are listed below: • Retune existing PSS. SVCs. For example. Neural nets provide another and perhaps complimentary solution to the nonlinear control problem through their capacity to learn from system conditions and model nonlinear effects. The key question is to find and evaluate different alternatives. they may rely on local signals.2-75]. References 2-34. 2-55. MATLAB’s Control System Toolbox offers usable tools for model reduction. require extensive computations. Reference 2-38 outlines how synchrony. 2-33 and 2-62. see also references 2-32. Either we adopt a reduced order model suitable for the design method. Therefore it is not feasible to use design models as detailed as those used for time domain simulation. Reference 2-63 describes a controller design and analysis approach to adjust the existing structure of a system by retuning the internal control loops to relocate critical zeros. Many advanced methods. especially for robust control. • Calculate system operating restrictions on-line. The residue of a transfer function is similar to the participation factor of a state space model. such as PSS design. Modeling and model reduction. a generalization of slowcoherency. can be used to construct dynamic equivalents by aggregation of generators. c) approximation of the nonlinear system by ignoring higher-order harmonics. or we are restricted to design methods with moderate computation requirements. It’s important to find a reasonable compromise between model complexity and the design method’s computational requirements.• Add a new power electronic device. Eigenvalue sensitivity [2-31] and participation factors [2-2] are well-known methods of locating control equipment. Retuning is based on an existing extension of modal analysis to linear system zeros. Residues provide information about which modes are most sensitive to gain variations. In automatic control. Structural aspects of controlling active loads are presented in reference 2-32. it is argued that the best model is the simplest one that is accurate enough to fulfill the design requirement. Reference 2-35 describes modeling and model reduction from a control perspective. thus removing the constraints that arise when zeros are at unsuitable locations. and what directions the poles will move when the gain is increased. and 2-36 discuss the use of transfer function residue information for placing and designing controllers. It’s pointed out that model reduction may involve: a) model order reduction in a linear system. This is actually the situation power engineers face when having a complex multi-machine simulation model that includes saturation nonlinearities and also nonlinearities in the power flow equations. Note that the case of model order reduction for high order nonlinear differential equations to low order nonlinear differential equations is not considered. This method is especially suitable for a design aiming at a certain frequency window. • Strengthen the primary system with a new transmission line. The method is reported to be effective in decomposing the eigenanalysis of electromechanical modes 2-5 . Design methods and model reduction are intimately related and some remarks are appropriate. b) model approximation of a nonlinear differential equations by linear systems. For case a). References 2-36 and 2-37 present time-scale decomposition applied to power systems. One advantage of this approach is that it can be applied to field data and does not strictly rely on simulation studies. The models obtained using these methods are generally reduced order because only the system modes observable in the output signal can be incorporated. What are the possible operating conditions? Where are the load centers? Where are the generation areas? Are the power flow directions always the same? For example in the Nordel system that connects the Scandinavian countries. for years with little rain the power flow direction can be reversed. it works with a very limited knowledge about the power system. Line resistance is temperature dependent and Load-Tap-Changers (LTC) can change the nominal transformer ratio. The normal trading pattern gives a power flow from Norway. In reference 2-55 Prony analysis.2-61]. Their voltage and frequency dependence is very different. is presented and in reference 2-56 this method is extended for robustness considerations.. In Sweden. An extension of Prony analysis for multiple output signals is discussed in references 2-78 and 2-79. • Varying load levels during the day. Another approach to obtain dynamic models for linear controller design in power systems is to use identification techniques on system input/output data. Some of these parameters are related to design. The need for robustness design depends on system properties.Identification of models. such as generator time constants and inductance. PSS. Robustness can include many different types of uncertainties and some are listed below.e. One must be aware. Here there is an obvious need for a robust design method that can handle two very different operating conditions.258. there is always a risk of subsequent modification without updating the model. For example direct load switching to damp generator 2-6 . or year. However. • The dynamic model of the power system always has some level of parameter uncertainty. Additional recent work in this area is reported [2-57. • Different load flow patterns. Once determined. such as voltage and frequency dependence. of the limited range of validity of these models [2-56]. Robustness. that might vary with seasons and time of day. Other parameters such as AVR. Once the transfer function model is obtained any standard linear control design procedure can be used. their change is negligible. i. • Load characteristics. week. modified for transfer function identification. • Some parameters change slightly during operation. • Uncertainty in the topology (structure) of the power system—some plants. The control principle itself might be inherently robust. a lot of electric heating is used during winter. through the Swedish west coast to Denmark. and turbine governor are tunable parameters that are easy to change. Even if these parameters have been identified. For other systems. there is a common market for trading electricity. however. such as the New South Wales system in Australia dominated by coal fired plants and well-defined load centers. the need for robust design is less pronounced. lines or transformers might be taken out for maintenance. and in summer air conditioning can be used. oscillations only needs two impedances and one switching level [2-32]. In contrast, the design in reference 2-39 is based on a linear multi-machine model of the entire power system. Many blackouts are caused by cascading disturbances that were not foreseen. Ultimately the power system should be robust to unforeseen disturbances. Power oscillations are often triggered by an initial disturbance that can give a range of possible input amplitudes or operating conditions to the system. The design should also be robust to variation in disturbance amplitude and operating conditions. Linear design methods. The linear control design literature is extensive. Many design methods exist for linear and non-linear systems, and some methods include uncertainty. See references 2-40–43. References 2-2 and 2-44 present overviews of design method for power system applications. The methods can be categorized in different ways such as: • Linear (linear output or state feedback) or nonlinear (on-off) control law. • Linear or nonlinear design method. For example LQ-design can use a nonlinear criteria to design a linear state feedback. • By the physical device the design is aiming for, that is, design for PSS, AVR, HVDC, SVC, or load switching. • By a development scale ranging from academic control methods, to methods used to design controllers implemented in the power system. The evolution of a design method goes through the evolutionary steps: theory, small illustrative simulation study, larger simulation study, redesign, preliminary field test, redesign, and finally working application in a power system. It always falls back to engineering judgment when deciding whether an advanced design method is really necessary, or if a simple control scheme would be sufficient. Measurements of time synchronized phasors opens new possibilities to feedback laws that can be inherently robust. The control design must be simple enough to be reliably applied to a physical system. LQG methodology. Linear quadratic (LQ) control design is an attractive theoretical approach that has not found wide application in practice. Reference 2-39 presents a linear quadratic (LQ) based design method used to find a feedback structure and parameters for PSS/AVR. MATLAB software [2-45] is the main modelling and design tool. A linearized multi-machine model is used to design an optimal LQ-controller with full state feedback. In LQ design a trade-off is done between input energy and performance. It’s suggested that the best generator to damp a certain mode is the one where the optimal controller uses most of its input energy. Instead of using a full state feedback, the feedback is restricted to a sparse structure where most signals are local and only a few strategic global signals are used. This structure is retuned by parametric LQ, that is, numerical minimisation of the loss criteria used in LQ-design. The method’s strength is that the design is done using a multi-machine model, so all PSS and AVRs design is coordinated and simultaneous. The weak points are that the design is done at one operation point and the method does not consider robustness. Reference 2-59 provides another example using LQ design on a very large power system. 2-7 LQG/LTR methodology. Linear quadratic regulators discussed in the previous section have appealing robustness properties, including guaranteed gain margins of 6 dB or greater and phase margins of at least 60 degrees. However such controllers require knowledge of all the system states which usually is not possible or practical in power system applications. In these cases loop transfer recovery (LTR) can be used to estimate unavailable states and still retain the robustness properties of full state feedback with LQG. LQG/LTR is used to design stabilizing controllers for a SVC in references 2-56 and 2-65, and an HVDC link in reference 2-64. Application to power systems is proceeded by identification of an effective low order transfer function which is used as the design model. Desensitized Control. In reference 2-47 a single-machine infinite bus model is used to design a robust regulator integrating AVR and PSS functions. In the design the controller is desensitized, i.e., made insensitive to parametric uncertainties. In this way robustness is included in the design. The design method was originally developed for the “Four-LoopsRegulator” structure used by Electricite de France (EdF), but reference 2-46 shows that the method can also be used for a standard AVR/PSS structure. The method is used to retune EdF’s voltage regulators, and the new values will soon be used in operation. Robust Control, µ-design, H∞.. Robust control is a well-established discipline with textbooks and MATLAB toolboxes [2-43,2-48]. Reference 2-49 proposes a framework for robust stability assessment of controls in multi-machine power systems. Structured Singular Value (SSV) is used to determine stability for varying operation conditions. In the companion paper [2-50], the method is used in a simulation study of a four-machine test system. The simulation results show excellent accuracy of robust stability assessment for a wide range of operating conditions. Reference 2-51 points out that robust controllers designed by µ-design can produce extremely fragile controllers in the sense that vanishing-small perturbations of the coefficients of the designed controller destabilize the closed-loop control system. Reference 2-60 is another study of H∞ control design to power systems. Design methods for active load controllers. Control of active load can be used to improve angle stability. Reference 2-30 describes control of end-user loads in the western USA to enhance stability. Reference 2-9 describes modulated controllable loads for power system stabilization. It’s found that a decentralised two-loop load stabilizer, using local bus voltage and frequency, adds damping to all oscillation modes. Reference 2-81 presents an on-off damping controller for a single machine system. It was used during a field test in southern Sweden to damp oscillations at a 0.9 MW hydro power generator. The controller used estimated machine frequency as input and controlled a 20 kW resistive load via thyristor switches. The results indicate that on-off control of active loads is effective in terms of added damping, and that it is simple to tune and implement. References 2-1 P. M. Anderson and A. A. Fouad, Power System Control and Stability, IEEE Press, revised printing 1994. 2-8 2-2 P. Kundur, Power System Stability and Control, McGraw-Hill, 1994. 2-3 F. P. DeMello and C. Concordia, “Concepts of Synchronous Machine Stability as Affected by Excitation Control,” IEEE Transactions Power Apparatus and Systems, Vol. PAS-88, No 4, April 1969. 2-4 N.B. Bhatt, “Field Experience with Momentary Fast Turbine Valving and Other Special Stability Controls Employed at AEP’s Rockport Plant,” IEEE Transactions on Power Systems, Vol. 11, No. 1, pp. 155–161 February 1996. 2-5 Matsuzawa, K. Yanagihashi, J. Tsukita, M. Sato, T. Nakamura, and A Takeuchi, “Stabilizing Control System Preventing Loss of Synchronism from Extension and Its Actual Operating Experience“, IEEE Transactions on Power Systems, Vol. 10, No. 3, pp. 1606–1613, August. 1995. 2-6 C.A. Stigers, C. S. Woods, J. R. Smith and R. D. Setterstrom, “The Acceleration Trend Relay for Generation Stabilization at Colstrip,” IEEE Transactions on Power Delivery, Vol. 12, No. 3, pp. 1074–1081, July 1997. 2-7 H. Ota, Y. Kitayama, H. Ito, N. Fukushima, K. Omata, K. Morita, and Y. Kokai, “Development of Transient Stability Control System (TSC System) Based on OnLine Stability Calculations,” IEEE Transactions on Power Systems, Vol. 11, No. 3, pp. 1463–1472, August 1996. 2-8 H. Bourlés, F. Colledani, and M. P. Houry, “Robust Continuous Speed Governor Control for Small-Signal and Transient Stability,” IEEE Transactions on Power Systems, Vol. 12, No 1, pp. 129–135, February 1997. 2-9 I. Kamwa, R. Grondin, D. Asber, J. P. Gingras, and G. Trudel, “Active Power Stabilizers for Multimachine Power Systems: Challenges and Prospects,” IEEE Transactions on Power Systems, Vol. 13, No 4, pp. 1352–1358, November 1998. 2-10 O. Samuelsson, B. Eliasson, G. Olsson, “Power Oscillation Damping with Controlled Active Loads,” paper SPT PS 09-04-0620, Proceedings of the IEEE/KTH Stockholm Power Tech Conference, Stockholm, Sweden, June 10–22, 1995, Vol. III, pp. 274–279. 2-11 D. J. Hill (Editor), Special Issue on Nonlinear Phenomena in Power Systems, Proceedings of the IEEE, Vol. 83, no. 11, November 1995. 2-12 L. Gao, L. Chen, Y. Fan, and H. Ma, “DFL-Nonlinear Control Design with Applications in Power Systems,” Automatica, Vol. 28, 1992, pp. 975–979. 2-13 F. Fatehi, J. R. Smith, D. A. Pierre and M. H. Nehrir, “Application of Feedback Linearization to Generator Speed Control in Multimachine Power Systems,” Proceedings of the 27th Annual North American Power Symposium, Bozeman Montana, Oct. 2–3, 1995, pp. 487–492. 2-14 J. R. Smith, D. A. Pierre, I. Sadighi, M. H. Nehrir, and J. F. Hauer, “A Supplementary Adaptive VAR Unit Controller for Power System Damping,” 2-9 IEEE Transactions on Power Systems, Vol. 4, No. 3, pp. 1017–1023, August 1989. 2-15 J. Reeve and M. Sultan, “Gain Scheduling Adaptive Control Strategies for HVDC Systems to Accommodate Large Disturbances,” IEEE Transactions on Power Systems, Vol. 9, No. 1, pp. 366–372, February 1994. 2-16 A. A. Ghandakly and A. M. Farhoud, “A Parametrically Optimized Self-Tuning Regulator for Power System Stabilizers,” IEEE Transactions on Power Systems, Vol. 7, No. 3, pp. 1245–1250, August 1992. 2-17 P. Shamsollahi and O. P. Malik, “An Adaptive Power System Stabilizer Using On-Line Trained Neural Networks,” IEEE Transactions on Energy Conversion, Vol. 12, No 4, pp. 382–387, December 1997. 2-18 M. Lown, E. Swidenbank, and B. W. Hogg, “Adaptive Fuzzy Logic Control of a Turbine Generator System,” IEEE Transactions on Energy Conversion, Vol. 12, No .4, pp. 394–399, December 1997. 2-19 Y. M. Park, U. C. Moon, and K. Y. Lee, “A Self Organizing Power System Stabilizer Using Fuzzy Auto-Regressive Moving Average (FARMA) Model,” IEEE Transactions on Energy Conversion, Vol. 11, No 2, pp. 442–448, June 1996. 2-20 P. S. Dolan, J. R. Smith and W. A. Mittelstadt, “Study of TCSC Optimal Damping Control Parameters for Different Operating Conditions,” IEEE Transactions on Power Systems, Vol. 10, No. 4, pp. 1972–1978, November 1995. 2-21 D. N. Kosterev and W. J. Kolodziej, “Bang-Bang Series Capacitor Transient Stability Control,” IEEE Transactions on Power Systems, Vol. 10, No. 2, pp. 915– 929, May 1995. 2-22 C. W. Taylor, J. R. Mechenbier, and C. E. Matthews, “Transient Excitation Boosting at Grand Coulee Third Power Plant: Power System Applications and Field Tests,” IEEE Transactions on Power Systems, Vol. 8, No. 3, pp. 1291–1298, August 1993. 2-23 M. L. Shelton, W.A Mittelstadt, P. F. Winkelman, and W. L. Bellerby, “Bonneville Power Administration 1400 MW Braking Resistor,” IEEE Transactions on Power Apparatus and Systems, Vol. PAS-94, pp. 602–611, March 1975. 2-24 J. Chang and J. H. Chow, “Time-Optimal Control of Power Systems Requiring Multiple Switchings of Series Capacitors,” IEEE Transactions on Power Systems, Vol. 13, No. 2, pp. 367-373, May 1998. 2-25 S. K. Starret and A. A. Fouad, “Nonlinear Measures of Mode-Machine Participation,” IEEE Transactions on Power Systems, Vol. 13, No. 2, pp. 389– 394, May 1998. 2-10 2-26 G. Jang, V. Vittal, and W. Kliemann, “Effects of Nonlinear Modal Interactions on Control Performance: Use of Normal Forms Technique in Control Design: Part I: General Theory and Procedure,” IEEE Transactions on Power Systems, Vol. 13, No. 2, pp. 401–407, May 1998. 2-27 G. Jang, V. Vittal, and W. Kliemann, “Effects of Nonlinear Modal Interactions On Control Performance: Use of Normal Forms Technique in Control Design: Part II: Case Studies,” IEEE Transactions on Power Systems, Vol. 13, No. 2, pp. 408– 413, May 1998. 2-28 K. A. El-Metwally, G. C. Hancock, and O. P. Malik, “Implementation of a Fuzzy Logic PSS Using a Micro-Controller and Experimental Test Results,” IEEE Transactions on Energy Conversion, Vol. 11, No. 1, pp. 91–96, March 1996. 2-29 M. A. Abido and Y. L. Abdel-Magid, “A Hybrid Neuro-Fuzzy Power System Stabilizer for Multimachine Power Systems,” IEEE Transactions on Power Systems, Vol. 13, No 4, pp. 1323-1330, November 1998. 2-30 J. E. Dagle, D. W. Winiarski, and M. K. Donnelly, End-Use Load Control for Power System Dynamic Stability Enhancement, Pacific Northwest Laboratory, Report PNNL-11488, prepared for the U.S. Department of Energy under Contract DE-AC02-76RLO 1830, February 1997. 2-31 B. Eliasson, Damping of Power Oscillations in Large Power Systems, Ph.D.thesis, CODEN: LUTFD2/(TFRT-1032)/1-155/(1990), Department of Automatic Control, Lund Institute of Technology, Lund, Sweden, 1990. 2-32 O. Samuelsson, Power System Damping—Structural Aspects of Controlling Active Power, Ph.D.-thesis, CODEN: LUTEDX/(TEIE-1014)/1-196/(1997), Department of Industrial Electrical Engineering and Automation, Lund Institute of Technology, Lund, Sweden, 1997. 2-33 N. Yang, J. D. McCalley, and Q. Liu, “TCSC Controller Design for Damping Interarea Oscillations,” IEEE Transactions on Power Systems, Vol. 13, No 4, pp. 1304–1310, November 1998. 2-34 M. E. Aboul-Ela, A. A. Sallam, J. D. McCalley and A. A. Fouad, “ Damping Controller Design for Power System Oscillations Using Global Signals,” IEEE Transactions on Power Systems, Vol. 11, No. 2, pp. 767–773, May 1996. 2-35 R. Johansson, System Modeling and Identification, Prentice-Hall, 1993. 2-36 J. H. Chow, Time-Scale Modeling of Dynamic Networks with Application to Power Systems, Lecture Notes in Control and Information Sciences, Vol. 46, Springer-Verlag, 1982. 2-37 P. W. Sauer, D. J. LaGesse, S. Ahmed-Zaid, and M. A. Pai, “Reduced Order Modeling of Interconnected Multimachine Power Systems Using Time-Scale Decomposition,” IEEE Transactions on Power Systems, Vol. PWRS-2, No. 2, pp. 310–320, May 1987. 2-11 2-38 G. N. Ramaswamy, G. C. Verghese, L. Rouco, C. Vialas, and C. L. DeMarco, “Synchrony, Aggregation and Multi-Area Eigenanalysis”, IEEE Transactions on Power Systems, Vol. 10, No. 4, pp. 1986–1993, November 1995. 2-39 M. Akke, Power System Stabilizers in Multimachine Systems, Tech. Lic.-thesis, CODEN: LUTFD2/(TFRT-3201)/1-103/(1989), Department of Automatic Control, Lund Institute of Technology, Lund, Sweden, 1989. 2-40 B. D. O. Anderson, and J. B. Moore, Optimal Control: Linear Quadratic Methods, Prentice-Hall, Englewoods Cliffs, N. J., USA, 1989. 2-41 J. M. Maciejowski, Multivariable Feedback Design, Addison Wesley, Wookingham, UK, 1989. 2-42 J. J. E. Slotine and W. Li, Applied Nonlinear Control, Prentice-Hall, 1991. 2-43 K. Zhou, J. C. Doyle, K. and Glover, Robust Optimal Control, Prentice-Hall, 1995. 2-44 CIGRÉ TF 38.01.07, Analysis and Control of Power System Oscillations, Brochure 111, December 1996. 2-45 MATLAB—User’s Guide, The MathWorks, Inc., Natick, MA, USA, 1992. 2-46 H. Bourlés, S. Peres, T. Margotin, M. P. Houry, “Analysis and Design of a Robust Coordinated AVR/PSS”, IEEE Transactions on Power Systems, Vol. 13, No 2, pp. 568–575, May 1998. 2-47 A. Heniche, H. Bourlés, and M. P. Houry, “A Desensitized Controller for Voltage Regulation of Power Systems,” IEEE Transactions on Power Systems, Vol. 10, No. 3, pp. 1461–1466, August, 1995. 2-48 G. J. Balas, J. C. Doyle, K. Glover, A. Packard, and R. Smith, µ-Analysis and Synthesis ToolBox, Mathworks, Natick, MA, 1993. 2-49 M. Djukanovic, M. Khammash, V. Vittal, “Application of the Structured Singular Value Theory for Robust Stability and Control Analysis in Multimachine Power Systems—Part-I: Framework Development,” IEEE Transactions on Power Systems, Vol. 13, No 4, pp. 1311–1316, November 1998. 2-50 M. Djukanovic, M. Khammash, V. Vittal, “Application of the Structured Singular Value Theory for Robust Stability and Control Analysis in Multimachine Power Systems—Part-II: Numerical Simulation and Results,” IEEE Transactions on Power Systems, Vol. 13, No 4, pp. 1317–1322, November 1998. 2-51 L. H. Keel and S. P. Bhattacharyya, “Robust, Fragile, or Optimal?,” IEEE Transactions on Automatic Control, Vol. 42, No. 8, August 1997. 2-52 A. M. Stankovic, P. C. Stefanov, G. Tadmor, And D. J. Sobajic, “Dissipativity as a Unifying Control Design Framework for Suppression of Low Frequency Oscillations in Power Systems,” IEEE Transactions on Power Systems, Vol. 14, No 1, pp. 192–199, February 1999. 2-12 3. J. 4. Vol. Kamwa. F. pp. J. 172–178. Illic. pp. Sanchez-Gasca. C. Chow. 1219–1225. 144–152. 6. February 1997. Eng. 1461–1467.” IEEE Transactions on Power Systems. Fatehi. Smith. No 1. Simoes Costa.12.” IEEE Power Engineering Review. G. L. pp. F. 19(1) pp. Vol. F. pp.” IEEE Transactions on Power Systems. 2-65 J. J. 62– 64.” IEEE Transactions on Power Systems. Fatehi. Andersson. and A. F. No. 2-63 L. Trudel. 2. 2-54 J. 2-56 F. pp. Jones. August 1993. 14. 2-55 J.” IEEE Transactions on Power Systems. J. May 1996. London. J. The Algebraic Eigenvalue Problem. A.” IEEE Transactions on Power Systems. and P. 1. Pierre. 46.” Proceedings of the 33rd Conference on Decision and Control. V. No. H. Klein. M. 2-57 A. Vol. 2-13 . Vol. Leirbukt. 1995. No. August 1991. pp. Vol. and D. J. H. 2-60 G. “Robust Power System Controller Design Based on Measured Models.” IEEE Transactions on Power Systems. R. A. and H. Woods. Pierre. 2-61 I. Trudnowski. No. N. 303–312. pp. Taranto and J.” IEEE Transactions on Power Systems. King. No. Vol . Wilkinson. 3. 1965. C. “A Fundamental Study of Interarea Oscillations in Power Systems. August. Hauer and D. No.E. A. February 1999. 3. S. pp. R. “Low-Order Black-Box Models for Control System Design in Large Power Systems. “Computation of Power System Low-Order Models from Time Domain Simulations Using a Hankel Matrix. 10.2-53 M. 3. September 1998. “Damping Control Design Based on Time-Domain Identified Models. No. L. Smith. No. S. Bourles and G. 914–921. G. Chapman. Freitas. Vol. Rogers.” IEEE Transactions on Power Systems. 1282–1290. 205–211. Vol. Andersson. H. “A Robust Frequency Domain Optimization Technique for Tuning Series Compensation Damping Controllers. December 1994. H. November 1997. 4055–4060. Vol. Smith and D.” IEEE Transactions on Power Systems. pp. Oxford University Press. 11. pp. Larsen. B. 12. “Application of Modal Analysis of Zeros to Power System Control and Stability. de Silva. D. and E. “Transfer Function Identification in Power System Applications. W. H. Fatehi. “Closed-Loop Identification and Tuning for Damping Interarea Modes. 8. D. J. 3. Kundur. R. “Nonconventional Control of HVDC Using LQG/LTR Methodology. J. 830–839. 774–780. J. “Stabilizing a Multimachine Power System via Decentralized Feedback Linearizing Excitation Control. pp. Kaufman. 2-64 L. Sanchez-Gasca and J. January 1999. Jones and G. Chow. 1. A. 8. August 1993.” Electric Power Systems Research. 2-59 A. 2-62 J. No. Chow. 2-58 J. “Design of Decentralized Controllers for Large Power Systems Considering Sparsity. Gerin-Lajoie. Vol 11. February 1996. 12. pp. No. Vol. W. presented at Electric Power Systems Operation and Management Conference. Electric Power Systems. Chow. Vol. 3.” IEEE Transactions on Power Systems. Y. December 1990. 4. Trudnowski. “Concepts for Design of FACTS Controllers to Damp Power Swings.” Paper No. 25.” IEEE Transactions on Power Systems. Chong. pp. 453–460. 2-68 E. Control of Power Oscillations in Transmission Systems Using Controllable Series Devices. Jones. K. 1993. Sethares. 2-14 . Pai. Andou.” IEEE Special Publication # 87TH0187-5-PWR. No. Klein.12. S. 2-72 M. Vol. No. “Lyapunov Functions of Lure’-Postnikov form for Structure Preserving Models of Power Systems. 787–799. “Integrated Fuzzy Logic Generator Controller for Stability Enhancement. “H∞ Damping Controller Design in Large Power Systems. 1989. “Effects of Generator and Static Load Nonlinearities on Electromechanical Oscillations. Zürich. 10. Kaprielian. Kluwer Academic Publishers. December 1997. 2-71 D. Lasseter. 10. Licentiate thesis. Energy Function Analysis for Power System Stability. E.158–166. ETH. Part 5.” IEEE Transactions on Power Systems. September 23–25. August 1995. Ueki. Royal Institute of Technology. H. TRITA-EES-9705. 2-75 J. 1989. 2. 1987. N. 1598–1605. “SVC Control Design Concepts for System Dynamic Performance. V. 3367–3372. Chow. and H. 2-74 M. Pai and H. pp. and C. and N. Turi. Application of SVC for System Dynamic Performance. Vol. E. Hill. Y. No. 3. and J. May 1995. F. No. “Power Oscillation Damping Control Strategies for FACTS Devices using Locally Measurable Quantities. 1. 2-70 T. L. A. H. Balu. G. 21. Hiyama. Vol. Larsen and J. Dagle. V. Ghandari. 2-77 M. Farrokhpay. J. Sanchez-Gasca and J. 18. and R. 1997. and J.” IEEE Transactions on Energy Conversion. Clements. pp. 10.” Proceedings: 29th Conference on Decision and Control.X. A. and G. J. J. Vol. 2-67 D. pp. L. S. 1998. 3. Grönquist. Lee. 1283–1289.2-66 S. Vol. R. H. pp. Rogers. Andersson. “Feedback stabilization for an AC/DC Power System Model. February 1995.” IEEE Transactions on Power Systems. Sastry. 2-76 L. 2-69 E. August 1997. J. 2-73 K.” in Sadhana. “A Structure Preserving Energy Function for Stability Analysis of AC/DC Systems. J. J. “Robust Controllers for Power Systems: Open and Closed Loop Design Approaches.” Automatica. No. Alvarado. pp. Larsen. 948–955. pp. 226–231. Johnson. 2-79 D. and J. and M. 2968–2972. 2-15 . F. pp.” Proceedings of the American Control Conference. paper 38-01. Akke. Hauer. “Making Prony Analysis More Accurate Using Multiple Signals. Samuelsson. Vol. 1986. Vol. M. J. 2. and J. Lee and P. “On-Off Control of an Active Load for Power System Damping-Theory and Field Test. M.” IEEE Transactions on Power Systems. Kundur. Trudnowski. Hauer. 14. 608–613. J. J. 2-81 O. June 1998. “SIMO System Identification from Measured Ringdowns. pp.2-78 D. Johnson.” IEEE Transactions on Power Systems. 2-80 D. “Advanced Excitation Controls for Power System Stability Enhancement. J. F. No 1. 14. Trudnowski. February 1999. C.” CIGRÉ. No. May 1999. Moreover. the microprocessor technology makes easy the addition of new functionality such as: • Adaptive or non-linear control. computing power and functionality.) for making the code development fast and easy. • On-line monitoring and the transient recording of meaningful control and process variables. • Simulation of the unit under operation for a check of control parameter values without interfering with the plant. low-cost digital hardware and powerful software tools.Chapter 3 State-of-the-Art in Digital Control The evolution of microprocessor technology—with the consequent availability of reliable. • Sophisticated auto-diagnostics on-board. The availability of reliable hardware and software products together with their high performance and the associated low costs makes the use of the digital technology convenient and feasible for most power system controls. I/O. The progress in microprocessor technology has led to continuously increasing performance in term of speed. . A similar evolution has taken place also in the software field: very powerful development environments have been carried out by specialized software houses and offered on the market. • Data communication with the supervisory systems. alarm and protection functions. and for allowing its independence from the hardware. high-performance. The benefits from the use of digital technology include: • Greater flexibility and adaptability to different practical needs. along with the growing difficulties in the maintenance of analog apparatus—has resulted in development of powerful digital control systems. • Easy and accurate setting and change of control parameter values. software analyzer. profiler. Such environments include a wide set of tools (debugger. etc. Table 3-1 compares analog and digital control. their constancy and independence from environmental conditions. • Reduced number of subset types (electronic boards) to be used for the practical realization. friendly and interactive. • Improvement in the user interface which becomes graphic. through the integration on one chip of functions that in the past required many external components. graphic and mathematical libraries. • Enhanced control. components. Adjustments can be automatically documented. although with a compromise for speed. Very expensive. acknowledgment. facilities for understanding and performing critical tasks. to get maximum accuracy. accuracy is practically unlimited. No extra hardware is required for implementing new functions. Facilities to test and perform new for complex laws. Adaptive control ideas. Fuzzy controllers are equally feasible. Digital implementation provides drift-free settings. Model identification Testing Commissioning Resolution Stability of Parameter values Interfaces Control Laws Direct translation of the digital control model from the studies and simulation platform. concepts such as: familiar and standard nomenclature. can Practically restricted to linear applications. integration.).. requiring an excessive number of simplifications. station can be used for pre-commissioning and training. Difficult execution. Requires the use of very expensive Only analog interfaces need expensive and low drift components in every circuit board. in some applications. At the present time 16-bit A/D converters are used and. according to the mathematical model specified. Parameters don’t change because of parameter limitations.. messages. Thermal drift causes parameters. A lot of instrumentation and other facilities are required. Facilities such as recording. Maintenance services produce parameter changes. drift. causes machine unavailability. considerably reducing costs and risks. etc. engineering units. Software emulates instrumentation for different kinds of testing. Functions are synthesized in software. Item Analog control Digital control Processing Each operation is synthesized by a physical device (sums. limitations. Accuracy depends on signal measuring quality Very expensive. New functions require the addition of new hardware. with severe constraints be easily implemented. etc. Many testing facilities can be made available. Analog time constants used in these Strongly affected by component interfaces are of little importance to the process aging. settings and readings make performance verification very easy. Theoretically infinite. maintenance services. Dataacquisition automatically provides recording of many physical and internal variables.Table 3-1: Comparison between control types. etc. such as step and sinusoidal signals. 24-bit are available. In practice Limited to the converters and to the length of the limited due to noise. graphics).. such as adaptive control. A computer-based simulation and time-consuming job. The need for a lot of instrumentation Commissioning can be carried on easily and in a and numerous calibration tasks make shorter time because parameters are expressed in the commissioning a quite strenuous per unit or seconds. 3-2 . Done through conventional switches Interface design using human factors engineering and instruments. Calculations are done in floating point. word used to make the calculations. Complex control laws. usability (dialogs. laws are difficult to implement. and with the available 80-bit co-processors. cost/benefit relationship may still be unfavorable in small size power plants. the times between the task switching and the latency for the interruptions must be smaller than the smallest control sampling rate. Use of assembly instructions to accelerate the process can bring serious maintenance problems. • Section 4 describes application of digital control for static var compensators.such as RS485 with protocols such as ModBus.1 Review of Digital Control of Dynamic Systems Figure 3-1 shows a computer-controlled dynamic system. remote data logging. Focus is held on serial interfaces . Decreasing. Self-diagnosis None. Maintenance Expensive due to the lack of selfdiagnosis. t k . Wide self-diagnosis resources. For good performance. The real time core manages the execution of the various tasks. Bandwidth Can be very large. The computer processes the measurements using an algorithm. • Section 2 describes the basic structure of a digital-control system. • Section 3 describes application of digital control for a generator excitation system. The conversion is done at the sampling times. The growing use of Ethernet TCP/IP as field bus (suitable cables and connectors are being developed) allows direct and remote connection with the analyst even via Internet. A trend towards not repairing printed circuit boards is being observed. 3-3 . Can be considerably cheaper because of the use of the self-diagnosis. This sequence is converted to an analog signal by a digital-to-analog (D-A) converter. The real-time clock in the computer synchronizes the events.Item Analog control Digital control Software None Connectivity The connection with the other plant devices is done through relays and analog signals. The output from the process y (t ) is a continuous-time signal. and customer protocols compatibility. Costs Stable and relatively low. New techniques are being developed based upon artificial intelligence concepts. Communication resources exist for remote control and integration with supervisory systems for purposes such as: remote settings changing. The output is converted into digital form {y (t k )} by the analog-to-digital (A-D) converter. Software for off-line analysis are available. This chapter examines the following topics: • Section 1 reviews the fundamentals of digital control of dynamic systems. Still limited for extremely fast and complex loops because the Real Time Operating Systems need to keep switching between multiple tasks. and gives a new sequence of numbers {u (t k )} . 3. 3-1. Consider the first order system: x& (t ) = α x + β u For periodic sampling with period τ . t k ) x(t k ) + Γ (t k +1 . 3. Schematic diagram of a computer-controlled system.1. The relationship between the system variables at the sampling instants can be determined.1 Sampling of continuous-time signals Assume that the continuous-time system is given in the following state-space form: x& (t ) = A x(t ) + B u (t ) y (t ) = C x(t ) + D u (t ) The system has r inputs.Computer A-D Clock y (t k ) Algorithm u (t k ) y (t k ) u (t k ) D-A Process Fig. tk = k τ 3-4 . t k ) = e A(tk +1 − tk ) Γ(t k +1 . t k ) = t k +1 − t k As ∫e ds B 0 Example 3-1. p outputs. Normally a D-A converter is constructed so that it holds the analog signal constant until a new conversion is ordered. Given the state at the sampling time t k the state at the next sampling time t k +1 is thus given by: x(t k +1 ) = e A(tk +1 −tk ) x(t k ) + t k +1 ∫e A(tk +1 − s′ ) Bu ( s ′)ds ′ tk The system equation of the sampled system is: x(t k +1 ) = Φ (t k +1 . and is of order n. t k ) u (t k ) y (t k ) = C x(t k ) + D u (t k ) where: Φ(t k +1 . 4752 y (k − 2) + 0. Because Φ = exp( Aτ ) it follows from the properties of matrix function that 3-5 . For a second-order single-input.0278 u (k − 2) Poles and zeros. As a numerical example. Digital filtering provides a great deal of flexibility. Use of the pulse-transfer operator allows the input-output relationship to be conveniently expressed as a rational function y (k ) = H (q ) u (k ) where: H (q) = C (qΙ − Φ ) −1 Γ + D where q is a shift operator with q x(k ) = x(k + 1) Example 3-2.. A digital filter has the general form: y (k ) = −a1 y (k − 1) − a 2 y (k − 2) − .0557 u (k − 1) + 0. since the filter characteristic can easily be changed by tuning a few parameters. a second order low-pass filter with a cutoff frequency of 300 Hz can be modeled by: y (k ) = − y (k − 1) − 1. The poles of a system are the zeros of the denominator of H (q ) or the eigenvalues of Φ ..0278 u (k ) + 0. single-output we have: H (q ) = C (qΙ − Φ ) −1 Γ + D B (q ) = A(q ) = b0 + b1 q −1 + b2 q −2 1 + a1 q −1 + a 2 q −2 This means that the input-output model can be written as: y (k ) + a1 y (k − 1) + a 2 y (k − 2) = b0 u (k ) + b1 u (k − 1) + b2 u (k − 2) Digital filtering. If some or all of a parameters are non-zero there is an auto-regressive filter which has an infinite impulse response. − a n y (k − n) + b0 u (k ) + b1 u (k − 1) + .Applying the formulas above we get: Φ = eατ τ Γ = ∫ eαs ds β = 0 β ατ (e − 1) α The samples system thus becomes: x(kτ + τ ) = eατ x(kτ ) + β ατ (e − 1)u (kτ ) α Pulse transfer operator... + bm u (k − m) where y is the filter output and u is the input measurement value. If all the a parameters are zero we will have a moving average filter with a finite impulse response. Thus. the criteria and algorithms for numerical integration of differential equations must result in numerical solutions close to the solutions of the corresponding continuous-time equations. The basic point to be deeply considered is the altered dynamics of the system under control when moving from the theoretical description by continuous-time differential equations to the practical implementation where finite-differences algebraic equations (discrete-time dynamic system) are used [3-9].2 Dynamic performance For real-time digital control. According to the above and the results shown in Appendix B. This aspect is very important for real-time applications. the adequacy of several numerical integration methods are considered in terms of altered poles and residues of the related system transfer functions. and also that the variations ∆λ. the better the discrete time system approaches the corresponding continuous model. 3-6 . The value of the integration step τ strongly affects the highest poles of the discrete-time system: the lower the value. Computation of the altered dynamics.λi (Φ) = eλ ( A)τ i The equation above gives the mapping from the continuous-time poles to the discretetime poles. it’s not always possible to use complex numerical integration algorithms combined with very short integration step length (τ ) .1. the correspondence of digital control to the nominal analog performance must be verified. ∆c of initial eigenvalues λ and of the related residue c are negligible. 3. Through this analysis it’s obvious that the left half of the s-plane is mapped into the unit disc of the z plane. Analyzing the dynamic behavior of discrete systems it should be guaranteed that the “spurious modes” due to the integration algorithm are stable and timely convergent. Because of computing time constraints. For small variation (sensitivity method) of original generic pole (∆λ/∆ «1). but the higher the digital hardware performance requirement. the critical factor affecting the dynamic behavior of digital control systems are the numerical integration method and the integration step length. In the following. the following relations allow evaluation of the corresponding altered dynamics: ∆λ ≅ Q(r) − 1 λ where r = λτ Q( sτ ) / s : is the “equivalent integrator” of the numerical integration method (see Appendix B). calculated for t=tk=kτ. selected according to u(t) and a term proportional to eλt: x(t)=aeλt +I(t). ) of a continuous-time function x (t ) given by: ) ) ) ) x ( t ) = ae λ t + I (t ) assuming ) 1 λ = ln ρ τ ) ) I ( kτ ) = I k We can see that the solution of the discrete system is formally similar to that of the ) continuous system.λ) is a function whose structure depends on the integration method. u(t )] = λ x (t ) + u(t) dt that allows a complete and correct analysis of a given mode (λ) of a linear “diagonal” dynamic system (see also the conclusion of the first section of Appendix B). in the most general case. (a being a suitable constant). By applying a single-step integration method. In particular. In fact. it’s evident that λ affects the solution of the discrete system like λ affects the solution of the continuous system. a linear combination of the input u(t) and its derivatives. calculated at instant tk and tk+1 (and possibly at instants between tk and tk+1).∆c ∆λ dQ ( r ) ≅ +r c λ r In case of Explicit Euler (EE) integration method: r ( e − 1) An exact calculation of the dynamic performance modification is possible for single step integration methods. sample-values. and αk is. λ ) x k + α k where ρ(τ. the solution of this equation can be expressed as the sum of a particular integral I(t). i. 3-7 . coming back to the first order linear differential equation: Q(r) = r dx (t ) = f [ x (t ). the differential equation becomes a finitedifference equations of type: x K +1 = ρ (τ . selected knowing the following values of αk and a term proportional to ρk.e. The theory of linear finite-difference equations shows the solution of the above equation to be sum of a “particular integral” Ik . where ρ is the root of the characteristic polynomial associated to the difference equation: ∩ x k = aρ k + I k ( a being a suitable constant) The xk values can be considered as the values. sends control signals. considering an oscillating system (imaginary eigenvalues). 3. 3-8 . that the EE integration algorithm results in instability. A digital control system works only on information in numerical form. Figure 3-2 shows the basic structure of a control system: The physical process is observed with sensors. with EE method and τ = 10 ms (integration step). Conversely.) λ can therefore be taken as the eigenvalue of the discrete system and can be compared with the corresponding eigenvalue of the continuous system. therefore the collected electrical variables have to be converted via analog to digital (A/D) converters. links the continuous system eigenvalue to the discrete system eigenvalue. let λ = λ + ∆λ . thus we have: ∆λ ln[ρ ( r )] = −1 λ r This equation highlights the exact transformation that. The central control unit interprets all incoming data from the physical process. Then. depending on the specific singlestep integration method. As an example. Information from different source points distributed in space is brought to the central unit via communication channels. the resulting modified time constant of the corresponding discrete model is 15 ms. the process is influenced through actuators. instead of 20 ms. Table 3-2 shows the results obtained for different integration algorithms.2 Basic Structure of a Digital Control Systems The general structure of a process computer interacting with a physical process consists of the following parts: • central data processing unit • process communication channels • A/D and D/A converters • sensors and actuators • physical process. the change of a 20 ms continuous model time constant is: ∆λ = λ/4. exchanges data with the human operators and accepts their commands. The table also shows. take decisions on the basis of the program instructions. ) In this regard. Figure 3-3 shows the conventional organization of a computer system.. 2 ∆λ ∆λ Re Im EUTRAP − 1 (τλ ) 2 +. It’s more effective to design a computer system where the peripheral units are more independent and have added computing capacity... 6 1 − τ 2 λ 3 +. In this configuration... the peripheral units may communicate directly only with the CPU and only one peripheral unit at the time may be active exchanging data. A computer system is normally built around a central processing unit (CPU) to which are connected peripheral units performing different functions: keyboard. 2 1 − τλ 2 +. 3-9 . 12 12 ∆λ ∆λ Re Computer structures..... 6 ∆λ ∆λ RungeKutta 3 Re Im − 1 1 (τλ ) 3 +... − τ 2 λ3 +.. video interface.. even if the CPU does not need it. 120 120 ∆λ ∆λ Re Im EXTRA − 5 5 (τλ ) 2 +.. 24 24 ∆λ ∆λ RungeKutta 4 Re Im − 1 1 (τλ ) 4 +.Table 3-2: Altered dynamics due to numerical integration method... − τ 3 λ 4 +. Figure 3-4 shows the principle of a busorganized computer system.. The peripheral units are connected together with a bus by which each unit can communicate with all the others. ∆λ/λ Integration ∆λ Pole Shift Method Im EE 1 − τλ +. The CPU-centered configuration is inherently inefficient because all data has to pass through the CPU..... − τ 4 λ5 +.. disk driver. input/output (I/O) cards. 3-10 . 3-3.Fig. In the next two sections we examine two applications of digital control in power systems. The basic structure of a digital-control system I/O card Disk Terminal RAM memory CPU Printer Clock Tape Fig. 3-2. The conventional organization of computer systems. alarms and protection. Referring to Figure 3-5: • The first block on the left includes circuits for measurement and computation of the following process quantities: active and reactive power. Adaptive or Non-Linear Control Measurements Transducers Operator Interface Control Logic Communication Automatic Regulator Gate-Pulse Control Alarms Protections On-Line Monitoring Three-phase thyristor Bridge Fig. Principle of bus organization.CPU RAM memory Tape Printer Clock I/O card Disk Terminal Bus Fig. The power unit supplies the excitation current to the field winding of the generator and mainly consists of a three-phase full-controlled thyristor bridge. It consists of two parts. automatic regulation. generator frequency or 3-11 . 3. respectively named control unit and power unit. and phase control of firing pulse. control logic.3 Evolution of Excitation Control Systems through Microprocessor Technology The general scheme of a modern static excitation system is shown in Figure 3-5. excitation voltage and current. as reference for the requirement specification and the correct design of a digital one. The blocks with dashed line show the additional functions that can be introduced using a digital control system. HV bus voltage. 3-4. Principle scheme of a modern static excitation system. In the following a short description of the characteristics and performances of the conventional analog control unit is given. electrical machine voltage and flux. In the control unit the blocks marked with solid lines represent the conventional functions such as measurement of process quantities. 3-5. the most critical problems. • The second block Automatic Regulator consists of several control loops. resolution and time response of measurement. Less critical data are managed by the modular terminal boards which achieve a distributed I/O. Auxiliary loops limit the working point of generator in over/under excitation and the maximum stator flux. • The dynamic performance of control loops taking into account the altered dynamics from the sample and hold of the I/O signals. logic and communication tasks as well as sampling and holding of the measurements and control variables requiring fast management. • The remaining two marked blocks represent the control and the protection logic. detect fault or incorrect operating conditions. It maintains the firing angle inside the allowed range. and thyristor firing pulse phase modulation. transducers. Passing from the analog to the microprocessor technology. overlapping the previous. PSS) and for compensating the reactive power drop (compounding). 3-12 . which communicate via a local bus.3. They manage the different operating modes of AVR. and provide proper alarm signals in order to improve the safety and the reliability of the system. filtering. A further possible auxiliary loop. regulates the machine reactive power. It performs measuring. regulation. requiring particular care in the design phase are: • The accuracy.speed.1 Hardware architecture The hardware configuration of the newest digital AVRs with decentralized architecture consists of a central system and of modular terminal boards which are placed close to the measurement points. The major part of these measurements requires high resolution (about 12 bit for digital transducers) and fast response (response time less than 20 ms). These peripheral boards communicate with the central system by a field bus. The main loop requires a bandwidth of 5–10 radians/second. • The reliability and availability of the practical realizations. The central system (see Figure 3-6) mainly consists of CPU and A/D-D/A conversion boards. The main loop regulates the stator voltage and has additional feedback for improving the electromechanical stability (power system stabilizers. • The block on the right controls the phase of thyristor firing pulses. compensates the gain variations (depending on supply voltage) and makes the bridge transfer characteristic linear. 3. They give the CPU control to the tasks with the right frequency according to priority level. “ad hoc” schedulers. They permit remote debugging by downloading the machine code to the CPU boards memory and by monitoring the execution as normal debuggers. implements measuringfiltering. Another advantage of operating systems is portability.2 Software organization and development environment The AVR software. in spite of lower use of CPU resources they are able to supply useful primitives to organize the software execution. but quite rigid solutions. real-time kernels or operating systems can be adopted. Operating systems are more flexible. AVR typical hardware architecture. Real-time kernels are an intermediate solution. to compile and to test the software. firing. The software is normally organized in tasks. General purpose real-time development environments are present on the market today. allowing a plain and structured solution. For example filtering functions have high execution frequency to avoid possible altered dynamics of the fast control loops. regulation. 3. Some are PC-based. in order to optimize the hardware resources and achieve the required dynamic performances. whereas high-level programming languages and structured programs lead to plain solutions. for example drivers and protocols for the communication with Local Area Networks (LAN). a few software changes can require a new plan of the scheduler. showing the high or low level code processing. executed by the central system CPU boards. 3-6. To manage the CPU time and the task execution. providing useful tools to write. monitoring and communication functions. 3-13 .3. Similar considerations characterize the software organization: low-level programming languages lead to optimized solutions. characterized by different execution frequencies.Modular Terminal Board 1 Modular Terminal Board n Field Bus Local Bus Exciter + Generator Central system Fig. whereas the communication with the human-machine interface can be executed with lower priority without decreasing the overall performance. logic. while the disadvantage is the nonoptimized use of resources and therefore the requirement of more powerful digital hardware. “Ad hoc” schedulers usually lead to optimized. They can also manage complex resources. It allows both easy and accurate setting of customized data and regulation parameters.3. Diversified AVR configurations are used for different plant sizes. The most popular redundancy configuration is two identical channels: one is active while the other is standby.4 Operator interface A friendly and effective operator interface can easily be implemented. for example cyclic redundancy checks and check-sums. 3.4 Application of Digital Control to SVCs As an example from one manufacturer. waiting to become active if the first malfunctions. Digital hardware able to run code resident on EPROM or flashEPROM and to store operative parameters on EEPROM is employed. filters the digital inputs and monitors the correct work of peripherals. Watchdog circuits are usually required on the boards to detect CPU crash and other fatal conditions. are used to control and to recover data errors. 3.3. Different Human-Machine Interfaces (HMI) are possible. going from a simple LCD display and dedicated push-button to a colored graphical monitor with standard or dedicated keyboard up to a portable PC. The software can also improve the reliability. Data exchange with the control room supervisory system is possible through a LAN.1 Control and protection design The control and protection system provides the following features: • control functions • valve control • protection • alarms • operator interface (locally and remotely) • transient fault recorder • internal supervision • remote interrogation 3-14 . The AVRs of the largest generators have two and sometimes three central controls.3 Reliability and safety Care in the hardware choice concerns the possible reliability improvement for embedded real-time systems. Figure 3-7 shows an overview of the ABB MACH 2 computer structure and indicates how the control system interfaces with the high voltage components of a SVC.4.3. and of the on-line display of the most important process variables. Other precautions. 3. It verifies the measurement coherence. The two field bus types that are used are CAN and TDM. e. while less demanding functions. it’s necessary to use field busses. Transmission speed is normally 100 Mbit/s. PS860. such as operators interface.g. are realized in DSPs. PS 845 Digital Signal Processor Board PS 801 Ethernet board TCP/IP 3-phase voltage measurements Delivery Limit 3-phase current measurements Switch Control PS 850 Breaker Interface Digital Input PS 851 Digital input Digital Output PS 853 Digital output Analog I/O PS 860 PC Motherboard Pentium processor Isol Analog Input PS 862 Power Supply PS 890 Industrial PC Bus Connection PS 870 PC LAN 110 V Communication with MACH2 PC EVT Bus Connection PS 930 GWS OWS Trinitro n Multiscan VCU COMPAQ DE S KPRO Modem Fig.2 Communication Local Area Networks (LANs) are used to connect together several locations (called nodes) so that they can all communicate with each other. firing control etc. The CAN bus is used for communication in both directions between the main computer and digital I/O circuit boards. e. It contains digital and analog I/O boards. and also a DSP board. MACH 2 computer structure. This bus uses the well-proven Carrier Sense Multiple Access with Collision Detection (CSMA/CD) principle to arbitrate access to the bus.. consisting of a main computer and an I/O system.Control & Protection Main Circuit Local Control Room Control & Protection PC I/O rack Backplane PS 880 PCI bus Supervision Board PS820 AC Voltage Meas.g. In designing a new and modern control equipment.3 (Ethernet). The LAN used is the wellknown IEEE 802.. The communication between the main computer and the I/O is by field busses. 3-15 .4. PS 841 AC Current Meas. TCP/IP. fast regulators. All functions within the control and protection system are realized with the MACH 2 building blocks. are realized in the main computer CPU. 3. The transfer speed of a CAN bus is less than that of a TDM bus. 3-7. High speed applications. The I/O rack serves as an intelligent interface between the main computer and the high voltage side equipment. This bus can transfer data using many different protocols (even at the same time). 3. which is fed to the voltage regulator. Another example of integrated self-supervision is the switch control unit.6 System voltage measurement The main objective of the data acquisition unit.3 Internal supervision Periodic maintenance is minimized by the extensive use of self supervision built into all microprocessor-based electronic units. which is measured from the high voltage bus. The regulator output is a susceptance reference value further distributed as an input to the control pulse generator. Feedback for the voltage control is the primary voltage. The voltage response. This can be explained by high preset regulator gain versus new power system impedance. memory test (both program and data memory) and supervision of the I/O system communication over the field busses. the gain supervisor will automatically reduce the voltage regulator gain until the SVC output becomes stable again. and the control rule generator. DAU. 3. In this unit the outputs to the breakers are continuously monitored to detect failure of the output circuits of the board.4. is to measure the voltage response on the primary side of the main transformer. Upon large changes of the impedance in the connecting network the SVC reactive power output may start to oscillate.4. When this occurs an alarm will be given and the gain can manually be reset to normal value from the HMI.4. The internal supervision of microprocessor-based systems includes auxiliary power supervision. program execution supervision (stall alarm).4.4 Automatic voltage control The automatic voltage control consists of a closed-loop voltage regulator formed by a positive sequence voltage response. The TDM bus operation status is continuously monitored by the receiving nodes in the control and protection system. and by the possibility to check all measured values during operation without disturbing the operation. a PI-regulator with variable gain. The reference range is limited by parameters and indicated on the HMI for operator feedback. Bref . and detected faults will give alarm. is processed in the DAU in order to meet the dynamic demands regarding speed and stability. 3-16 .The TDM bus is single direction and used for high-speed measurement signals.5 Gain supervisor The control system provides a gain supervisor function for supervision of the SVC MVAr output. The operation of the field busses is monitored by a supervisory function in the control and protection system that continuously writes and reads to/from each individual node of the system. The voltage reference signal from the HMI is transformed into a reference for the voltage regulator. For oscillations detected in the susceptance reference. 3. 3. CPG. The susceptance reference serves as the control reference value from the voltage regulator while the SVC bus voltage is used for synchronization of triggering pulses and simulation of TCR and TSC current. The voltage space vector is thereafter fed to a function that can extract both the positive and negative sequence components from the voltage space vector. These are typically Windows NT computers interfaced to the main computer via the local area network. Therefore an α/β-transformation is employed in order to transform the three-phase voltage into a rotating vector system in the alpha/beta plane.8 Operators interface The operator interfaces are provided by workstations for local and remote control (Figure 3-8). 3-17 .4. VCU. the TCR must have different controllers for positive and negative sequence voltage components. LAN.7 Control pulse generator The main objective of the Control Pulse Generator.If a TCR is operated with symmetrical firing.4. The other control functions are as follows: • power oscillation damper • control of external devices • loss minimization functions • TCR direct current control • sequence control of breakers • protective control functions • undervoltage control strategy • supervision of faults in the thyristor triggering system 3. 3. The most important input quantities are the susceptance reference from the voltage regulator and the measured SVC-bus voltage. if the task is to control unsymmetries. a so-called space vector representation (see Appendix C). the true voltage response fed to the closed control loop should not contain negative sequence components or harmonics other than fundamental. is to generate control pulses for further distribution to the Valve Control Unit. On the other hand. • Alarms—all events classified as alarms in order of severity. The operator interface may also provide high performance transient fault recording. An on-line graphical debugger allows the user to view several graphical programming tool drawings at the same time and inspect any internal software “signal” in real time by just double-clicking on the line that represents the signal.4. 3-8.g. Operator work station. setpoints. The graphical debugger also allows all thresholds. The VCU is realized by two special boards giving a compact design. but also for maintenance and debugging.10 Valve control The Valve Control Unit (VCU) is the electrical/optical interface between the firing control system and the thyristor valves. • Fault list—all persistent alarms in chronological order. • Sequence of events—all events/alarms including logging of orders..Trinitron Multiscan 20 se COMPAQ DESKPRO Fig. This fact makes the graphical debugger a very useful not only for monitoring.4. 3.9 Remote interrogation Remote interrogation of the control system may be provided by modem communication. The main functions are: • Full graphic status displays of various views. 3-18 . 3. • Display and adjustment of protection settings and control parameters. and timer settings to be easily displayed in various formats (e. as tables). 3.11 Application software development The application software for the MACH 2 control and protection system are produced using a fully graphical code generating tool called HiDraw. 3. It is Windows-based software that is very easy to use as it is based on the easiest possible select. for further analysis. It’s designed to produce code either in a high level language (PL/M or ANSI standard C) or in assembly language. It’s very difficult to make long-term predictions. Parameters can easily be changed by double clicking on their value. There are also a number of supporting functions such as single or multiple stepping of tasks (one page is normally a task) and coordinated sampling of signals. • The improvement in speed and reliability of the communication channels will allow the creation of safe methods for remote commissioning and maintenance.3. and stored in the flash PROMs. a fully symbolic debugger is available either on a computer or a dumb terminal. a double click on the page reference will automatically open the new page and allow the trace to continue immediately on the new page.5 Trends in Digital Control Electronics have evolved at an astounding rate in the last years. For fault tracking. 3-19 . A schematic drawn in HiDraw consists of a number of pages.12 Debugging facilities For debugging. it’s also possible to transfer sets of signal values to other Windows-compatible programs. The result is a file that is ready to be downloaded from the computer to the target.4. One page specifies cycle times and execution order of the other pages. The fact that the debugger allows inspection of signals while the application is running makes it very useful not only for debugging but also facilitates maintenance. Because the debugger works in the Windows environment. although trends are apparent. HiDraw includes an on-screen reasonability check of the drawn schematic. For functions not available in a comprehensive library (one for each type of processor board) it’s very easy to design a new block and link to the schematic with a simple name reference. a fully graphical debugger operating under Windows is used.4. As output. it’s easy to follow a signal through several pages because when a signal passes from one page to another. The next step in the workflow is to run the make file (on the same computer) which means invocation of the necessary compiler/assembler and link locate programs (usually obtained from the chip manufacturers). As a complement. place method. it produces code and a “make” file ready to be processed. The debugger allows the operator to view several HiDraw pages at the same time. and automatic cross reference between the pages. such as Excel. and look at any internal software “signal” in real time by just double clicking on the line that represents the signal. drag. Basically it can be said that: • Availability of resources from digital controllers will shorten the time between the development of a new control law and its practical implementation. C. Macmillan New Electronics. Computer Systems for Automation and Control. as well as new technologies and the enhancement of software techniques. G. Second Edition.• The adoption of common use. Raffaelli. Laplante. Pozzi. Real-Time Systems Design and Analysis. • There will be a need to develop more system analysis tools to handle the large diversity of control laws performing in different machines of the system. S. Levine. “The ENEL’s Experience on the Evolution of Excitation Control 3-20 . Virk. “A New Digital Simulator of the TurbineAlternator-Grid System (STAR) for Control Apparatus Closed-Loop Tests. 3-9 V. Ottaviani. Std 421.” EPE. Baroffio. 1995. Theory and Design. For instance. Berlin. • On-site implementing upgrades should be easier. • An easy access to a superior hierarchical level can be provided by object-oriented technologies. On the other hand. S. 1992. the use of OPC (Object Linking and Embedding for Process Control). 1991 International Editions. and E. 3-3 G. Olsson and G. Piani. CRC Press. making systems integration much easier and lowering maintenance costs because of high scale production. 1997. Åström and Björn Wittenmark. 3-2 G.41990. IEEE Press. Arcidiancono. 3-5 Corbetta and G. because no hardware changes will be needed. Guide for the Preparation of Excitation System Specifications. Corsi. 1991. 3-4 IEEE.S. more flexible hardware and software is an observed trend. References 3-1 Karl J. and makes available all of its resources (adjustments and commands) without the system integrator worrying about the knowledge of the controller implementation details. G. 1999. Computer-Controlled Systems. Togno. The Control Handbook. “Digital Measurement Procedures in a MicroprocessorBased Excitation System. Rosa. as a result of the electronic circuit large scale integration increase. S. 3-7 Phillip A. • The costs will drop. • PC-based systems are becoming more cost-effective. 3-8 S.” IEEE/PES Summer Meeting. where the controller opens a window in a higher level supervisory system. Prentice Hall. Digital Computer Control Systems. Corsi. design and documentation efforts demanded by software modification could be large. and have been occupying traditional PLC space. 3-6 W. and G. Tagliabue. Ottaviani. 1999. Florence. M. 13. Technical Report RU 8037 AU.Systems through Microprocessor Technology. 3-10 ABB Power Systems. 3. Vol. 292–299. 3-21 . 1999. MACH 2 Description. No. pp.” IEEE Transactions on Energy Conversion. September 1998. . The future operators also need to have the ability to specify the operating strategy in qualitative form. which is then translated into quantitative form in order to be processed by the computer control. Data and rules are formulated on a symbolic level in pseudo-natural language. or has been. Beside the control-center applications. the formulation of goals and the subsequent application of rules. The large quantity of information required can be provided in many cases through advances in telecommunications and computing techniques.e. • Artificial Neural Networks (ANN) which infer quantitative information through approximation techniques and classify quantitative data into higher-order qualitative categories. the case in present and past practice. the “reasoning process. One of the main motivations for using intelligent systems is to provide this important interface between qualitative and quantitative information. • Decision Trees (DT) which classifies quantitative data into discrete sets of qualitative categories. In the ideal case. • Fuzzy Systems (FS) which quantify qualitative knowledge including uncertainties. . For example consider closed-loop generator control. Heuristic ∗ Control is typically verified by nonlinear simulation for a limited number of operating conditions and disturbances.∗ In case of a large disturbance.Chapter 4 State-of-the-Art in Intelligent Controls Deregulation requires that utilities exercise less conservative operation regimes and more precise power-flow control. This is possible only by monitoring and controlling the system in much more detail than is. A consideration with existing control methods is that the control law is based mainly on a linearized model and the control parameters are tuned for certain operating conditions. and the controller parameters may no longer be valid. Expert System techniques are often associated with the software engineering concept of intelligent computing environments.” i. who can be overwhelmed in emergency situations when fast decisions are needed. Up until now. such as negative damping. In this case the controller may even add a destabilizing effect. There is still the need for evaluation techniques that extract the salient information from the large amount of raw data to use for higher-order processing. intelligent control can be applied in a decentralized manner. the extraction of qualitative information is still done by the human expert. are transparent to the user. Intelligent Systems can be categorized as: • Expert Systems (ES) which process qualitative as well as quantitative knowledge with emphasis on the qualitative results. the system conditions will deviate significantly from the linearized condition. deduction) process. shape and width of the membership function by empirical rules. a2. and have a triangular shape and a maximal width o σ of 20 F as shown in Figure 4-1. Fuzzy systems come in two flavors: • Empirical or rule-based fuzzy systems • Self-adaptive fuzzy systems (self-organized or unsupervised fuzzy systems) In the literature. 4-2 . However. fuzzy sets and fuzzy control are mostly discussed in terms of qualitative attributes like cold or warm and qualitative rules like “if temperature is cold with a likelihood of 0. Expert system techniques are therefore usually implemented as off-line decision aids. Instead of defining center. It’s therefore necessary to establish the fuzzy sets and rules in a more systematic. in the area of power system control.e. in the case of load forecasting. autonomous manner and the corresponding fuzzy systems are referred to as self-adaptive fuzzy systems. one can choose a more systematic approach using data analysis. and a3. the membership function describing the three fuzzy sets cold. In the following we will concentrate on the applications of Fuzzy Systems..” These empirical rules are often established from existing expertise in manual control and the corresponding fuzzy systems are referred to as empirical fuzzy systems. expert system techniques are often discussed in the context of an intelligent user-friendly human-machine interface. A clustering algorithm might have identified three typical temperatures a1.06 [4-3–7]. µ2 and µ3. Reference 4-2 discusses a voltage-control expert system for the off-line changes of onload tap changer settings. Other examples of applications of expert systems for power system off-line monitoring and control can be found in reports published by several task forces of CIGRÉ WG 38. a2 = 70oF and a3 = 100oF. warm and hot may be centered at a1 = 40oF. Let us briefly illustrate these concepts by looking at the example of fuzzy temperature sets [4-9]. as for example power system stabilizers. sampling of the load data might indicate that the load exhibits three different behaviors correlated with the temperature.1 Fuzzy Systems for Power System Control Fuzzy sets and systems were first introduced by Zadeh [4-8]. where not only real data and network topology maps but also abstract reasoning concepts like rules and decision trees are displayed graphically [4-1]. this expertise may not exist for unusual operating conditions. Artificial Neural Networks and Decision Trees to power system control.reasoning (inspired by rules of thumb) are implemented in order to limit the number of branches of the decision tree to be exploited during the reasoning (i. For example. Due to the nature of this approach. 4.7 then increase heating fast. with the width of the cluster defining the width of the membership functions µ1. It specifically draws attention to the fact that the heuristic nature of the off-line control rules limits their range of validity. If the initial input set is the range of temperatures from 0oF to 120oF. a mapping from a crisp number to the fuzzy set can be defined consisting of this number only (singleton fuzzification). 4-3 . One can further define fuzzy rules either by establishing these rules empirically or in a self-adaptive manner. for example. Self-organizing fuzzy controllers therefore fall into the class of adaptive controllers and the related stability issues can be explored with adaptive control techniques. intersection and complement. Also a mapping from a fuzzy set onto a number can be defined by choosing this number as the center of average of the integral defined by the fuzzy membership function (center of average defuzzification). one can define union. 3 1 20 30 40 50 60 70 a1 80 90 a2 100 a3 110 120 Temperature e[F] Fig. Whether one defines the membership function empirically or self-adaptively.t). 2. Membership function for fuzzy temperature sets.µi(e) i = 1. In addition. F(e) = u This mapping F will be constructed as an approximation to the controller φ(e. For the purpose of power system control it is sufficient to note that the fuzzy system is a mapping F: Un ⊆ ℜn ->ℜ. Stability of power system controllers is discussed in more detail in reference 4-10. which is continuously differentiable. intersection and complement of two fuzzy sets A and B by defining the membership functions corresponding to union. It is shown [4-9] that there is a class of self-adaptive fuzzy systems F with Gaussian membership functions φj that can be written in a closed form as: m u = F(e ) = ∑ b j φ j (e ) j=1 Self-adaptive fuzzy systems given in closed form have the advantage that stability analysis can be performed and tasks like optimal control can be addressed. the number of fuzzy sets and membership functions. 4-1. instead of the triangular (or the sometimes used trapezoidal shape) without altering the degree of membership of any given temperature significantly. As an analogy to crisp sets. there are always some degrees of freedom. one can choose a Gaussian function. Figure 4-2 shows the structure of fuzzy system. Finally. Structure of a fuzzy system. µB) → u u∈ℜ Fig. load.1. and generation can change stochastically and discontinuously. and all projects report better tracking capabilities of the fuzzy controllers compared to conventional controllers. In most cases. the membership functions are established based on data samples. For feasibility studies. Neither approach to fuzzy systems necessarily needs a detailed state-space model of the controller. The self-adaptive controllers can be easily tuned to different operating conditions. The advantage of the empirical approach is that heuristics and human knowledge can be incorporated. The comparison of fuzzy controllers and conventional controllers stresses advantages of fuzzy controllers as being “generic” parametric models instead of circuit-based state space models. most authors experiment with empirical 4-4 . µA) A Fuzzy Rule Base and Fuzzy Inference Engine IF e is A AND e1 is A1 ... The majority of fuzzy controllers can be found in the area of excitation control. However. This is especially important for power system control where topology. especially power system stabilizers (PSS). Although the majority of investigations perform feasibility studies using computer simulation only. the sensitivity issues concerning the range of validity of the tuning and the detection of changes of operating conditions still needs to be investigated for conventional as well as for fuzzy controllers. However.1 State-of-the-art of fuzzy control for power systems We now give an overview of studies of fuzzy systems in the area of power system or generation control [4-11]. 4. 4-2. A lot of progress has been made concerning the application of fuzzy systems to power system control problems.e ∈ ℜn Fuzzifier e → Fuzzy Set (A. An upcoming important area is control of power electronic devices. AND en is An THEN u is B B Defuzzifier Fuzzy Set (B. the demonstration of stability for this type of controller is very tedious if not impossible. several authors study the implementation of the fuzzy controller on a PC or DSP in order to control actual small generators or motors in a laboratory environment. Instead. the output stabilizing signal was calculated based on the representation of the alternator state in the phase plane. interacting oscillation modes. Experience with the more sophisticated types of fuzzy logic control is even more limited.rules and data.4 MVA hydro units in the Kyushu Electric Power System [4-14]. Although the fuzzy logic power system stabilizers are field tested as described above. the self-organizing Fuzzy Auto-Regressive Moving Average (FARMA) controller was studied to enhance the low frequency damping of a synchronous machine [4-15].1.2 Implementation of fuzzy logic PSS In joint research. Kugimiya. 4. a twoyear evaluation of the FLPSS was finished in March 1996 on 30. Damping of oscillations were significantly increased. See Figure 4-3. They took into account the PID information of the generator speed. The proposed fuzzy logic power system stabilizer (FLPSS) is set up by using a microcomputer with AD and DA conversion interfaces. 4.1. Hassan. the FARMA FLC needs no expert in making control rules. In addition. In power systems. 1997 on a hydro unit with the rating of 90 MVA at the Hitotsuse Hydro Power Station in the Kyushu Electric Power System. there is limited experience.3 The future of fuzzy logic power system stability controls There is continued debate on the fuzzy versus conventional control (reference 4-59 is entertaining and instructive). The FLPSS has been in service since June 19. the effectiveness of the FLPSS was demonstrated [4-13]. many actual industrial applications (in other industries) are for higher level or supervisory control [4-60. All the signal conditioning and the generation of stabilizing signals are performed by the on-line microcomputer. even in the simulation world. have been installed on a microprocessor and tested in a research lab environment either in academia or a utility. The next section describes an operational application of fuzzy control in a real power system. In this method. of fuzzy logic power system stability controls in large power systems with multiple. Malik and Hope applied the fuzzy logic control (FLC) to PSS design [4-12]. where the rule base and membership functions are supplied by an expert or tuned off-line through experiment. experiments on a 5 kVA laboratory system. rules are generated using the history of input-output pairs. Although most of the literature on power system fuzzy logic control is on replacement of conventional control.2 and 23. Hiyama. and implementation on an actual 5 MVA hydro unit. Kumamoto University and the Kyushu Electric Power Company proposed a microcomputer-based fuzzy logic power system stabilizer (FLPSS) to enhance power system stability through control of thyristor exciters. Through simulation studies.4-61]. however. Recently. and Satoh proposed PID type fuzzy logic PSS [4-13]. using self-organizing techniques. In contrast with a conventional FLC. Additional parameters were also tuned off-line to minimize the performance index. fuzzy logic controls 4-5 . The generated rules are stored in the fuzzy rule space and updated on-line by a self-organizing procedure. A few projects. and discrete controls. It can be used in nearly every area of power systems where a task can be formulated as an approximation problem.2 ANN for Power System Control Artificial neural networks have been applied in technical areas since the early 1960s. As a classifier (approximation of the Boolean function secure/insecure or trip/no-trip signal) it is applied in power system security assessment [4-16] and on-line security control to initiate load shedding at a bus [4-17]. • unsupervised ANNs. The multi-layer perceptron (MLP) is probably the most heavily investigated supervised ANN model.may be attractive for higher level. rather than as replacement of essentially linear continuous controls. 4-6 . ANNs come in two major categories: • supervised ANNS. when Widrow and Hoff developed an adaptive least square estimator called ADALINE.A/D computer Power Transducer FLPSS Fig. The MLP is then used as a regression tool in order to estimate additional parameters [4-18]. In this framework. Basic configuration of PSS prototype and its overview. They therefore solve problems similar to problems solved by regression and parameter estimation techniques. 4-3. for example TRUE or FALSE coded with binary numbers. classification tasks can be formulated as the task of finding a regression model for the function which maps an input vector x onto its class label. Supervised neural networks perform approximation tasks using a special combination of non-linear basis functions called sigmoid functions. nonlinear. 4. The MLP is often used in combination with Fuzzy Systems where qualitative attributes like hot or cold temperatures are first translated into numbers. Protection Unit Voltage Detectors Timer WT AVR Exciter ‚ r AC 100V G Monitoring Unit UPS PQVF AC 100V D/A Micro. This survey was updated by Niebur and Dillon [4-26] based on a review of more than 400 publications regrouped into 200 different projects published before April 1995. Therefore power system control is still done in the most conservative manner.Unsupervised networks reduce the complexity of the data sets by either reducing the dimensionality of the input data or by grouping input data into categories of “typical” data and by constructing a typical presentation (code vector) for each class.2. Unsupervised neural nets fall into the same class of tools as statistical non-parametric data analysis.06. [4-25] assessed their potential for transient stability and static security assessment. Field tests for 4-7 .4-22].1 ANN applications In the 1970s simple ANN-based machine-learning techniques were explored for transient stability [4-23]. it’s the practice of some experienced operators to even remove conventional controllers like power system stabilizers. It was mainly motivated by the lack of automated tools in the utilities and by the expected economic gain. The potential of ANNs for non-linear adaptive filtering and control stimulated research in the area of control of highly non-linear power system behavior. A bibliographical survey covering 1988–1993 world-wide is presented in the paper by the CIGRÉ Task Force 38. 4. In critical situations. and encoding or decoding techniques. the control tool. Research in other major application areas like security assessment attempts to exploit the data reduction. and Aggoune et al. Unsupervised ANNs which quantize data into categories provide a choice of free parameters. and regression capabilities of ANN in combination with conventional simulation techniques. In the area of power system security assessment the ART network [4-19] and the Kohonen map [4-20] are used to reduce the space of all feasible operating points into a finite set of typical operating points. Unsupervised ANNs are often used in combination with supervised approaches or conventional tools. The ART networks fixes the radius of the class but allows a variable number of classes. whether conventional or ANN has to be operated on-line. Time-series prediction in the area of load forecasting has been one of the most examined areas for ANN applications.06 on Artificial Neural Net Applications in Power Systems [4-4]. New control tools need to be extensively tested before they can be integrated into the existing complex power system. For power system control. classification. With the emergence of more powerful computers. [4-24]. clustering algorithms. Available reaction time is extremely limited and control errors can easily lead to a breakdown in a substantial portion of the interconnected system. The unsupervised net serves as pre-processing tool for data reduction and the supervised net estimates associated parameters like security classes [4-21. when Sobajic et al. ANN gained renewed interest from 1988 on.. whereas Kohonen’s self-organizing feature map fixes the number of categories but allows varying class sizes. These projects have led to a sudden upsurge in applying neural net approaches to many power system problems. Classification does not in itself indicate distance from the operating condition to the insecure conditions. Further. a pilot installation is running successfully in the island of Lemnos. In many North American utilities. The main disadvantages of this approach include intensive labor requirement. in order to maneuver the system with respect to security boundaries. Classification involves determining whether the system is secure or insecure under pre-specified contingencies. and little flexibility in integrating with the energy management system (EMS). These precontingency parameters are called critical parameters. The procedure for boundary visualization consists of the following major steps: 1. as well as how to adjust them. varying one critical parameter while keeping the other constant. 4-8 .2 ANN application in security assessment Security assessment can be divided into two levels: classification and boundary determination. on the other hand. In the area of control. Once the boundary is identified. involves quantifying this distance. Base case construction: Construct a base case power flow solution that appropriately models the system conditions. however. Similar remarks apply to the area of security assessment. A boundary is represented by constraints imposed on parameters characterizing pre-contingency conditions. the traditional boundary characterization is a twodimensional graph called a nomogram [4-29–31]. An ANN technique has been used in a security boundary visualization method to overcome these disadvantages [432. Greece [4-28]. Security problem identification: Identify the specific set of security problems to be characterized and operating parameter candidates that may have influence on them.2. security assessment for any operating point can be given as the “distance” between the current operating point and the boundary.control. inaccurate boundary representation. in both areas. Points on the nomogram curve are determined by repeating computer simulations. The non-critical parameters are set to constant values. To develop a nomogram. have been reported for isolated components like photovoltaic storage [4-27]. For fast dynamic security monitoring in a medium scale network with diesel and wind power production.4-33]. Boundary determination. 4. Japan [4-27]. Assessment in terms of pre-contingency operating parameters instead of the post-contingency performance measure is more meaningful to the operator as it directly identifies the parameters to control. data covering significant periods of operation are not readily available and have to be collected for the specific ANN applications. two critical parameters are chosen and all other critical parameters are set to selected values within a typical operating range. 2. field tests are reported by Kumamoto University and Sanyo Electric. The inaccuracy of the nomogram results mainly from linear interpolation between boundary points and insufficient information contained in critical parameters. and is therefore a function of u. 4. A systematic method.3. 5.e.g. Standard MLP networks have been used for this application. Data generation: Automatically generate a database with each record consisting of pre-contingency operating parameters and the corresponding post-contingency measure. Feature selection: Select the best subset of pre-contingency operating parameters for use in predicting the post-contingency performance measure. The ultimate boundary captured by the whole procedure will characterize the data that is provided to the neural network. i. x is the critical parameter vector. the computation used in solving equations (1) and (2) is based on a derived form of the neural network mapping function. resulting in the relation R = f(x). the boundary will be incorrect. in identifying the boundary. 6.. and u is the input parameter vector to the power flow program. Visualization: Provide an easily understood automatic visualization of the security boundary in the space of operating parameters that can be monitored and controlled by the system operator. Once the neural network is trained. the problem of boundary identification is solved by finding x that simultaneously satisfies: f (x) . This data is used to train a neural network to compute the post-contingency performance measure R as output given the pre-contingency operating parameters x as input. expressed as 4-9 . the relationship between the post-contingency performance and the pre-contingency operating parameters can be inverted. The vector x may include both independent critical parameters (e. subject to the power flow equations. That is. the boundary for a single security problem under a given contingency. with each sample corresponding to a simulation of the same contingency but for different operating conditions.. Rb is the threshold value of R. Because the presented parameters (those corresponding to the two coordinate axes) must be varied in drawing the boundary. This data consists of a large number of samples. (2) represents the power flow equations. and consisting of values for pre-contingency operating parameters together with the post-contingency performance measure. the influence of these variations on dependent critical parameters should be considered accordingly. For visualization of an individual boundary.Rb = 0 (1) h ( u) = 0 (2) where (1) represents the neural network mapping function. If this data does not reflect what actually occurs in system operations. flows). call ASAS [4-35] has been developed to generate the data for neural network training. where f represents the neural network mapping function.g.. Neural network training: Train a neural network using the selected parameters and the database to map the relationship from the pre-contingency operating parameters to the post-contingency performance measure. Data generation is a very important step [4-34]. real power injections) and dependent critical parameters (e. if there are no other individual boundary functions between the two binding functions identified in the previous interval. 4-4. y0 is the dependent critical parameter vector corresponding to a specific operating condition. To do this. and repeats until it reaches the maximum of z2. Then it increases z1 by a fixed step. Neural networks have attractive capacity in handling sensory information. The use of an artificial neural network is very attractive because of its nonlinear mapping ability.3 ANN application in power system stabilization Control of large-scale systems such as power systems has been recognized as a foremost challenges in control engineering due to its nonlinearity and complexity. In the next interval. gy (y0. Algorithm illustration for composite boundary visualization. ∆z2)) . z is the independent critical parameter vector. As shown in Fig. The composite boundary is therefore identified as this pair of individual boundaries. it is not necessary to perform the check for this interval.f (z. and gy models the influence of the z1 and z2 changes on the dependent critical parameters y. and performing collective learning from the data sets given for a subsystem in the decentralized control approach. 4-4. updates y. The approximation property of neural networks can make it possible to organize subsystem dynamics to a 4-10 . we rank the functions in descending order of z2. where z1 and z2 represent the two presented parameters. For complexity coming from high dimension or from the spatial distribution of a large-scale system. z2 B3 B2 B1 z1 Fig. ∆z1. we check an arbitrarily selected point (marked with crosses) between them to see if it is secure for all security constraints. then the algorithm stops.2. 4.y]. In this case. Once it is no longer possible to find any secure point. The visualization algorithm starts from the minimum value of z1 and solves equation (3) for z2. and the corresponding neighboring individual boundary functions must be the binding functions for the composite boundary for this interval. for each interval ∆z1. In visualizing a boundary comprised of two or more constraints. we first identify the two individual boundary functions that are binding for the composite boundary. If so. then these functions are also binding for the new interval. For each pair of neighboring functions in this rank. we proceed as follows. this point is inside the secure region. decentralized control is a practical approach. y is the dependent critical parameter vector.Rb = 0 (3) where x=[z. solves for z2. represent output and input variables. We consider a system in the form of the general nonlinear auto-regressive moving average (NARMA) model: y ( k + 1) = f ( y ( k ). During the low-frequency oscillation. Thus. The use of general quadratic cost function provides “optimal” performance with respect to trade-off between the tracking error and control effort.e. y ( k − 1). (4) where y and u . However.⋅ ⋅ ⋅. and n and m represent the respective output and input delay orders. a neural network based power system stabilizer can be designed for a large-scale power system when only local input/output information data for a subsystem. The above control objectives can be achieved by minimizing the following well-known quadratic cost function: 4-11 . Most cases are limited to speed deviation control with supplementary excitation signal for a single generator–infinite bus system. and this is a key variable affecting the rotor dynamics. The use of neural networks’ learning ability avoids complex mathematical analysis in solving control problems when plant dynamics are complex and highly nonlinear. the MRAC approach has difficulty in selecting an appropriate reference model. k represents time index. u ( k ). Optimal tracking neuro-controller. Since the cost function is defined over a finite time interval. u( k − m + 1)) . Electrical power has nonlinear properties. A practical power system stabilizer to enhance the damping of the low-frequency oscillations must be robust over a wide range of operating conditions. u( k − 1). The feedforward neuro-controller (FFNC) generates the steadystate control input to keep the plant output to a given reference value. and the feedback neuro-controller (FBNC) generates the transient control input to stabilize error dynamics along the optimal path while minimizing the cost function. respectively. However. Recently. power plant data.. is available. conventional PSS design approaches based on linearization around the normal operating point have deficiencies and difficulties coming from nonlinearities in the system. A novel inverse mapping concept is developed to design the FFNC using a neuro-identifier. rotor oscillates due to the unbalance between mechanical and electrical powers. was developed to minimize a general quadratic cost function of tracking errors and control efforts [4-41]. Neural networks in control has mainly used Model Reference Adaptive Control (MRAC) [4-36–40]. Difficulties in a power system stabilizer design come from the handling of nonlinearities and interactions among generators. This results in a hybrid of feedback and feedforward neurocontrollers in parallel. Recently.certain degree by training the input/output relationships obtained in the full system operation. a Generalized Backpropagation-Through-Time (GBTT) algorithm was developed to train the feedback controller. neural networks have been investigated for power system stabilizing control. a general purpose controller. an Optimal Tracking NeuroController. y ( k − n + 1). From this point of view. handling the nonlinear power flow properly is the key to the PSS design for a multimachine power system. i.⋅ ⋅ ⋅. A feedback neuro-controller (FBNC) is constructed to generate feedback control input. (5) where yref is a reference output.. and trained by a Generalized BTT (GBTT) algorithm to minimize the quadratic performance index.. {x(i ) } to a r vector output with an appropriate dimension defined as x(i −1) = ( x(i −1) . 4-5 shows an architecture for the optimal tracking neuro-controller for a nonlinear plant. x( x − p ) ) . The control objective is to improve system damping by using a supplementary excitation control applied to the second generator. The power system has sustained low-frequency oscillations due to disturbances. The power system consists of three power plants: two are thermal units and one is a hydro unit. This quadratic cost function or performance index not only forces the plant output to follow the reference. 4-12 . where p = n for the output variable y. y ref y (k) u (k-1) ufb (k) u (k) Σ Feedback Neuro-Controller u (k) ∆ u (k-1) y (k) Plant Dynamics y(k+1) Tapped Delay Operator ∆ y(k) yref Feedforward Neuro-Controller u ff u (k-1) y(k+1) Neuro-Identifier u (k) Fig. A feedforward neuro-controller (FFNC) is constructed to generate feedforward control input corresponding to the set point.J= 1 N ∑ (Q( yref − y ( k 2 k =1 + 1)) 2 + R (uref − u( k )) 2 ) . This network is trained to emulate a plant dynamics and to backpropagate an equivalent error or generalized delta [4-36] to the controllers under training. the tapped delay operator ∆ is defined as a delay mapping from a sequence of scalar input.. uref is the steady-state input corresponding to yref . The study power system. Fig.. 4-5 Block diagram for the optimal tracking neuro-controller. but also forces the plant input to be close to the steady-state value in maintaining the plant output to its reference value. and Q and R are positive weighting factors. In the figure. and p= m-1 for the input variable u. An independent neural network named neuro-identifier is used when the above two neuro-controllers are in training mode. An optimal tracking neuro-controller (OTNC) is designed with two neuro-controllers in order to control a nonlinear plant that has a non-zero set point in steady-state [4-41]. The neuro-controller is applied to a 5-bus power system [4-42] to stabilize low-frequency oscillations (Figure 4-6). and trained by the well-known error Backpropagation algorithm. x(i −2) . Three of the seven input nodes are for its output history. and two for ∆ω ( k ) .0 1 0.2(0.06+j0. and the power flow.2+j0.02) 0. angle.12(0.15 L4 L3 3 4 0. Therefore. a 9th order model for thermal plants and a 10th model for the hydro plant are used to represent the nonlinear characteristics and the low-frequency oscillations in simulations. The power system with 3 generators and 5 buses.0+j0.75+j0.6+j0.04+j0.06+j0.15 Voltage : 1. u ( k ) . ∆Pe( k −1) . an input layer with 6 input nodes and an output layer with one node.02+j0.67 G1 G3 1. Training of the neural networks. Typical IEEE governor and turbine models are used: TGOV1 (2nd order) for the thermal plant and IEEEG2 (3rd order) for the hydro unit [4-43]. ∆δ ( k ) .4-6. As a result.25 Hz.03) 0.1 L2 G2 0. frequency. The Neuro-Identifier consists of one hidden layer with 40 nodes. The Neuro-Controller has one hidden layer with 40 nodes.18 (0. it’s assumed that the exciter can be approximated as a second-order model. The Optimal Tracking Neuro-Controller is applied to Generator 2 to provide supplementary excitation signal as a power system stabilizer. The IEEE exciter and voltage regulator model EXST1 (4th order) is used for this study on which supplementary excitation control input is to be injected.( Thermal Plant ) ( Thermal Plant ) 0.4+j1. are all deviations from the respective references.01) 0. two are for control input history. During the lowfrequency oscillation in the range of 1~2 Hz. u ( k −1) . Three of the six input 4-13 .03(0.02) 2 0.18(0. the feedforward controller was not used.015) Power Flow Power: 0. ∆Pe( k ) . This allows at least twenty sampling points in a cycle of the low-frequency oscillation under 1. Since the output variables.02) 0.01+j0.67+j0.2 L5 Install PSS Fig. an input layer with 7 input nodes and an output layer with one node.5 ( Hydro Plant ) 5 0. The discrete-time training patterns are obtained with the time step of 0. The training patterns of the Neuro-Identifier are generated by the power system simulations starting from the steady-state initial value in a wide range of operating conditions and randomly generated control inputs history within the conventional PSS operation region.8+j0. ∆Pe( k −2) .24(0.04 sec in simulation.06+j0.08+j0.06(0.025) 0. the Neuro-Identifier is constructed to emulate the power flow dynamics as a third-order model that includes the dynamics of exciter and the excitation field voltage. Then. Figure 4-7 shows the speed deviation of Generator 2 for a three-phase ground fault at midpoint of a half the line 4–5.75[p.5 [Hz] 0. ∆Pe( k ) . generating power) and Figure 4-9 shows speed deviation for a heavy loading condition (1.).] ) 0. The cost function for the N-step ahead controller is set with the weightings Q = 1. Training of the Neuro-Controller is done in two phases. The speed deviation of generator 2 for the line fault disturbance in a normal load condition. Speed-dev. and the Neuro-PSS. The figure compares the cases without a control and with supplementary excitation controls by the conventional PSS.02. Training is carried on with a gradually increasing N until it reaches 8 so that the system can be controlled for a longer duration of time. ( 0. STAB4 [4-43]. 4-7. It takes about 30 minutes on an IBM-PC 486 computer to train two neural networks: the Neuro-Identifier and the Neuro-Controller. Figure 4-8 shows the speed deviation for the same disturbance when the power system is in a light loading condition (0.0 p.u. 4-14 . of the 2-nd Gen. The ITE performance of the conventional PSS shows larger variation to loading conditions because the parameters in the STAB4 were optimized in the normal loading condition. training is done with a small N ( = 3) since in the beginning it has little knowledge of control.2 sec. Observations from the table show that the Neuro-PSS works very well judging from the ITE performance in both the heavy or the light load compared to the normal load condition.0 and R = 0. The figures show that both controllers work very well judging from small swings with large damping. To avoid oscillation during training stage.5 p. The performance of the controllers are compared in Table 1with the integral-time-error (ITE) computed with the cost function (5). weight parameters in the Neuro-Identifier are corrected with the average of corrections calculated for ten patterns. which cleared after 0.3 0.5 0 1 Without Control 2 3 Time [Sec] STAB4 4 5 6 Neuro-PSS Fig. First. Comparison of the control results. ∆Pe( k −1) . training is carried on with N fixed at 8.u.u.1 -0. ∆δ (k) .3 -0. one is for previous control input u ( k −1) and two are ∆ω ( k ) . A small number of steps prevents the system from diverging. ∆Pe( k −2) .1 -0.nodes are for output history. 24 sec.5 [p. ( 0.1 -0.3 0. of the 2-nd Gen. increase at 0.5 [Hz] 0. and cleared at 1. of the 2-nd Gen.44 sec when the power system is in the heavy loading condition.4 [Hz] 0.1 -0.4 0 1 2 Without Control 3 Time [Sec] 4 5 6 Neuro-PSS STAB4 Fig. 4-15 .3 -0.Speed-dev. The speed deviation of generator 2 for the line fault disturbance in a light load condition.u.15 p.] ) 0. The figure shows that the Neuro-PSS works very well judging from small swings.5 0 1 Without Control 2 3 Time [Sec] STAB4 4 5 6 Neuro-PSS Fig. 4-8. decrease at 0. Speed-dev.2 0.u.0[p.0 -0.u.2 -0. ( 1.] ) 0. Figure 4-10 shows the speed deviation for other disturbances coming from stepwise loading conditions: 0. The speed deviation of generator 2 for the line fault disturbance in a heavy load condition.96 sec. 4-9. u. i.3 0. and they are more general than decision trees. Another advantage of decision trees is that when you have training data with maybe 250 variables in each input vector. Table 1. 4-16 . ITE performance evaluation for the line fault disturbance Loading 0.75 p.81 30. But when a problem can be reduced to a small number of choices.03 100(%) 22.3 Decision Trees for Power System Control Decision trees (DTs) are learn-by-example classifiers which are particularly well suited for discrete event control [4-44.0[p.6(%) 4.19 8.92 8. Artificial neural networks (ANNs) can also be used for discrete event controls.67 27. Neural networks can associate their input vectors with a continuous range of output values.83 12.5 [Hz] 0.24 100(%) STAB4 1. we can see which threshold criteria were met.u.4-45]. then decision trees have important advantages. the DT training algorithm usually selects a much smaller subset.0(%) 2.89 15. 0. Without Control 6. The speed deviation of generator 2 for the load change disturbance in a heavy load condition.1 -0. why the case was classified and how the outcome would have changed if certain input variables had been different. 1.. to be used for classification.5 p.Speed-dev.2(%) 2. whereas decision trees are only suited for classification problems having a small number of output categories such as stable/unstable.] ) 0. 4-10.6(%) 1. perhaps 25 variables. of the 2-nd Gen. The decision trees reported in [4-46–50] require only a few minutes to train whereas neural networks usually require much more computation for the training. When a particular case is classified by a DT.7(5) 1.04 100(%) 12.3 -0.u.0 p.1 -0.7(%) Neuro-PSS 1.5 0 1 2 3 4 5 6 Time [Sec] Without Control STAB4 Neuro-PSS Fig.u. ( 1.e. The DTs for real-time remedial action control [4-46–50] could be trained either from offline simulations or from on-line simulation tools that are being developed to perform online DSA. Power system protection and large-scale stability controls have traditionally relied upon off-line simulations that are transformed into decision rules by engineers.800 transient stability simulations for 50 randomly generated operating points. 4.800 contingencies. An emerging possibility is to train the classifiers using on-line DSA [4-54.3. Simulated generator angle measurements were taken over an eight cycle window immediately after fault clearing. Training sets are extracted from off-line simulations of critical contingencies applied to a large number of pre-fault equilibrium conditions. three4-17 . it was proposed to train DTs off-line to handle a specific range of operating conditions. From this snapshot immediately after fault clearing. In simulations of a 176 bus model of the western U.4-55]. The resulting controls are custom tailored to the current operating conditions. These on-line simulations can already be used to program discrete event controls such as generator tripping (see Chapter 5). In that work. Accuracy in excess of 95% was obtained for the 40. remedial action control. and then two velocities and one acceleration were computed from the angle measurements of each generator. Training sets were created by simulating three-phase faults of various duration on all the buses and transmission lines.2 Decision trees for real-time transient stability prediction The earliest research on DTs for real-time control investigated prediction of angle instability using synchronized phase angle measurements from all 10 generators in the New England 39 bus test system [4-46.4-47].S. These DTs are designed to perform on-line dynamic security assessment (DSA). One way to use DTs for real-time control is to train a DT to predict whether loss of synchronism will occur without control and train another DT to predict whether loss of synchronism will occur with some particular control. The input vector contains various static parameters from the prefault equilibrium point such as key generation and transfer levels. a combination of generator tripping at Palo Verde and load shedding at Tesla and Vaca-Dixon was found to stabilize long duration three-phase faults for five transmission lines in the Arizona area [4-48]. The same simulation capabilities could generate the training sets for DTs that perform real-time. A test set of 500 random duration.1 Relation of angle stability decision trees to on-line dynamic security assessment Decision trees have been developed for on-line preventive control and also for real-time remedial action control.. Classifier training algorithms can perform the same tasks using large numbers of simulations and predictor variables. The first research and industrial use of DTs for angle stability control was in the area of on-line preventive control [4-51–53].3. The desired output reflects whether any of the contingencies caused instability for that equilibrium. Robustness to variations in the operating point was investigated using a test set of 40. The DTs are then used on-line to predict the vulnerability of the power system in its present equilibrium state to those contingencies.4. Three successive measurements of the generator angles were used. the decision trees correctly predicted whether loss of synchronism would occur in the next four seconds with over 97% accuracy. the designers learned to differentiate between stable and unstable swings based on their trajectories in the R-Rdot phase plane.3 Decision trees for response-based control Prior to 1996. in the development of the R-Rdot out-of-step relay [4-56. Instead of using only the immediate post-event electrical measurements. More recently.728 data points extracted from 168 transient simulations on the 176 bus model. one cycle fault followed by loss of the Pacific DC Intertie (PDCI). For example. The 168 contingencies in the training set contained 6 different fault scenarios for each of 28 transmission lines: one-cycle fault. decision trees have been adapted to continuously follow the measurements and select control action as soon as the need becomes apparent [4-50. If control is applied when the DTs predict stable with control and unstable without.phase faults on these lines without control contained 232 stable cases and 268 unstable cases. The training sets can be generated using industry standard power system models. Decision boundaries were then drawn to classify new swings as either stable or unstable and to order circuit breaker operation as appropriate. apparent resistance R and its rate of change Rdot were plotted for both stable and unstable transient events. The remaining 5 cases had very long fault durations and hence were too serious to control.4-58]. a DT was trained to associate each pair of R and Rdot measurements with whether the angle across the PACI exceeded 90 degrees when the measurements were taken. The controller operated in all 268 of the unstable cases. and never failed to trip on an unstable event. and stabilized 263 of them. correctly refrained from tripping on 704 events. In addition to achieving response based control. four cycle fault. Using 28. Using large-scale simulations. the research on DTs for real-time control had assumed there would be some way to detect that an event had just occurred so that the immediate post-event measurements could be fed into the decision tree. 4-18 . This response-based operation effectively turns the classifier approach into a natural generalization of the way engineers determine relay settings and discrete-event control laws. Decision tree training algorithms can draw decision boundaries in phase planes as well as in higher dimensional spaces. three cycle fault. six cycle fault. Each simulation in the training set was three seconds long. 4. each involving two of the 28 study lines. these DTs also respond appropriately to single-phase faults.3. The resulting DT tripped correctly on 70 events. response-based DT control is achieved by using every time sample in the simulation for an input-output pair. All faults were three-phase short circuit to ground with the faulted line removed at clearing time. The apparent resistance was measured at Malin substation near the electrical center of the Pacific AC Intertie (PACI) in order to detect loss of synchronism across the PACI. The test set contained data extracted from 784 simulations which were five seconds long. and one cycle fault followed by loss of the Intermountain Power Project (IPP) DC line. then 215 of the 232 stable cases have no unnecessary control intervention. The R-Rdot relay provides a good demonstration of DTs for response-based control. tripped incorrectly on 10 events. The remaining 17 stable cases had control intervention without adverse effect.4-57]. 756 of the test set events were double contingency outages. A circuit breaker controlled by this DT will be programmed to trip and stay open once the DT outputs “trip. T J =∫ 0 ∑i M i (δ i − δ coa ) 2 dt 4-19 . If.600 Yes No No Trip Yes Rdot < . 4-11. The resulting DT will only trip if the trajectory enters a region where stable trajectories almost never enter.Specifying misclassification costs during the training has been particularly helpful for building DTs to perform response-based control. For training the DT shown in Figure 4-11. For a truly unstable event. once the breaker opens it must stay open.4-50]. If a control makes the difference between stability and instability. then it still has the option of tripping later. Decision tree for an R-Rdot out-of-step relay. When instability is not an issue and the goal is to improve the dynamic performance. it’s necessary for a computer algorithm to determine which control to assign each case in the training set.143 Yes No Trip No Trip Trip Rdot < .4 Decision trees for improving dynamic performance Decision trees can perform response-based discrete-event control to improve the dynamic performance of stable transient events [4-49.64 Yes R < 21 Trip Yes No No Yes Rdot < . then the choice is clear. There will always be an area of uncertainty between when the DT should trip versus not trip. Yes Yes No No R<0 No Trip R < 38 No Trip Rdot < . the DT fails to trip on a case where the intertie angle has in fact exceeded 90 degrees. 4. the need to trip should become more obvious over time and it would be desirable to train the DT to wait until the need to trip is nearly certain. an objective measure of the post-event behavior must be used.13 R < 17 No No No Trip No Trip Fig. This behavior can be obtained by assigning a high misclassification cost to false trips. the misclassification cost of false trips was set 50 times higher than the misclassification cost of failures to trip.3. however. In order to automatically train a classifier to associate the incoming measurements with an appropriate discrete-event control. The following objective function is used to calculate the severity of simulated transient events with and without control.” Hence there is no remedy for a false trip. Each contingency was simulated with and without a 500 MW fast power increase on the IPP DC line immediately after fault clearing. 4-4 CIGRÉ TF 38-06-06 on Artificial Neural Network Applications for Power Systems.) Expert System Applications to Power Systems. “Neural Network Applications in Power 4-20 . a 500 MW IPP DC ramp in response to the initial events would have prevented the cascading outage that occurred on December 14. Vol. The DT ordered a 500 MW fast power increase at some point in 44 of the 62 simulations and had a positive effect in 42 of the 44 simulations it tried to control. (eds. Laughton. References 4-1 R. pp. 210–252.” IEEE Transactions on Power Systems. and H.This performance index is like the weighted sum squared “error” comparing the simulated swing curves to a hypothetical “ideal” trajectory where all the generator angles are constant with no angle differences. PAS-3. The decision tree was tested on three cycle. The sum does not have to contain all the generators in the model.” in T. Y Kohno. T. A sampling on the order of 10–100 of the larger generators distributed throughout the power system is sufficient to have J be a fairly good numerical measure of the amount of interarea oscillation following a disturbance. Dagmar Niebur (convener). “A Voltage Control Expert System and Its Performance Evaluation.1% resulting from the DC fast power change. the improvement from the DT controller is roughly 2. Sakaguchi. Most of the stable events had performance index scores between 40 and 80. 4-2 C. Dillon and M. Dillon (convener). C. Marathe. Controls that reduce J tend to have the strongest smoothing and stabilizing effects on the post-event oscillations. Fujiwara.4/60 = 4. and a DT was trained to predict from real-time phasor measurements whether the numerical improvement in dynamic performance would exceed a threshold [4-50]. December 1992. 302–307. The average performance index improvement for the 39 stable contingencies was 2. Between two simulations. 1994 by reducing overloads which caused some of the transmission lines to trip [4-49. pp. “An Intelligent Load Flow Engine for Power System Planning. Performance index calculations applied to the large-scale simulations of the initial December 14 events showed an improvement of 4.7. In addition to improving dynamic performance. Prentice-Hall. pp. three-phase faults and five cycle single line to ground faults applied to the same 31 transmission lines used in the training set. 142–156. 1989. August 1986.4-50]. Fifty-one of the 62 simulations were stable for the first two seconds. Suzuki.” Electra.4 and the maximum improvement was 4. and 39 of the 44 DT operations occurred during stable events. the performance index can also be used to determine powerful combinations of discrete event controls for stabilizing strongly unstable events [4-48]. Using 60 as a rough estimate of the average score for stable cases. For comparison. “A Survey on Expert Systems for Alarm Handling. Liu. everything is held fixed except for some control action that needs to be evaluated.0%. T. 4-3 CIGRÉ TF 38-06-02 on Expert Systems Applied to Monitoring and Alarm Processing in Control Centers. Tomsovic. H. Decision trees were trained to improve dynamic performance using data extracted from 93 transient simulations on the 176 bus model. and K. 3. “Advanced PID type Fuzzy Logic Power System Stabilizer. Hassan. Vol. Journal of Engineering Intelligent Systems. Universal Approximation and Orthogonal Least-Square Learning. Moon. M. 1 No. Chow (eds.125–131. Park. 338–353. 4-11 D. “A Self-Organizing Power System Stabilizer Using Fuzzy Auto-Regressive Moving Average (FARMA) Model. Vol. A. 807–814. Wang and J.” IEEE Transactions on Energy Conversion. 4-13 T. pp. M. pp.” in K. April 1995. 4-12 M. A. Columbus. 1997. 1965. Tomsovic and M. Niebur and O. Y. September1994. April 1995.” Electra. 3. tutorial publication of the 1997 IEEE Power Industry Computer Applications Conference. September 1992. Hiyama. IEEE Transactions on Energy Conversion. 4-7 CIGRÉ TF 38-06-03.” IEEE Transactions on Energy Conversion. pp. pp. pp. September 1991. May 11–15. pp. “Fuzzy Basis Functions. and H. 50–75. OH. No. 5. pp.Dillon (convener). Hope.” Int. OH. “Applications of Artificial Neural Networks to Power System Security Assessment. 4-5 CIGRÉ TF38-06-06 on Artificial Neural Network Applications for Power Systems. 3. Tutorial on Fuzzy Logic Applications to Power Systems. Chow (eds. 11. Lee. P. June 1996. 2. on Practical Applications of Expert System. “Summary of Final Report. Y. pp. Vol. 4-9 L. Lee.”Inf.). Tomsovic and M. 514–520. S.Systems.” to appear in Electra. “Fault Diagnosis in Electric Power Systems through AI Techniques. Nagashima. 6. D. No. tutorial publication of the 1997 IEEE Power Industry Computer Applications Conference. Malik and G. Niebur. El-Sharkawi. U. Malik. Vol. Y. Satoh. D.-W. 4-15 Y. 4-6 CIGRÉ TF 38-06-02 on Expert Systems Applied to Monitoring and Alarm Processing in Control Centers.” in M. Fischl. Oniki. 76–101. 9. 133– 158. O. 407–413.. Niebur (convener). No. No. M. 4-16 R.” Electra. Vol. 1. May 11–15. Tutorial on Fuzzy Logic Applications to Power Systems.” in K. S. Zadeh. March 1996. 159. 1997. “Artificial Neural Networks for Power Systems. pp. and K. 442–448. “Evaluation of Advanced Fuzzy Logic PSS on Analog Network Simulator and Actual Installation on Hydro Generators”. “Stability of Fuzzy Controllers. No.). “A Fuzzy Logic Based Stabilizer for a Synchronous Machine. 159. and D. Columbus.” IEEE Transactions on Neural Networks. Kugimiya and H. C. 3. No. 4-14 T. Mendel.” IEEE Transactions on Energy Conversion. Control 8. Hiyama. and M. 4-10 K. December 1993. 4-8 L. M. 4-21 . 11. “Fuzzy sets. “Control Applications of Fuzzy Systems. No. T. El-Sharkawi. Vol. and R.. 1990. “Artificial Neural-Net Based Dynamic Security Assessment for Electric Power Systems. 3. Seattle. 4-18 A. Niebur and A.” IEEE Transactions on Power Systems. “Evaluation of Neural Network Based Real Time Maximum Power Tracking Controller for PV System. No.-H.” IEEE International Symposium on Circuit and Systems. Vol. 4-22 . No. Piras. Leics. Aggoune. Imakubo. May 1992. J. pp. J. B.” IEEE Proceedings of 1989 PICA. New Orleans. 4. Salt Lake City. pp. Sobajic and Y. July 1991.S.” IEEE Transactions on Power Systems. 5. Dolce “Real-time Security Monitoring of Electric Power Systems Using Parallel Associative Memories. 252–258. Sobajic and Y. February 1989. El-Sharkawi. A. 104–127. M. Seattle WA.” IEEE Transactions on Power Delivery. UK. Kouzuma. 94. J. 425–433. 2. 543–548. 4.” IEEE Transactions on Power Systems.A. H. M. pp. No. Y. 9.. pp. King. 4-22 S. Koizumi. 4-25 M. M. February 1989. 10. J. M. U. and Tsuji. Vol. and T. “Security Monitoring Systems Including Fast Transient Stability Studies. J. “Using Artificial Neural Networks for Load Shedding to Alleviate Overloaded Lines.-H. Dillon and D. and Y. T.Niebur (eds. “Artificial Neural-Net Based Dynamic Security Assessment for Electric Power Systems. Marks II. January 1994.” IEEE Transactions on Energy Conversion. May 1989. pp. May 1–3. Novosel and R. 59–64. 4-21 D. September/October 1975. A. 1995. Vol. 220–228. Germond. No.) Neural Net Applications in Power Systems. Pao. 865–872. Hiyama.” First International Forum on Applications of Neural Networks to Power Systems. 4-20 D. Imhof. 1789–1805. Vol. 4-27 T. Vol. Saito. A. “Preliminary Results on Using Artificial Neural Networks for Security Assessment. 11. Udo. Pao. Niebur (eds. Louisana. 4-17 D. Utah. and J. D. 1996.-H. Buchenel. S. Vol. 4-23 O. Washington. 4-26 T. Sobajic. pp. No. 4.” IEEE Transactions on Power Apparatus and Systems. September 1995. Sato. Mukae. 7. Damborg. 2929-2932. K. 4-24 D. “Heterogeneous Artificial Neural Networks for Short Term Load Forecasting. No.” Proceedings of IEEE PICA ‘95. Park. “Power System Static Security Assessment Using the Kohonen Neural Network Classifier.) Tutorial Course on Applications of Artificial Neural Networks to Power Systems. Jaccard. pp. IEEE Catalog Number 96 TP 112-0. J. Weerasooriya and M. CRL Publishing Ltd. “Use of Karhunen-Loève Expansion in Training Neural Networks for Static Security Assessment. Pao. Germond. K. 4-19 D. 4. El-Sharkawi. 220–228. C. McCalley. Treinen. The MIT Press. Kanetkar and S. 940–947. Papalexopoulos. B. pp. pp. McCalley. Vol. McCalley.” Proceedings of the North American Power Symposium.” IEEE Transactions on Power Systems. 78. July 1992. Vol. Q. 4-35 V. Neural Networks for Control. S. Ku and K. “Diagonal Recurrent Neural Network for Dynamic System Control”. October 1997. 1463–1468. G. Miller. November 1994. “Security boundary visualization for systems operation. S. 39–44.Results in the Lemnos Power System. Zhou.” Proc. 4-37 P. Yamada and T. Jacquemart. Parthasarathy.6. C. “Identification and Control of Dynamic Systems Using Neural Networks. Wang. S. Mitchell. Van Cutsem. Werbos. D. pp. and A. pp. Sutton and P. 10. 4-32 J.” Proceedings of IFAC Symposium on Control of Power Plants and Power Systems. J. Chicago. 595– 601. Laramie. “Present Day Power System Stability Analysis Methods in the Western United States. 4-29 P. Iowa. T. No. pp. et al. 4-23 . pp. 1985. T. December 1995. J. Ames.” IEEE Transactions on Neural Networks. 4-33 J. Zhao. “Dynamic Security Assessment using Pattern Recognition. Neural Networks and Decision Trees . “Compact Representation of Power System Security— a review. pp.-A. Ranade. May 1997. R. of the Conference on Rough Sets and Soft Computing (RSSC'94). Shanahan and S. 4-34 Y. Krause. 4-40 C. 4-36 W. 37–45. 1. 4-28. Vol.” Proceedings of the International Symposium on Power System Stability. A. 1990. G. Mexico. 1992.” Proceedings of the 1995 American Power Conference. Van Acker. Narendra and K. October 1990.” Electric Power Systems Research.1550–1560. “Statistical Approaches to Dynamic Security Assessment: The Database Generation Problem.. C. Lee. pp. Farmer. Zhou. Vol. “Neuro-controller Using Autotuning Method for Nonlinear Functions. 1995. Peças-Lopez. May 13–15. WY. “Data Generation using Automated Security Assessment for Neural Network Training. 12. 28–65. 4-38 K. “Backpropagation Through Time: What it does and how to do it. and M.4-28 J. 2. March 1990. S. 4-30 J.” IEEE Transactions on Neural Networks. Naumann. Y. Pruvot. 3. “Rapid Transmission Capacity Margin Determination for Dynamic Security Assessment using Artificial Neural Networks. 144–156. Vol. No. Cancun. L. J. D. IL. “Evaluation of Simultaneous Transfer Capabilities.” Proceedings of the 29th North American Power Symposium. 34. Wang. 4-31 R. and P. Werbos. Yabuta.” Proceedings of IEEE. 4-39 T. pp. April 1995. January 1995. R. T. Vol. Wehenkel. IEEE Transactions on Neural Networks. S. and J. “An optimal tracking neuro-controller for nonlinear dynamic systems. E. Mansour. February 1994. Brown. pp. Vol. Van Cutsem. 9. T. and K. E. W. Ribbens-Pavella. 1. pp. Taylor. Breiman et al. Rovnyak. J. 2. Mechenbier.” IEEE Transactions on Power Systems. Choi and K. 1099–1109. 4-49 S. October 25–29. PWRS-9. “Decision Trees and Transient Stability of Electric Power Systems. California. Pavella. S.” Precise Measurements in Power Systems Conference. M. pp. Chang. Th. Kretsinger. 4-42 Stagg and El-Abiad. S. Kretsinger. 4-47 S. Wehenkel. May 1997.” IEEE Transactions on Power Systems. C-26. M.” IEEE Transactions on Power Systems. 4-48 S. Virginia. and D. 1968. S. M. Thorp. B. Rovnyak. C. A. Y. 5. Hydro’s On-line Transient Stability Assessment 4-24 . 27. Kostyniack. PSS/E Program Operation Manual. R.. 8–10.A Case Study. No. E. Garrett. Wehenkel. pp. E. August 1994. pp. and B. S. Wadsworth. 4-44 L. pp. Arlington. “B. Vol. Classification and Regression Trees. 1. 1996.” Control Engineering Practice. 1977. B. Discrete-Event Control. M. 789–800. Belmont. 4-54 Y. “A Recursive Partitioning Decision Rule for Nonparametric Classification. Power Technology Inc. “Decision Tree Based Transient Stability Method . 4. pp. 404–408. 1984. S.1417–1426. Rovnyak. No. 4-51 L. No. E. 4-43 T. Corns. R. Vol. Pavella. Lee. and J. Vaahedi.. “Decision Trees for Real-Time Transient Stability Prediction. C. M. No. H. 1991. 4-53 L.” Fault and Disturbance Analysis & Precise Measurements in Power Systems Conference. Rovnyak. S. “Parallel Decision Trees for Predicting Groups of Unstable Generators from Synchronized Phasor Measurements. Arlington. Vol. Heilbronn. Vol. September. No. K. 91–99. 4-46 S. 1997. 115–134. J. Vol. Euxibie. E. Nov. pp. E. 1983. Wehenkel. Athay.” Automatica. 5. EPRI Final Report TR-107908. 1993. Thorp. Brown. and M. Cheung. M. No. 3. Friedman. W. October 31. 387. 7. C. Taylor. Thorp. “Plans to Demonstrate Decision Tree Control Using Phasor Measurements for HVDC Fast Power Changes. “Performance Index and Classifier Approaches to Real-Time. McGraw-Hill. and J. M.” IEEE Transactions on Neural Networks. D. Park. Thorp. W. Virginia.4-41 M. November 1989.” IEEE Transactions on Computers. Computer Methods in Power System Analysis. 1. 459-469. Demaree. 1995. “An Artificial Intelligence Framework for On-Line Transient Stability Assessment of Power Systems. Y. 4-50 Synchronized Phasor Measurements for the Western Systems Coordinating Council. 4-52 L. 4-45 J. W. N. No. 1. 3. 83. 3. G. K. Bushnell. October 1998. July 18–22. Cresap. 11. pp. pp. L. 35–39. 2. March 1983. “A New Out-of-Step Relay With Rate of Change of Apparent Resistance Augmentation. No. 4-57 J. Taylor. T. 1463–1472. D. 4-55 H. 10. 88–91. pp. Vol. pp. Edmonton.” Proceedings of IEEE/PES 1999 Summer Meeting. Morita.” IEEE Control Systems. Kokai. Laughlin. Hill. Haner.M. Bonissone. 241–253. Vol. Analysis. W. pp. “Development of Transient Stability Control System (TSC System) Based on OnLine Stability Calculation. A.” IEEE Transactions on Power Apparatus and Systems. A. M. “Using Fuzzy Logic in Control Applications: Beyond Fuzzy PID Control. Vol. Mittelstadt. Ota. and Post-Processing. 10-15. et al. W. Chiu. Sheng. Taylor. M. “Industrial Applications of Fuzzy Logic at General Electric. Abramovitch and L. “Report on the Fuzzy versus Conventional Control Debate.. April 1986. Omata. No. pp. pp. March 1995.” IEEE Transactions on Power Systems. and C. and Y.” IEEE Transactions on Power Delivery. “Experience with the R-Rdot Outof-Step Relay.” IEEE Transactions on Power Systems. P. Vol.(TSA) Model Development. H. pp. Kitayama. 4-60 P. L. 4-59 D. No. August 1996. PWRD-1. Ito. 450–465. PAS102. 4-58 S. 631–639. February 1995. K. July 1999 (debate between Michael Athans and Lotfi Zadeh).” IEEE Control Systems. 1999. “Using Measurements and Decision Tree Processing for Response-Based Discrete-Event Control. 3. and R. Vol. 4-56 C. 100–104. Fukushima. 4-25 . 4-61 S. Haner. Y.” Proceedings of the IEEE. J. Rovnyak and Y. No. Y. The limits derived are conservative.5-6. • Performance. Transfer limits are determined by selecting extreme system conditions and simulating critical contingencies. preventive or corrective remedial actions are designed. Preventive actions are applied in the pre-contingency system so that after any credible contingency the system remains secure. This chapter describes on-line dynamic security assessment methods as part of the remedial stability control determination. 5. To ensure that a system remains dynamically secure. • Accuracy. on-line dynamic security assessment consists of the following four elements: • preprocessing .5-19. The accuracy of DSA is of ultimate importance to ensure the dynamic security of a power system. and capacitor bank or reactor switching.5-21] a few of which have found their way to real system implementation [5-1–3.Chapter 5 Integration of Dynamic Security Assessment and Stability Controls Dynamic security assessment (DSA) or transient security assessment (TSA) determines a system’s ability to survive contingencies with a safety factor (margin). and describes in detail its different components. Traditionally. Other performance requirements for DSA include flexibility in data input/output and good user interface. In order to achieve the best computation speed. Examples include restrictions on interface flows. advanced techniques in software and hardware design must be used. and total generation out of a plant. Transient security assessment for arming of generation tripping stability control is described. Recently. since they are based on extreme system conditions. As described in the following subsections. on-line dynamic security assessment tools have been developed [5-1–7. These tools differ in the methodology but they share the same concepts and fundamental blocks. Stability control is made adaptive based on the on-line security assessment. In particular.1 On-Line Transient Stability Assessment Design An on-line DSA tool should meet the following requirements: • Reliability. angle differences across a particular interface. The general trend is that on-line DSA should give comparable results with the best off-line study tools for a given system model.5-7]. The processing speed of DSA is often critical in meeting the requirements for on-line real-time or near real-time operations. as well as for disturbances that may include autoreclosures and other complex switching actions. Both the hardware and software of on-line DSA should perform reliably under all feasible system operating conditions. preventive and corrective dynamic security measures have been developed from numerous off-line simulations. Corrective remedial actions (stability controls) are those taken following a contingency. Examples include generator or load tripping. it should accommodate detailed models for system components. There are other difficulties associated with measurement accuracy. expert systems [5-2. bus power flow states are known.5-7. and contingency screening and ranking. with high quality bus voltage magnitude and angle measurements. unbalance operation. While all these methods can be used for contingency screening and ranking. It’s impossible to assess all the credible contingencies within the confines of available computational resources and required response times. In addition to power flow data. A major impediment to DSA for interarea stability problems in large interconnections is the difficulty of state estimation to obtain the on-line power flow base case.5-30]. etc. • False alarm: a stable contingency that is identified as an unstable one (critical one) 5-2 .1 Preprocessing The task of preprocessing includes static state estimation. Therefore. The common requirements of candidate contingency screening and ranking methods are high speed and accuracy of the final results.• security assessment • post-processing • process control and integration 5. housekeeping for base case development. For a particular control center. the final limit computations should be done using more accurate methods. External network models for on-line DSA is obtained by selecting from a number of previously stored dynamically reduced system models [5-1. Misclassification consists of two components of False Alarm and False Dismissal as described below. the main difficulty is with the external network model. The performance of a contingency screening and ranking method can be evaluated in terms of its misclassification. Conceptually. the list of credible contingencies has to be reduced to make it manageable by the security assessment module. Considerable inter-utility data exchange is required. 5-26]. other data required for DSA may also need to be updated for a new system snapshot. State estimation can be improved by high quality digital measurements from throughout the interconnection. network parameter uncertainty. The contingencies may also need update when the network topology or system operation condition changes. This will facilitate base case initialization and helps maintain the base case within certain size limit so that the computation speed requirements can be met. Synchronized positive sequence phasor measurements are valuable [5-22. neural networks [5-8. Alternatively dynamic reduction techniques can be used in real-time to develop the external model.5-25. For instance.5-2].5-8.5-10]. or indices derived from energy properties or fast time domain simulations using simplified models [5-6. [5-24].1. The contingency screening and ranking method could be based on transient energy functions [5-10–17].5-9]. extended equal area criteria [5-17–20]. the settings of a PSS for a pumped storage generation unit may need adjustment for the different modes of operation of the unit. 5. small-signal stability in the form of sustained or growing oscillations in part or all of the system. Commercial-grade software is available that can normally be customized for a utility’s required modeling and disturbances.. This is still an area with room for research. Normally. The minimum of the corrected kinetic energy.g. unstable). In the following.2 Security assessment Detailed time domain simulation is the most reliable security assessment approach [5-1–7.e. One approach based on the time domain simulation technique is to obtain an estimate of the critical mode by post-processing simulation results. the margin is calculated from the value of the corrected kinetic energy at this point. 5. As power systems operate in more and more stressed conditions. If the minimum is greater than zero (system unstable). kinetic energy and corrected kinetic energy. at Tkemin1. the original concept [5-2] is described below.. or the ability of the system to maintain synchronism after a credible contingency. Second-kick method [5-2. • Determining the transfer limit or preplanned stability control actions (e. arming of generator tripping).• False Dismissal: an unstable contingency that is identified as a stable one (non-critical one) An acceptable contingency screening and ranking method should have zero false dismissals and a very low number of false alarms. DSA assesses transient stability of a power system.5-29]. may become critically restricting to the system operating limits. • Determining the sensitivity of the margin to key variables (transfer limit and generation tripping).1. This allows on-line security assessment with unlimited modeling capabilities capable of handling a full-scaled power system. The requirement to address this type of stability problem calls for an efficient and reliable method to compute the critical mode of the system. • Determining the degree of stability or instability (margin). Detailed time domain simulation is performed with calculation of potential energy. The second kick method was based on energy concepts for determining stability margin and other useful information from the simulations. 5-4]. several methods that have been used in the post-processing stage are described. a second fault (second kick) which is 5-3 .3 Post-processing Online implementation of time simulation requires a built-in intelligence for the following: • Assessing the system dynamic performance (stable. Kemin1 is identified.5-19. If the system is stable (corrected Kemin = 0). i. This is further described in the following section.1. It was inspired by the hybrid method [5-3]. another form of angle stability. Although there are different implementation methods available for this algorithm [5-5. No modeling assumptions are made and no analytical equation is used to calculate potential energy. after the contingency. This has already happened in some parts of the North American interconnected power systems. 5-21]. each of which is characterized by a special version of the method: • Static method 5-4 . This point also reflects the crossing of the potential energy boundary surface (PEBS). is obtained. (PEBS crossing) should give the transient energy margin. Extended equal area criterion (EEAC) [5-17. This value should be adjusted for the potential energy change during the second kick. Corrected kinetic variations.Kemin2 + Dpe where Dpe is the change in potential energy during the second kick. Kemin2. The transient energy margin. therefore is calculated by: TEM = Kerec2. The EEAC was developed based on the fact that the loss of synchronism in a power system is always initiated from the splitting of the system into the following two parts: • Critical cluster of generators (CCG) • Rest of the system These methods have evolved in their developments through three major stages. The basic idea here is that the kinetic energy injected into the system by the second kick minus the value of the kinetic energy left in the system at the crest of potential energy hill. 5-1. 5-18].long enough to make the system unstable. as shown in Figure 5-1. The transient energy margin is then calculated using the values of the corrected kinetic energy at the second minimum of kinetic energy (Kemin2) and the value after the second fault recovery (Kerec2) taking into account adjustments due to potential energy change during the second kick. is applied and simulation is continued until the second minimum of kinetic energy. Figure 5-1 shows the system trajectory on the potential energy surface. Fig. the integrated method implementation can be made to check the system status during the simulation. In the integrated method. and η ≤ 0 if the system is unstable η > 0 if the system is stable The computation of this index requires a straightforward implementation of the integrated method on top of the time-domain simulation engine. and by using the Taylor-series expansion technique to obtain the approximate trajectory of the system. the transformation is integrated with the detailed time-domain simulations. -100 ≤ η ≤ 100. This is achieved by simplifying power system modeling.• Dynamic method • Integrated method The SIngle Machine Equivalent (SIME) method belongs to the last type [5-21]. Thus. For all other cases a complete simulation is required. such an index is not subject to any modeling restrictions and it is also able to identify multi-swing stability problem. 5-5 . The stability index η of the system can then be defined as A dec − A inc  If the system is stable ( A dec > A inc ) 100 × A dec η=  A − A inc 100 × dec If the system is unstable ( A inc > A dec )  A inc Thus. The dynamic method improves the accuracy by using several transformations. In systems for which angle stability is the only concern to the dynamic security. no modeling compromise is required and the stability index so computed is very accurate. If the system is found to be definitely stable or unstable prior to the end of the simulation. Figure 5-2 shows the principle of the integrated EEAC. the simulation is terminated. The basic difference among these versions is the number of Critical Cluster Center of Inertia (CCCOI) transformations that are performed to obtain the parametric One-Machine Infinite Bus (OMIB) system. The static method does only one static transformation and therefore its accuracy is usually not satisfactory. This is an area for which more research is needed to equip this class of methods with more sophisticated early termination techniques. System snapshots are taken from the conventional time-domain simulation results (Figure 5-2 (a)) and for each snapshot a CCCOI transformation is performed to obtain the parametric OMIB system trajectory (Figure 5-2 (b)). As described earlier. Other methods to determine stability margin. Integrated EEAC. Stability Limit Calculation: The final outcome of TSA are guidelines for system operation in the form of pre-contingency transfer limits and generation tripping remedial action immediately following a fault. and accelerating powers obtained from time-domain simulations. and has since been extended. 5-2. It uses one time domain simulation and with the help of the generator’s angular swing and kinetic energy estimates the required tripping. the authors have verified the performance of the method on a developed prototype. The idea is based on the path of the post-fault system trajectory: the system is stable if its trajectory “swings back” before reaching PEBS or unstable if its trajectory “exits” the PEBS. The authors successfully applied this index to determine the transient stability transfer limit for the Hydro Quebec system. Several approaches have proposed to determine the power transfer limits.δ (a) Multi-machine system trajectory (Time-domain) (b) Parametric OMIB system trajectory (Power-angle characteristic) Pe Pm Adec Ainc Electric power Mechanical power Kinetic energy decreasing area Kinetic energy increasing area System snapshot t CCCOI transformation P Pe Adec Pm Ainc δ Fig. Reference 5-31 uses the signal energy obtained from time-domain simulations to define a stability index. speeds. The “Transient Stability Control” system has been in service since June 1995. all of which make use of the stability margin or index as a measure of system stability: 5-6 . Using a conservative threshold. Reference 5-6 defines a stability index using dot products of generator rotor angles. A DSA system using this index has been developed and is now operational on-line. Reference 5-7 describes an approach for establishing the required generator tripping. based on design limitations and experience. • Binary and accelerated binary search method [5-31]. and DSA needs to handle these problems. or generation change in the case of power transfer limit calculations.. Figure 5-3 shows the application of the sensitivity-based method (in this case. although the system is transiently stable.• Sensitivity-based method [5-5. 5-7 .e. These analytical equations can only be used in the first step to calculate the conditions for the next one and be abandoned afterwards. Historically. while ensuring the maximum allowable transfer between areas. the least stable mode) in a system by using the multi-channel Prony analysis. sensitivity values are calculated for stability margin with respect to generation tripping. To find the power transfer limit and the required generation tripping. utilities have established remedial action controls.5-20]. an increasing concern on the angle stability of power systems is the oscillation problems. The multi-channel Prony algorithm helps improve drastically the accuracy of the results as compared with the commonly used single-channel analysis. It’s therefore important to identify the critical mode in order to assess the security of the system. After the second run linear interpolation is used to obtain the sensitivity values from the two previous stability margin calculations. MW -1 Iteration 1 Fig. e. • Curve fitting method [5-6]. 1 Energy Margin Iteration 2 Iteration 0 0 200 100 Generation Rejection . using a system model consisting of 6139 buses and 798 generators. Reference 5-28 has moved in that direction by establishing the corrective actions necessary to stabilize all dangerous contingencies simultaneously. Figure 5-4 shows the simulation results for selected generator rotor angles for a typical contingency. 5-3. It can be seen that. Limit calculation using sensitivity factors. units for generation shedding. This method is illustrated in the following. The advantage of this approach is that it makes use of the time-domain simulation results with very small computational overhead. Critical damping estimate: As mentioned earlier. sustained oscillations exist. Reference 5-32 presents a method of estimating the damping of the critical mode (i. the stability margin obtain by the second kick approach is used). There seems to be a need for rigorous methods which can establish the most effective remedial actions in real-time.g. (Hz) 0. Further complicating the matter is the tendency that utilities are using larger and larger base case models in their EMS.1. the results from the full eigen-value analysis and the conventional Prony analysis on each individual channel are also shown in the table. It is inevitable that on-line DSA systems have to work with EMS models of 5. Generator angle (degrees) 150 100 50 0 -50 -100 0 1 2 3 4 5 Time in seconds Fig.62 -3.20 Prony on rotor angle of generator ‘C’ 0.75 -0. As typical requirements. For comparison. the results from individual generator rotor angles are apparently not reliable. Simulation results for the sample system.4 DSA performance enhancements As an on-line application.75 Prony on rotor angle of generator ‘D’ 0. ‘C’. so does the four-channel Prony algorithm.03 5. ‘D’ 0.Table 5-1 contains the critical mode identified using a four-channel Prony analysis algorithm.71 Damping (%) 2.000 5-8 . 5-4. Table 5-1: Critical mode comparison from different methods Computation Method Prony on rotor angle of generator ‘A’ Freq. However.11 Four-channel Prony on rotor angles of ‘A’. the computation speed of DSA has been of critical importance to the end users.57 Prony on rotor angle of generator ‘B’ 0.80 -3. ‘B’.58 Full eigen-value analysis 0.79 0. an on-line DSA system should be able to process hundreds of contingencies for dozens of transactions within a 10 to 20 minutes cycle.79 -3.000 to 10.79 Hz with almost zero damping. The eigen-value analysis clearly shows the critical mode at 0. The main steps of this approach are the prediction (say. commissioning. and for stability control analysts and developers [5-23].2 Other Integration of DSA and Stability Controls The previous section described DSA methods where the output is used for stability control adaptation—namely. and be used for direct monitoring and stability control. This can be thought of as very slow. see discussion in §4. monitor-based DSA is valuable for both system operators.buses and up to 1. parallel or distributed simulations for transactions or contingencies can be easily achieved. Complementing the computer-based contingency analysis described in the previous section. multi-contingency simulations. Another potential use of DSA in advanced stability control is pattern recognition based control where DSA provides the database. Any fast contingency screening method that can reduce the false alarm rate while maintaining zero false dismissals can drastically reduce the computation time needed for detailed analysis of the critical contingencies. In this work. it’s desirable to terminate the simulation as soon as the stability of the system can be identified using a stability index. the real-time transient stability emergency controls are derived by feeding the Single-Machine Equivalent method [SIME] with real-time measurements taken at the power plants to control the system transient stability in real-time and in a closed-loop fashion. Phasor measurements may be sufficiently related to dynamic states such as rotor angles and speeds to be useful for stability control. fast digital measurements support stability control development. This is described in Chapter 4. Since DSA involves multi-transaction. and monitoring as discussed in Chapter 2. 5-9 . techniques that speed up DSA performance need to be developed and deployed in order to meet the requirements. Thus.000 generators. This is the classical method of improving the speed of simulations. the following are noteworthy: • Development of better contingency screening method. Synchronized positive sequence phasor measurements are one type of digital measurements. 5. Other synergism is possible between DSA and stability controls. When using time-domain simulations for detailed analysis of the critical contingencies.3. In the latter case. outer-loop adaptive supervisory control. • Enhancement of Early termination techniques. the amount of generation tripping required to compensate for this margin is assessed and the system status after the corrective action has been triggered is monitored to establish whether this action is sufficient or additional remedial action is required. 150 to 200 ms ahead) of the transient stability status of a system after a fault occurrence and its clearance by protective relays and its degree of instability if instability is detected. arming of the correct number of generating units for tripping. In addition. Another application of DSA for stability controls using measurements is found in reference [527]. • Parallel or distributed computations. Among the techniques that aim at improving DSA performances. High quality digital measurements can both improve state estimation as described above. ” IEEE Transactions on Power Systems. Tamby. Palo Alto. V. Ejebe. E. Mansour.C. 384–393. Berlin. Demaree. R. D. Fouad and V. 5-7 H. Ito. Vittal. G. Analysis. 5-13 A. 5-9 Y. August 1996. pp. A. April 1994. 5-10 . Energy Function Analysis for Power System Stability. 4. Hirsch. H. T. A. 1524–1530.” IEEE Transactions on Power Systems. Omata. Prentice Hall. Y. Mokhtari. “Development of Transient Stability Control System (TSC) Based on On-line Stability Calculations. Kitayama.” presented at the IEEE Summer Meeting. Athay. 1989. Athay. No. 9. 241–253. 5-11 T. 3. “B. Alden. 12. Vol. and A.” EPRI Workshop on DSA/VSA. M. K. pp.” IEEE Transactions on Power Systems. K. and M. J. Kumar. B. A. 5-2 Y. “Hybrid Transient Stability Analysis. No. Power System Transient Stability Analysis Using Transient Energy Function Method.954–962. Vaahedi.” IEEE Transactions on Power Systems. “On-line Dynamic Security Assessment: Transient Energy Based Screening and Monitoring for Stability Limits. PAS-98. B. Tang. and R. pp. Cheung. K.” IEEE Transactions on Power Systems. Chang. pp. H. B. “Large Scale Dynamic Security Screening and Ranking Using Neural Networks. and S. 1463–1472. Vittal. and B. Morita. Graham. Y. 1716–1722. T. and K. No. G. Cheung. R. Germany. C. Kluwer Academic Publishers. 5-5 E. 573– 584. K. May 1990. and P. A. EPRI Final Report. February 1995. 5-3 G. and Y. Vaahedi. Tang. Vaahedi. F. pp. Mansour. E. Ipakchi. Vol. 5-10 S. 11. El-Sharkawi. J. Pai. Vol. A. Waight. C.” IEEE Transactions on Power Systems. C. and Post-processing. 2. July 1997. B. No. Y. November 1994. 5-12 M. 5-8 A. RP3103-02. Ota. T. Vaahedi. Athay. “An On-line Dynamic Security Analysis System Implementation. Podmore. C. 2. 1.” IEEE Transactions on Power Systems. Demaree. No. March/April1979. 5. Mansour. K. Virmani. Power System Dynamic Security Assessment Using Artificial Intelligence Systems. “A General Purpose Online Dynamic Security Assessment Method. Kim. “A Practical Method for Direct Analysis of Transient Stability. Jing. Kokaki. Garrett. and J. 10. El-Kady. Jamshidian. May 94. Corns. Vol. Pieper.References 5-1 K. Tse. Maria. 1990. E. A. Corns. Chang. October 9–10. 5-6 G. May 1997. Vol. and Y. Corns. “Transient Stability Index from Conventional Time Domain Simulation. 5-4 C. Sobajic. V. Hydro’s On-line Transient Stability Assessment (TSA) Model Development. Mansour. No. EPRI Final Report. pp. August 1994. Analytical Methods for Contingency Selection and Ranking for Dynamic Security Assessment. RP3103-03. Brandwajan. Vol. 2. E. Chang. pp. ” IFAC/CIGRE Symposium on Control of Power Systems and Power Plants. September 1987. Gonzalez. “Keeping an Eye on Power System Dynamics. J. E. P. Ding. W. Z. 660–668. 11. 5-17 A. pp. Vol. B. J. Kundur. No.” International Journal of Electrical Power & Energy Systems. W. 2. Karimi. August 1987. M. C. B. 1997 and also appeared in Control Engineering Practice. 19. W. Xue and M. B. 1497–1529. Chu. D. 1995. Kafka. L. January 1997. J. Litzenberger. Y. Pavella. Vol. “Extended Equal-Area Criterion: An Analytical Ultra-fast Method for Transient Stability Assessment and Preventive Control of Power Systems.” Proceedings of IEEE. G.” IEEE Report No: 87TH0169-3PWR. Luo. “Rapid Analysis of Transient Stability. S. 604–609. No. R. 5-21 Y. Y.” Control Engineering Practice. “State Estimation with Phasor Measurement.” IEEE Transactions on Power Systems. 5-11 . 26–30. Gonzalez Provost.” IEEE Computer Applications in Power.” to be published in IEEE special publication on Techniques for Transient Stability Limit Searches. 5. “Power System Transient Stability Index for On-line Analysis of ‘Worst -Case’ Dynamic Contingencies. and J. 83. 5-20 L. and Y. “SIME : A Hybrid Approach to Fast Transient Stability Assessment and Contingency Selection.” EPES. Vol. Kundur. No. Wang. Schaffer. 1997. Vol. No. Beijing. No. Li. G. 5-19 Y. Mittelstadt. F. 233– 241. “Fast Simulation of Power System Dynamic in General-purpose Simulator. M. Utah. J. 1511–1516. J. pp. M. 1. Morison. No. PWRS-1.” IEEE Transactions on Power Systems. C. Salt Lake City. Damonte. PWRS-2. 1. Wang. “Implementation of Phasor Measurements in State Estimator at Sevillana de Electricidad. November 1995. Zhang.3. and P. 1998. Applications. 5-16 P. Xue. pp. Phadke. Yu. 11. 5-18 Y. 5-15 H. Vol.5-14 W. Adibi and R. Hauer. and J. Xue. 131– 149. G. 123–129. 10.” IEEE Transactions on Power Systems. K. G. 195–208. Perez. P. M. pp. May 7–12. 1997. Cova. Sierra. pp. S. Baratella. F. Wang. F. K. “Direct Stability Analysis of Electric Power Systems Using Energy Functions: Theory. Thorp. Morison. and Prospective. Pavella. Ding. A. No. Price. No. Vol. J. 5-22 I. Gao. “Quantitative Search of Transient Stability Limits Using EEAC. Vol.3. 5-24 M. Rousseaux. Jaques. Vol. pp. Trudnowski. Marconato. pp. 3. Rahimi and G. and K. Wehenkel. Figueroa. L. “Minimization of Uncertainties in Analog Measurements for Use in State Estimation. 902–910. Xue. 1989. and G. Chiang. Cauley. pp. L. Mokhtari. Scarpellini. Slutsker. 5. M. P. China. Gaglioti. 5-25 A. Johnson. X. Xue. and M. C. February 1986. “A New Tool for Dynamic Security Assessment of Power Systems. Vol. pp.” IEEE/PES Power Industry Computer Applications Conference. 1. Rogers. W. pp. 6. August 1990. 5-23 J. M. “Real Time Voltage-Phasor Measurements for Static State Estimation. W. 5-29 K. V. R. China. G.” IEEE Transactions on Power Systems. Wehenkel and M. 5-31 R. Pavella. Martins. N. 5-30 K. 5-27 Y.” presented at the panel session at the 1998 IEEE/PES Summer Meeting in San Diego.” Proc. on Control of Power Systems and Power Plants. Uchida. and K. November 1985. 5-12 . 5-28 A. of CPSPP’97. 2. Cheung.” Proceeding of International Conference on Intelligent System Application to Power Systems (ISAP’94). Wang. Marceau. Vittal. Sanchez-Gasca. T. 11.” to be published in IEEE special publication on Techniques for Transient Stability Limit Searches. No. Soumare. Gibbard. pp. 673–678. J. W. Paliza. S. and X. “A New Hybrid Method for On-line Dynamic Security Assessment. “Recent Applications Of Linear Analysis Techniques.” IFAC Control Engineering Practice. 654–659. L. Zhang. S. No. 5-32 M.” IEEE Transactions on Power Apparatus and Systems. pp. Montpellier. Vol. PAS-104. Do. M.5-26 J. 6 (1998). Ma. “A Method for Real-Time Transient Stability Emergency Control. L. Karimi. August 1997. Bettiol. Beijing. “An Expert System Guided On-line Dynamic Security Assessment System. Siraogue. Pavella. Zuk. Phadke. D. K. J. pp. A. IFAC/CIGRE Symp. 3098–3106. France. J. 263– 270. pp. Thorp. 14. May 1999. Athay. L. 1373–1380. J. N. Cheung. Wehenkel and M. “Transient Stability-Constrained Maximum Allowable Transfer. pp. L. T. “Signal Energy Search Strategies for Transient Stability Transfer Limit Determination. Vol. There are many aspects of the controller environment which cannot be predicted from model studies. even when reinforced by extensive modeling studies. It’s also a potential path for disruptive interactions between the actuator and dynamic processes other than those targeted for . Providing the controller (and the control engineer) an ample reserve of directly measured dynamic information can increase controller performance and robustness. a source of undesirable noise disturbances. or equally difficult to transmit reliably. or from the sensors (instrument transformers) that provide inputs to those transducers. has conducted numerous measurements of system dynamics [6-3–8]. It’s useful to distinguish among: • Modulation signals used to directly shape the controller output. False measurements will. and which may not be measurable until the controller itself is available for system dynamics testing. It’s likely that any extraneous signals emitting from the transducer will be amplified by the actuator and re-injected into the power system. u(t) • System status flags used in • - rule-based control laws (e. Some. especially those in which modulation or sampling processes can translate signal components from one frequency to another. The Bonneville Power Administration (BPA). Fortunately.Chapter 6 Measurement and Communication Technology A stability controller addressing global objectives needs a reliable source of global information.g. through long involvement in stability control projects. parameter scheduling) - coordination with other controls - remote controller supervision Secondary response signals for - direct testing of power system dynamics and controller effects - local monitoring of controller performance - alternate or supplemental modulation signals Requiring that all modulation signals be local can make controller siting a difficult robustness issue [6-1. may have involved false outputs from the transducers being used. Local signals may prove inadequate for this. This has produced many examples of apparently anomalous system behavior. at best. This is. however. The situation is more serious when false measurements enter the modulation loop of a major control system (Figure 6-1). produce an erroneous view of the power system. not all remote signals are equally vital to controller performance. Most have been attributable to the power system itself. at best. Other likely sources include communication channels and secondary control loops. This can readily lead to inappropriate engineering or operational decisions..6-2]. Advanced designs can achieve bandwidths approaching 30 Hz. have sharply increased the risks in this respect. for example. The trend toward fast power electronic actuators. Transducer effects in a closed loop controller. 6-1. It also mandates the use of such transducers in monitoring of controller performance. The dynamics under control rarely occur at bandwidths above 2 Hz. Desired characteristics for next-generation transducers include: • Rigorous protection against out-of-band input signals. The actuator. Table 6-1. This encourages the use of “high speed” transducers with bandwidths in this same frequency range. such as spurious outputs. or else the voltage support that the generators provide to the transmission network. for versatility of application. indicates that there are many candidate sources for extraneous transducer outputs. to obtain full actuator performance. HVDC links and SVCs. Increasing transducer bandwidth also increases the likelihood that such effects will be present in the transducer output. Enhanced analog transducers used at the BPA have a bandwidth of 20 Hz. for example. • Absence of processing artifacts. external inputs u E (t) unmeasured response y’(t) POWER load noise u L (t) measured response y(t) SYSTEM input u(t) nonlinear interactions extraneous signals Sensors & Transducers measurement noise u m (t) – linear response u(t) ~ nonlinear response u(t) ACTUATOR actuator noise υu (t ) ^ command u(t) CONTROL LAW controller input y m (t) Fig. may have controllable responses up to 20–25 Hz. Field experience suggests that the interpretation and use of outputs from these or any other transducers should be approached carefully. in principal at least. based in part from reference 6-11. and the control law driving it. The vast majority of the transducers now in service are analog devices with bandwidths in the range of 1 to 2 Hz. together with more aggressive control objectives. 6-2 . Most stability controls seek to influence generator “swing” activity. permit mechanical or network resonances to mimic swing dynamics. • Programmable outputs. may very well have bandwidths higher than this.control. Transducers of all types may be subject to aliasing effects that might. At the other extreme. a digital relay is designed to detect and assess dynamic events with no more accuracy than reliability demands. Chief among these are rms (root mean square) voltage.1 Introduction to Transducers For our immediate purposes a transducer is a signal processing device that translates instantaneous “point on wave” current and voltage signals into averaged measures of electrical behavior. which sacrifice dynamic response for high reliability and accuracy. Table 6-1. waveform frequency. contingent upon the: • • Kind of dynamic process to be stabilized. Suitable choices among these measures—and for the averaging times used in calculating them— are determined by the information that is needed. and relative angles for voltage and current. • Option for synchronizing measurements against a precise external reference.Extraneous dynamics in the transducer environment Dynamic Activity Frequency Range -Hz Torsional oscillations Transient torques Turbine blade vibrations Fast bus transfer Controller interactions Harmonic interactions and resonances Ferroresonance Network resonances 5 – 120 5 – 50 80 – 250 1 – 1000 10 – 30 60 – 600 1 – 1000 10 – 300 6. considering all cost elements. Such transducers will almost certainly require digital technology. in control applications. both local and wide area. • Overall affordability. in metering applications. The slow end of the spectrum is occupied by revenue meters. rms current. • Good networking options. Transducers for stability control occupy a broad middle range. Possibilities include - modulation signal for feedback control 6-3 . Existing transducer technology reflects a broad range of information needs. Possibilities include - local swing modes (with modulation on individual generators) - interarea swing modes (with modulation on HVDC or FACTS device) - voltage dynamics Role of the transducer in the control process. rms power. • Assured high accuracy. Reference 6-12 provides an overview of standard transducer types.• High resolution and bandwidth. It would probably be more lightly filtered. insertion of Chief Joseph dynamic brake on August 10. By contrast. 6-3.- monitoring of power system conditions and behavior . 6-4 . by choice. 6-2. a transducer for monitoring controller performance should be capable of detecting such extraneous dynamic activity. and it might have a bandwidth approaching 25 or 30 Hz.monitoring of controller activity. Malin-Round Mountain #1 MW : PG&E Malin Sum MW : PG&E Olinda MW 150 100 50 0 -50 -100 Malin-Round Mtn1 MW swing PG&E Malin Sum MW/2 -150 -200 168 PG&E Olinda MW 170 172 174 176 178 Time in Seconds Fig. especially anomalous interactions The transducer for a feedback modulation system would. Time response of Malin area transducers. Measured current at Olinda substation for COTP Test Fault #3 (1715 h on 03/23/93). If interactions are sensed at such frequencies it is highly desirable that direct waveform recordings be made on a local digital fault recorder (DFR) or similar device. Figure 6-2 provides an example of current waveforms under moderately disturbed conditions [6-13]. 1996. A Phase B Phase C Phase Fig. be equipped with internal and external filters protecting the feedback loop from interactions with extraneous dynamics. and that communicates on a 20 Hz channel. it appears that the much slow Olinda transducer has a -3 dB bandwidth near 0. but more accurate than what would be obtained by direct inspection. These differences in bandwidth and transient performance are not always evident. based upon ambient noise outputs. compares the slowest signal against the fastest.g. as observed through the microwave channels that communicate their outputs to BPA’s control center. Since the fast transducer has a much higher bandwidth (e..5 Hz or lower). At the other extreme. its response is nearly flat to well past 1 Hz). All three transducers measure components of the real power export to Pacific Gas & Electric (PG&E) on the California – Oregon Interconnection. Its relative time delay. Differences among the signals in Figure 6-3 are almost entirely due to the instrumentation. The signal for Malin – Round Mountain circuit 1 is taken from an enhanced analog transducer that has a bandwidth of 12–14 Hz. Quantitative measures of relative response can be obtained by correlating the transducer signals against one another. or COI. or major handicaps.85) = 0.4 Hz. This value is consistent with Figure 6-3.85 Hz. Amplitude differences for the spectral peaks can be corrected through knowledge of the transducer filtering and channel response. the transducer for the PG&E – Olinda exchange is a conventional analog transducer communicating through a low-bandwidth channel (probably 1. in other forms of analysis. Olinda transducer response relative to Malin transducer. 6-4.Figures 6-3 through 6-5 indicate the relative performance of three kinds of transducers. and the PG&E – Olinda signal should be very similar except for magnitude. evident in the linear phase characteristic which produces a lag of 180 degrees near 0. as can 6-5 . Figure 6-4. Figure 6-5 shows that all three transducer signals produce useful and consistent spectral characterizations for important WSCC swing modes up to perhaps 1.9 Hz. and that the waveforms also exhibit appreciable delays. PG&E Olinda MW relative to Malin-Round Mountain #1 MW 180 0 TF Gain in dB -10 60 PHASE 0 -20 -60 -30 TF Phase in Degrees 90 -3 dB -90 GAIN -180 -40 0 1 Frequency in Hertz 2 Fig. Ignoring transducer/channel dynamics. the PG&E – Malin signal should be very close to twice that for Malin – Round Mountain circuit 1. can be estimated as (180)/(360*0. It is apparent that much of the waveform detail is not tracked very well by the slower instrumentation.59 second. some of the phase and timing differences. even during steady operation. 6-5. a rms transducer for real or reactive power may have as many as 3 voltage and 3 current signals as its inputs. 60 Hz) carriers by a combination of amplitude modulation and frequency modulation. Malin-Round Mountain #1 MW : PG&E Malin Sum MW : PG&E Olinda MW 10 Malin-Round Mtn1 MW swing PG&E Malin Sum MW/2 Autospectra in dB 0 PG&E Olinda MW -10 -20 -30 0 1 Frequency in Hertz 2 Fig. At the other extreme. These include: • modulated harmonics of 60 Hz • extraneous carriers (not necessarily at harmonics of 60 Hz) • modulated extraneous carriers • additive transients. Such corrections add considerably to computational demands and staff workload. There is no assurance that their underlying 3-phase carriers will be balanced. however.g. and they are rarely possible in an on-line environment.. Ambient noise autospectra for Malin area transducers 6. In the simplest cases the transducer will have just one input. For this and other reasons. determining the physical significance of system activity may necessitate decomposition of the signals into “symmetrical components” plus accessory filtering specific to their application.2 The Signal Environment for Power System Transducers A power system transducer is intended to extract information that has been impressed upon a set of fundamental-frequency (e. Reference 6-11 surveys physical sources for such extraneous components. The input signals may also contain components produced by mechanisms other than modulation of the fundamental frequency carriers (Figure 6-6). It’s better to avoid them through use of quality instrumentation. 6-6 . Signal environment for a power system transducer. Figures 6-7 through 6-9 show a sequence of autospectra for A-phase current. system frequency offsets. 1996. d) 30 Hz modulation of a 120 Hz carrier produces waveform components at 90 Hz and at 150 Hz (overlapping case c). The governing relations are simple: sin(x) sin(y) = 2 [cos(x − y) − cos(x + y)] 1 1 2 sin(x) cos(y) = [sin(x − y) + sin(x + y)] (6. at 59 Hz and at 61 Hz). Sampling effects in digital transducers considerably expand the possibilities for frequency aliasing. 6-6. e) Squaring any of the above waveforms produces terms at 0 Hz. and digital decimation. controls.e. On April 24. etc) other carriers (harmonics. Amplitude modulation can affect both analog and digital transducers. Those based upon amplitude modulation are discussed more thoroughly in reference 6-14. direct measurements were performed on enhanced transducers at BPA’s Slatt substation [6-15]. Related theory is available in [6-16–18].. This provides a number of ways in which a signal might enter a transducer at one frequency and be shifted to another.1B) The following examples show some consequences of these relations (see also Appendix D): a) 1 Hz modulation of a 60 Hz carrier produces waveform components at 60±1 Hz (i. etc) Fig. as determined with a Scientific Atlanta SD390 4-channel dynamic signal analyzer. saturation.other modulation (shafts. Field observations confirm that transducers operate in a very demanding signal environment. Then the original modulation can be recovered by lowpass filtering. In Figure 6-7 the peaks 6-7 . c) 30 Hz modulation of a 60 Hz carrier produces waveform components at 30 Hz and at 90 Hz. The most likely sources of frequency aliasing seem to be amplitude modulation. etc) fundamental carrier + (60 Hz) multiplier RMS Transducer Response to Inputs Processing Artifacts additive signals (LC resonances. etc) multiplier fundamental modulation (generator swings. b) Re-modulation of the above waveform produces components at ±1 Hz and at 120±1 Hz.1A) (6. 04/24/96 -20 -60 -80 91. 04/24/96. The numerous spectral peaks. Slatt Substation. 0 A-phase current Slatt substation. The PDCI was deliberately operated in a high harmonic configuration in order to test transducer performance [6-19]. which is directly connected to the Celilo converter of the Pacific HVDC Intertie (PDCI). Slatt Substation. 6-8.near 28 Hz and 92 Hz are probably associated with a modulating source at 32 Hz (very likely a generator shaft). many not at integer harmonics of the 60 Hz power frequency. Autospectrum for A-phase current. 04/24/96 500 Hz processing -40 -60 -80 -100 -120 0 100 200 300 400 Frequency in Hertz Fig. 6-7.00 Hz Amplitude in dB 100 Hz processing -40 -100 -120 0 25 50 75 100 Frequency in Hertz Fig. 04/24/96. 6-8 500 . Corresponding transducer spectra are shown in a later section.94 Hz 28. The spectrum in this figure was obtained at Big Eddy substation. These spectra are in close agreement with MATLAB analysis of signals extracted from the BEN 5000 digital fault recorder at Slatt. and voltage spectra were similarly complex. Autospectrum for A-phase current. further indicate just how harsh the transducer operating environment can be. Figure 6-10 extends the observations made at Slatt substation. 0 A-phase current Amplitude in dB -20 Slatt substation. HVDC controls in high harmonic configuration. indicate that protracted operation at anomalous frequencies is another challenge to transducer performance. 6-9 . 04/24/96. showing frequency records for individual islands formed during WSCC breakups in 1994. Big Eddy Substation. 6-9. In the case of Figure 6-11 the island frequency remains below 59.9 Hz for roughly 25 minutes. Fr equency in Hert z Fig. 04/24/96 Amplitude in dB 2000 Hz processing -40 -60 -80 -100 -120 0 10 20 30 Frequency in Multiples of 60 Hertz Fig. 10/25/96. Autospectrum for A-phase current. It’s possible that transducers not designed for such operation would experience filtering or timing problems under such conditions. Figures 6-11 and 6-12. Autospectrum for A-phase current. 6-10. and produce spurious outputs. Slatt Substation.0 A-phase current -20 Slatt substation. 5 59. Phoenix.0 59. 6-11.108 60.274 FREQUENCY IN HZ 60. Vancouver. but digital if some or all of its outputs are in the form of multi-level digital data. into: • “Algebraic” or “point on wave” transducers that perform simple calculations upon v(t) and i(t). 1994.3 Signal Processing in Power System Transducers Let v(t) and i(t) denote the instantaneous voltage and current signals that are processed within a particular transducer.2 60. 6-10 .0 59. 6. BPA system frequency following Los Angeles earthquake of January 17. 1994 (BPA control center.3 Kyrene substation. a bit cavalierly. Washington).1 60. 60. Transducers can be categorized.221 Dittmer PPSM #1 59.0 0 10 20 30 TIME IN MINUTES Fig.8 59. Arizona area system frequency following WSCC breakup of December 14.5 60. 6-12.60.9 59. We will consider a transducer to be of analog type if all of its output signals are analog.6 0 10 20 30 40 50 TIME IN MINUTES Fig. Phoenix AZ FREQUENCY IN HERTZ 60.7 59. The two kinds of modulation. Vin Iin INPUT ISOLATION INPUT ISOLATION TRIANGULAR WAVE GENERATOR COMPARATOR CIRCUIT AMPLITUDE FILTER MODULATOR OUTPUT ISOLATION Prms Fig.• Phasor transducers that project v(t) and i(t) onto reference waveforms. constitute an analog multiplier circuit and yield a signal that is pulse-ratio modulated (PRM). General architecture for a pulse ratio modulation (PRM) megawatt transducer. 6-13. which is then amplitude modulated by the current signal. The processing in Figure 6-14 is representative of modern analog transducers that BPA uses to measure real and reactive power. Phasor determination via projections. thereby generating associated voltage and current phasors that are used in all further calculations (see Figure 6-13). 6-11 . 6-14. Filtering requirements are greatly reduced by first combining the PRM signals for all three phases. and specific hardware products may well contain some for each category. PWM followed by AM. Whereas algebraic transducers may be either analog or digital. The voltage input width modulates a train of square pulses. contemporary phasor transducers appear to be entirely digital. Cosine Reference V=|V|∠θ = V r + jV i (pol ar f or m) (rectrangular form) Vi θ Vr Sine Reference Fig. The associated logic can be organized in many different ways. current signals voltage signals . voltage signals current signals Figure 6-15 represents the functional organization (showing just one phase) for an algebraic digital transducer. PROTECTIVE INTERFACE PRECONVERSION FILTERS A/D CONVERTER (*Grey shading indicates optional element) POSTCONVERSION FILTERS * INPUT DECIMATION GUARD RMS FILTERS CALCULATIONS PREDECIMATION FILTERS #2 OUTPUT DECIMATION rms outputs Fig. which is broadly representative of digital transducers based upon phasor calculations. the guard filters warrant special mention. and it permits use of this estimate to adjust the reference signals upon which voltage and current signals are projected. 6-16. 6-15. (*Grey shading indicates optional element) PROTECTIVE INTERFACE PRECONVERSION FILTERS BUS FREQUENCY ESTIMATOR frequency A/D CONVERTER REFERENCE POSTCONVERSION FILTERS SECONDARY CALCULATIONS SIGNALS * INPUT GUARD FOURIER DECIMATION FILTERS FILTERS SYMMETRICAL COMPONENTS LOGIC PREDECIMATION FILTERS #2 OUTPUT DECIMATION phasors multiplier Fig. General architecture for an algebraic digital transducer. The structure provides several points where bus frequency can be estimated. This is extended in Figure 6-16. General architecture for a phasor transducer. Among the other optional features in Figure 6-16. While their function might actually be absorbed into the post-conversion or the 6-12 . 6. network condition monitoring. bandwidth. • Large-signal dynamic performance. This is particularly likely in high performance stability control.17. other filtering considerations. Technical performance factors in control applications are resolution. accuracy. where both the control law and the monitoring equipment should be well protected against spurious activity. Emphasis upon remedial action (feedforward) control. 6.Fourier filtering. of course. Emphasis upon precise measurements under normal system conditions. Allocation of transducer signals in power system control. and transient behavior. As illustrated in Figure 6-17. delay. The proposed test procedures will focus upon small-signal dynamic performance. which is critical in those applications having the greatest exposure to parasitic interactions. 6-13 . that signal processing within the source transducer is fast enough to track large signal dynamics in the first place. This assumes. the appropriate settings may change with the application. • Small-signal dynamic performance. POWER measured response y(t) SYSTEM Transducer Setpoint slow trends LowPass Filter Controls Damping small-signal HighPass Controls activity Filter Emergency large-signal Controls activity Fig. noise. disturbance monitoring. and much the same as that used in a dynamic signal analyzer. Distinctions are also made between kinds of information to be obtained from transducers in a control environment. Emphasis upon feedback control and interactions monitoring. The signal processing in a phasor transducer is directly based upon Fourier analysis.4 Criteria and Procedures for Evaluating Transducer Performance Distinctions are made here between the following kinds of transducer performance: • metering performance. See also Appendix E. the output of a transducer that is used to track large signal dynamics might also be low-pass filtered to display slow trends and high-pass filtered to display small-signal activity. protection from aliasing. b) Resolution and dynamic range: equivalent to 14–16 bits. and often removed by highpass filtering. f) Positive sequence response: transducers intended to respond just to positive sequence activity should perform accordingly. e) Harmonic modulation: information imposed upon power frequency harmonics above the first must be outside the nominal bandwidth of the transducer. k) Drift: often removed by highpass filtering. h) Unbalanced operation: the above criteria (a–f) should be met during sustained unbalances of three phase voltage and/or current. g) Off-frequency performance: the above criteria (a–f) should be met during sustained ramps and offsets of system frequency. It should be noted that no criteria are recommended for discriminating between additive signals on the power system (such as network resonances) and signals associated with carrier modulation.The criteria for evaluating transducers in control applications are necessarily different from those used in static or slowly changing measurements. 6. c) Delay: must be essentially constant. the higher the better. The following are recommended as high priority performance targets in control applications: a) Bandwidth: in the range of 10–25 Hz.5 Transducer Modeling and Simulation Laboratory tests and field examination of transducer performance have been reinforced through the use of computer models. The following are recommended as lower priority performance targets for control applications: i) Accuracy: something on the order of 0. it’s unlikely that existing transducers can provide it. not to exceed 30–45 degrees within the primary control band. but must be outside the nominal bandwidth of the transducer and small enough to remove through accessory filtering. Appendix F describes laboratory evaluation of transducers and Appendix G describes field evaluation of transducers. it must not vary so rapidly as to mimic power system activity. Desirable as such a capability would be. The general approach involves: 6-14 .5% of reading is usually sufficient. and small enough to remove through accessory filtering. j) Fixed offset: usually measured. l) Instrument noise: must not have strong peaks within the primary control band. d) Carrier filtering (including harmonics): carrier effects will likely be visible to sophisticated analysis. and must be small enough to remove through accessory filtering. If drift is substantial. Macrodyne Phasor Measurement Unit (PMU) [6-24–26. Also. • Local recording capabilities as a basic “snapshot” monitor. with 16 bit digitization. • An evolving interface for direct local networking. not actual. is valuable even when global phase angles are not produced.• Use of SPICE computer software [6-21. such as filter type and settings. • Output sample rates of 30. As the term is used here. user-selectable. It’s equally true that any well-filtered algebraic digital transducer can probably be converted into a phasor transducer. It’s also apparent that that the desired class of transducers represents a functional extension of the conventional technology. by using the global reference in the phasor projection. The MATLAB codes permit direct changes to signal processing parameters. A transducer that is directly networkable. Considered as an individual device. it’s clear that any technology capable of calculating them is a good candidate for developing better transducers. and that performs measurements in synchronization with some precise global reference. • Use of MATLAB computer software [6-23] to examine the generic signal processing Models were developed for the (analog) PRM megawatt transducer of Figure 6-14. however. not just an improvement.6 Digital Transducers and Phasor Measurements The chapter introduction listed desired characteristics for next-generation transducers. plus digital megawatt transducers of both algebraic and phasor types (Figures 6-15 and 6-16). Apart from the special values that phasors themselves may have in stability control. The essential integrity of phasor processing. this is a phasor transducer plus: • Optional synchronization against precise time references. can be neither developed nor evaluated without considering its role in the overall measurement network. General performance features of the PMU include: • An input sample rate of 12 samples per cycle (720 samples per second at 60 Hz) after prefiltering. and 6 samples per second.6-37]. 12. This section assesses available products having potential for advanced stability control. and they support a broad menu of test waveforms. all phasors in the network provide consistent angle information. 6-15 . Appendix H provides results of transducer modeling and simulation. 6. All sampling is referenced against nominal system frequency.6-22] to examine interface and circuit performance. the distinguishing characteristic of phasor technology and phasor transducers is the explicit calculation of the phasors themselves. and related analysis. The WAMS project [6-10] assessed the several digital transducers. 6-27. 6-16 . Evaluating these devices—or. 7700 ION Programmable Transducer System. which the user can decimate under program control. The DSM is primarily designed to operate independently. quantities such as rms power or apparent resistance must be calculated later. The PMU is expressly designed to operate in wide-area networks. produced by Power Technologies Inc.• Voltage and current positive sequence phasors and bus frequency as standard outputs. the technology approaches that they represent—remains an ongoing process. Operational details and observed performance are described in [6-13. (PTI) [629]. General characteristics of the DSM’s transducer include: • An input sample rate of 4 samples per cycle (nominally 240 samples per second) after prefiltering. • A very wide range of rms outputs. While new digital transducer technology is appearing on the market with increasing frequency. They are also different enough in their processing details to span a good range of the basic possibilities. In this case phasor transducer logic is imbedded into a general purpose monitor. Voltage and current phasors can be obtained for the same global references that are accessed by the PMU. which include central recording plus modem connections into wide area networks.6-25. • Output sample rates ranging to 60 samples per second. with 16 bit digitization. programmable by the user. but with modem connections into wide area networks. The ION 7700 is highly evolved for operation in local area networks. programmable by the user. as an FFT-based harmonic analyzer. All sampling is referenced against a fast running estimate of system frequency. at a lower output rate. produced by Power Measurement Ltd.6-37]. more precisely. via the harmonic analysis and using a local reference. usually with 12 bit digitization. the three devices above are well established in the field and thereby of special interest. • A very wide range of rms outputs. Present logic provides voltage and current phasors at low rates only. • An output sample rate of 1 sample per cycle. The Dynamic System Monitor (DSM). General performance features of the 7700 ION include: • Input sample rate of 128 samples per cycle (7680 samples per second) after prefiltering. All sampling is referenced against a running estimate of system frequency. The 7700 ION (Integrated Object Network) functions as a high bandwidth algebraic digital transducer and. (PML) [6-28]. according to type and user selection. Similar logic.. it is not unusual for a modern excitation controller to contain transducer logic within a digital control law. however. and a building block for wide-area measurement networks.” or IED [6-28. with sensing and transducing logic occupying the lower hierarchies. 6.g. reliability. These needs are most easily met if the signals are produced locally to the controller site.6. • system status flags used in • - rule-based control laws (e. the term “transducer” no longer has very explicit meaning when the base technology is digital. where its status and performance are supervised and coordinated with those of other controls. and even the sensors that provide signals to higher levels of the measurement system. It can be misleading to call several different things “transducers” when the functionality they offer are so diverse. sometimes in optical form. and perhaps to other locations. There are many aspects of the controller environment 6-17 . and are combined into products with different functionality combinations.6-30]. can make controller siting a difficult robustness issue. controllers. At another extreme. and security. The Macrodyne PMU is an outstanding case of this.7 The Transducer as an Intelligent Electronic Device We have shown that transducers and transducer logic take many different forms. Selecting IEDs with the right functionality combinations lies at the heart of the value engineering process. Requiring this in advance. There is a useful trend now to just designate any such device as an “intelligent electronic device. It’s a transducer for producing rms signals. In short. From this perspective a wide-area measurements network is an integrated structure of IEDs. Similar problems are encountered with power system monitors.8 Role of Communication Channels in Wide-Area Control A fully evolved stability controller for wide area dynamics requires access to signals of the following kinds: • modulation signals used to directly shape the controller output u(t). is also appearing in such mundane devices as electrical bushings and circuit breakers. a simple monitor. parameter scheduling) - coordination with other controls - remote controller supervision secondary response signals for - direct testing of power system dynamics and controller effects - local monitoring of controller performance - use as alternate or supplemental modulation signals Signals are also required from the controller to operation centers. The signals used as modulating inputs are the most demanding in terms of quality. analogous to noise filtering in analog technology. The earliest modems transmitted digital signals on 6-18 . There are no delays other than those produced by distance and by filter effects. or frequency shift. by contrast. and reasonable immunity to undetected tampering. Modems. minimal communication delay. Telephone systems are a case in point. The digital format also allows noise free data recovery and positive verification of data integrity. first at the backbone (long distance) level and more recently at the local level.which cannot be predicted from model studies. thereby. it can also cause new problems associated with digital elements of the overall system and with digital/analog interfaces. Channel gains and offsets directly enter the received signals. introduce communication delays and expose the information system to penetration by unauthorized persons. For this discussion. so analog channels require frequent and precise calibration to maintain accuracy. This calls for some kind of data repair. an analog channel is one that accepts an analog signal as an input and carries the signal in large part as a continuously varying analog signal. in communication of digital data. superficially. digital communications modulate analog processes between a finite number of states (often just two states) that the detection logic is designed to recognize. all communication systems are analog in the sense that the physical processes can assume an infinite number of states. But. Channels for modulation signals can. phase shift. Traditional analog communications take an analog signal from a transducer and transports it as a continuously present and continuously varying voltage. At present. At the very lowest level. microwave. Means for separating noise components in the signal from the information are less effective. There are few artifacts in the information equivalent to the aliasing and quantizing errors sometimes introduced in digital systems. involve a conversion of the signal to digital format and a commensurate increase in delay. if present. they have a counterpart in occasional message loss. Modern communications frequently convert analog signals to digital quantities for long distance transmission. Analog channels usually offer the advantages of high bandwidth relative to that of the measured signal. Digital channels. and it is more difficult to detect dropouts. While digital channels do not experience “noise” in the same sense as analog channels. digital communications may also be more expensive than analog for the same bandwidth. and fiber optic links. At higher levels. They also have a lower signal bandwidth for a given channel bandwidth but require less channel calibration. Distortion and noise at the analog level can produce errors in demodulation and. and maintenance intensive. though. current. while this hybrid approach mitigates some of the difficulties found in completely analog systems. be categorized as analog or digital. they have been progressively converted to digital technology. Once entirely analog. This is largely transparent to the user. and which may not be measurable until the controller itself is available for system dynamics testing. This mixture of technologies means that data transmission over telecommunication systems may encounter one or more conversions between analog and digital formats. The transport system can be as simple as a pair of wires or as complex as multi stage exchange involving satellite. They also tend to be noisy. The following section describes utility experience. The PMU at Sylmar belongs to the Los Angeles Department of Water and Power. as points very near zero. All channels are frequency division multiplexed microwave. 6.analog links by shifting the phase or frequency of a tone that was detected as digital 0’s and 1’s at the other end. owned and operated by BPA. Accessory data from the PDC indicate these defective data precisely. The PDC data acquired for the test consists of 23 raw data files (45 megabytes) spanning two recording intervals of roughly 80 minutes each. or some brief loss of synchronism at modem level. John Day. but at the expense of increased processing delay. Each PMU was configured to produce one positive sequence voltage phasor and four positive sequence current phasors at a rate of 30 samples per second (sps). The signal. USA (see Figure 6-18). the issue is how to plan and manage the transition to digital technology. Five Phasor Measurement Units (PMUs) communicated data to a Phasor Data Concentrator (PDC) located near the Dittmer Control Center in Vancouver. which extends across all of recording interval #1. rather than both. Problems aside. 6-19 . 1997 [6-27]. Most of these files contain occasional “outliers” in the data. Appendix I describes utility experience with older analog communication channels. In doing this they monitor communication errors and re-train if the errors increase unduly. Colstrip. The PMU locations were Grand Coulee.9 Observed Performance of Digital Communications in the BPA Phasor Measurement Network This section shows the performance of digital communication channels within BPA’s phasor measurement network. shows just 8 outliers among 144.000 rms power calculations. The performance data were obtained from test insertions of the Chief Joseph dynamic brake on September 4. There was no added communication delay other than the time needed to assemble a set of binary digits into a word for processing. the issue is not digital technology versus analog. To maintain the high data rate modems must “train” with each other to reduce errors. Rather. based upon observed performance of a phasor measurement system spanning a broad region of the western North American power system. More sophisticated coding has now been developed to make better use of the (analog) channel capacity. Breaks can also occur through data re-transmission commanded by error detection logic. and Sylmar. These usually represent data packets (messages) that were lost in the digital communication system. The PDC was developed by Ken Martin of BPA. Malin. It’s possible that some of these represent defects in only the voltage or the current phasor. Figure 6-19 indicates that these outliers often tend to be conspicuous in the signals themselves. The result is more digital capacity. Washington. This can cause an unanticipated break in communication service. The source file for this record shows similar defects in all data extracted from the Colstrip PMU within this particular time frame. 6-18. outliers near zero are easily patched through linear interpolation. 6-20 . and (where possible) repair bad data at the signal analysis level. Effective standards and mechanisms for this are required at both levels. but it is not equipped to deal with the rather suspicious data that lies between the two outlier segments that it did recognize. Considerably more can be done to detect. or on the basis of data validity tags produced by the PDC as accessory output.GPS Synchronization & Timing DITTMER CONTROL CENTER CANADA PMU GRAND COULEE PDC #1 PMU COLSTRIP PMU JOHN DAY PMU MALIN PMU SYLMAR jfh MEXICO HVDC Terminal Fig. flag. The signal in Figure 6-20 demonstrates that so elementary an approach is not always enough. However. It also recognized and repaired 5 later points (near 4000 seconds). 1997. most of this is better done at PDC level. Configuration of BPA’s Phasor Measurement Network for brake test of September 4. For analysis. however. In this case the repair algorithm recognized a “blank” segment of 498 points and performed a linear interpolation across it. Provisions for (degraded) local control for communication failure is normally required. 6-19. For control.Raw Plot for Malin Round Mountain #1 MW 1400 Brake insertion #1 1200 1000 1 blank point 800 1 blank point 6 blank points 600 400 Brake insertion #1. detection of bad data typically causes freezing of the signal at the last good value. 09/04/97 Data collected on Dittmer PDC sample rate = 30/second 200 0 0 50K 100K 150K Time in Samples Fig. 6000 patched data Phasor Magnitude linear interpolation Colstrip Broadview #1 raw PDC file = 09050425. Sustained bad data may cause wide-area control suspension. 6-20. 6-21 .MAT 4000 suspect data 5 outliers 2000 blank data (498 points) 0 0 2000 4000 Time in Seconds 6000 8000 Fig. Raw data from PDC recording segment #1. Partial repair of a PDC file with modem retraining. 07. Field evaluations have been successful [6-35. The measured fiber optic link time delay was about 21 ms. 6. CIGRE Brochure 111.. January 1989.g. These techniques may make direct load control (e. compatibility with legacy electromechanical relays and meters. BPA has several phasor measurement links employing modems and analog microwave. wide dynamic range. heaters.6-10]. Advantages include smaller device size and weight. Some designs are “self-healing. related to change out of existing instrument transformers. The measured modem/analog time delay (latency) averaged about 70 ms.01. Acknowledgement: Much of the material in this chapter is based on findings of the DOE/EPRI Wide Area Measurement Systems (WAMS) project [6-9. pp. References 6-1 J.11 Optical Sensors The transducers described above normally use the outputs of conventional instrument transformers (magnetic current and voltage transformers. corresponding to about 7° of a 1 Hz signal. air conditioners) for stability more practical. elimination of hazardous oil-filled transformers. Optical sensors may be used with digital IEDs in the substations of the future. wide bandwidth. This is a very important development that greatly facilitates wide-area stability control feasibility. are commercially available from several manufacturers. and one link employing fiber optics. Optical voltage and current sensors. potentially lower cost. Analysis and Control of Power System Oscillations. elimination of instrument transformer burden limitations. Voltage and current sensors may be combined in a single device. Other emerging communication techniques promise fiber optic like performance. Another is digital communications over power lines. One is low earth orbit satellites [6-34] and possibly other wireless technology. 6-22 .6-36]. The advantages of optical sensors are only indirectly related to use of digital transducers for stability controls. “Robust Damping Controls for Large Power Systems. and capacitor voltage transformers).6. F. December 1996.10 Future Digital Communication for Stability Control Power companies and telecommunication companies are rapidly installing long distance fiber-optic communication. electrical isolation. 12–19. Material is reproduced here with the permission of BPA. 6-2 CIGRE Task Force 38. Challenges are mainly economic. however.” IEEE Control Systems Magazine. Direct load control is described in the next chapter. high accuracy. Hauer. and the volume production needed for cost reduction. and compatibility with digital technology. elimination of substation secondary electrical cabling.” with transfer to an alternate path in 50–120 ms for a failure. Hickman and J. Hauer. “Modeling and Analysis Guidelines for Slow Transients: Part 1 (Torsional Oscillations. Hauer.” IEEE Computer Applications in Power.” IEEE Transactions on Power Delivery. D. to be published in IEEE Transactions on Power Systems. Taylor and D. Damsky. 4. Mittelstadt. IEEE Publication 95 TP 101. and W. J. pp. “BPA Experience in the Measurement of Power System Dynamics. October 1995. 10. “General Characteristics of Power System Transducers. T. Transient Torques. October 1997. W. F. J. “Recording and Analyzing the July 2 Cascading Outage. Recife (PE) Brazil. D. Hauer and J. 11. Overholt. D.S. W. Vol.” IEEE Transactions on Power Delivery. 6-9 W. “The DOE Wide Area Measurement System (WAMS) Project—Demonstration of Dynamic Information Technology for the Future Power System. A. 6-7 J. Hauer.” WAMS Working Note. July 1996. C. Piwko. Vol. 1995. Litzenberger. H. Overholt. W. No. F. 3. No. W. W. 6-5 J. 6-8 D. 1. 158– 163. L. “Validation of Phasor Calculation in the Macrodyne PMU for California-Oregon Transmission Project Tests of March 1993. C. Turbine Blade Vibrations. 1996 WSCC System Outage. pp. Rizy . Hunt.. P. Mittelstadt.” Inter-Area Oscillations in Power Systems. Kosterev. Litzenberger. J. “Keeping an Eye on Power System Dynamics. D. R. J. Eden “Modulation and SSR Tests Performed on the BPA 500 kV Thyristor Controlled Series Capacitor Unit at Slatt Substation. 1995. pp. This report and attachments are available from BPA on compact disk. F. January 1997. in association with the WSCC System Oscillations Work Groups. Wilson. Hamai. Sobajic. N. 6-11 M.” V Symposium of Specialists in Electric Operational and Expansion Planning (V SEPOPE). W.6-3 J. “Model Validation for the August 10. A. 11. E. Vol. and D. April 9–11. D. Mittelstadt.C. pp. Department of Energy by Bonneville Power Administration and Western Area Power Administration. Hauer. 6-13 J. C. “Extending the Realism of Planning Models for the Western North America Power System. A. pp. Hauer. A. Mittelstadt. Trudnowski. 1224–1231. 10. Taylor.” IEEE Computer Applications in Power. 26–30. 50–54. A. Erickson. et al. Rogers. R.” IEEE/PES paper PE-226-PWRS-0-161997. Clemans. April 1999. M. H. G.. J. and P. pp. and J. F.” EPRI Conference on the Future of Power Delivery. Vol. 1996. 6-10 J. R. W.” IEEE Transactions on Power Systems. P. N. F. Irvani. Krause. attachment to reference 6-10. 1950– 1955. 6-12 B. May 1996. 6-4 J. F. 6-23 . Hauer. Fast Bus Transfer). F. 801–806. May 19–24. Hauer. Wide Area Measurements for Real-Time Control and Operation of Large Electric Power Systems: Evaluation and Demonstration of Technology for the New Power System. J. B. December 20. Report prepared for U. and J. Washington. 1996. 6-6 C. R. No. F. Mittelstadt. E. Johnson. attachment to reference 6-10 6-20 J. Bendat and A. June 5. 1992. A Guide to Circuit Simulation & Analysis Using PSpice.” Third Virginia Tech Conference on Computers in Electric Power Engineering. pp.” WAMS Working Note. March 3.” Electrical Business. 6-17 J. Thorpe. VA.” WAMS Working Note. Computerized Circuit Analysis with SPICE. W. 1997. 1995. Griess. “Signal Processing Aspects of Power System Transducers.” BPA Working Note. Manolakis. E. October 27–29. J. 1996. 1995. and B. “Phasor Measurements on the BPA Transmission System. 1988. Digital Signal Processing—Principles.” WAMS Information Manager Working Note. 1996. 6-24 . Hauer. S. “A Preliminary Report on Transducer Measurements Performed at Slatt Substation on April 24. “Performance of the PML 7700 ION Programmable Transducer System. The Fast Fourier Transform and Its Applications. Engineering Applications of Correlation and Spectral Analysis. F. June 18. March 12. 6-23 MATLAB – High-Performance Numeric Computation and Visualization Software (Reference Guide). F. attachment to reference 6-10 6-16 E. Prentice Hall. as Tested at BPA's Big Eddy Substation on October 25 1996. F. “Research Database from BPA’s Phasor Measurement Network for Test Insertions of the Chief Joseph Dynamic Brake on September 4. attachment to reference 6-10 6-21 Thomas W. and B. SPICE. 6-28 R. Phadke. “Nonintrusive Procedures for Measuring Dynamic Performance of Enhanced Transducers at Slatt Substation. John Wiley & Sons. R. Martin. Hauer. April 1993. January 1997. 6-22 P. 6-25 R. Inc. 14–15. 1992. Mass. Hauer. The Math Works. May 1997. Hickman. Algorithms. Tuinenga. NJ: Prentice-Hall. 6-18 J. P.. G. Piersol. pp. attachment to reference 6-10. O. 6-24 A. E.. G. Englewood Cliffs. December 7.. 6-19 M.” IEEE Computer Applications on Power Systems.” WAMS Working Note. “Synchronized Phasor Measurements in Power Systems. Proakis and D. Third Edition. 6-27 J. Hauer et al. Carolsfeld. G. G. Dagle.6-14 J. Brigham.” WAMS Working Note. 1997. 6-26 K. S. F. “GPS Synchronized Power System Phase Angle Measurements Recorded During 500 kV Staged Fault Testing. New York: John Wiley. 10–15. Natick. New York: McMillan. Donnelly. Wilson. Bunch. K. W. 1992. “To Measure is to Control. 1993. 1998. Sterlina. Arlington. 1996. and Applications (Second edition). 1980. 6-15 J. “Satellites Free the Mobile Phone. Hall. Vol. March 3. 9. pp. 17–24. L. “Experience with Dynamic System Monitors to Enhance System Stability Analysis. 6-31 J. K. Mittelstadt. W. Tillett.org/spectrum/mar98/features/leo. 693–701. F. D. J.” IEEE Transactions on Power Apparatus and Systems. Washington USA. C. W. W. Arlington. 4. D.” IEEE Spectrum. Hauer. L. “Phasor Measurement Unit Performance Tests. Smith. and D. N. “Damping of Pacific AC Intertie Oscillations via Modulation of the Parallel Pacific HVDC Intertie.” IEEE Transactions on Power Systems. “Experience with Optical PTs and CTs for Relaying and Metering. W. PWRS-7. L. Taylor. Chatrefau. “Operating Experience with Modulation of the Pacific HVDC Intertie. and D. Pease. 6-36 D. pp. Sutphin. Martin and R. Scott. 6-35 J. May 1992. No. 1998. Clark. October 1996. July/August 1978. 33–36. Vol. Scott. Loutan. 1/97.ieee. “Application of Optical Sensors in Extra High Voltage Substations. 6-32 R. No. and C. A. Gupta. 6-37 K. Miller. Taylor. pp.” IEEE Computer Applications in Power. Spokane. N. Virginia. 1997. J. Mittelstadt. Kwee.. 6-33 R. pp. 6-34 B. Vol.” GEC Alsthom T&D Review. and C. November 8–10. 3. K. 6-25 . A.html).spectrum.” Proceedings of Precise Measurements in Power Systems Conference. Research Database from BPA’s PPSM Network for Test Insertions of the Chief Joseph Dynamic Brake on September 4. 6-30 H. WAMS Information Manager Working Note. Bradley. et al. 1978. R.6-29 H. “Substation Automation Problems and Possibilities. October 23–25. attachment to reference 610. 1053–1059. 35. Cresap.” CIGRE 14-05. March 1998 (http://www. R. 1995. PAS-98. Vol.” Proceedings of Western Protective Relay Conference. Cresap. Section 7. depending on the actual hydrologic conditions. respectively. Technical and economical feasibility of the interconnection was studied since 1992. 7.300 MW. Main aspects. together with extensive study results. . The installed generation capacity in South/Southeast and North/Northeast systems is about 48 GW and 14 GW. A TCSC control system for transient and dynamic stability improvement was designed and. with suitable operation required from no load up to maximum flow in both directions. The north–south interconnection connects Imperatriz substation (in the State of Maranhão) to Serra da Mesa (in the State of Goiás).5 presents energy source power system stabilizers.3 explores how new distributed-measurement technology can be used to improve dynamic and transient system stability. Power flows will occur in both directions. See Figure 7-1. and eliminated the technical restriction on the AC transmission alternative.2 presents the control analysis of a potential 500-kV TCSC installation in China. Section 7.020 km long. In both cases. Section 7. formed the basis for the TCSC locations and equipment specification. the interconnection links the 500-kV substation of Imperatriz (north system) to the Serra da Mesa power plant (south system).1 Brazilian North–South Interconnection—Application of Thyristor Controlled Series Compensation (TCSC) to Damp Interarea Oscillation Mode This example deals with a pioneer commercial application of TCSC to damp the low frequency interarea oscillation mode in the Brazilian north–south interconnection [7-1]. The interconnection is a single 500kV line and is 1. Use of two small TCSCs (6% compensation each) proved to be very effective in damping the interarea mode. The line is designed to transmit up to 1. There are two main electric power systems in Brazil which were not previously interconnected: the south/southeast (or south system) and the north/northeast (or north system) systems. Section 7. Two transmission alternatives were considered and analyzed to establish the North–South Interconnection: a DC ±400-kV bipole and a single 500-kV AC compact transmission line (4x954 MCM bundle).020 km long. The “North–South Interconnection” will exploit hydrologic diversity between the systems.4 describes active-load modulation to improve stability. The interconnection was commissioned in early 1999 and reduces the risk of energy deficits. 1. They are essentially hydroelectric systems and include more than 95% of the total national production and consumption. achieving energetic benefits estimated at about 600 MW-year. The first example is from Brazil where thyristor controlled series compensation (TCSC) is used to damp interarea oscillations.Chapter 7 Applications of Advanced Controls This chapter gives examples of engineering projects where advanced control methods have been used or studied. On the other hand. 7-1. this alternative presented significant advantage in terms of costs. and other 500-kV AC transmission links are planned to cater for this additional generation. poorly damped interarea oscillation mode. When comparing the technical behavior of the two alternatives. and for future generation developments located over a vast geographic area having enormous economic potential. this long. From a purely technical viewpoint. This oscillation of wide amplitude (± 300 MW) represented a serious technical restriction for the AC alternative. Brazilian North–South Interconnection—geographic location.18 Hz). From a strategic and political viewpoint.5 to 2. Six hydroelectric plants may be built along the same route in the next two decades.Fig.0 Hz has been solved by power system stabilizers (PSS) in the main synchronous generators. however. low capacity interconnection between two large systems having different planning and operating criteria is a textbook application for HVDC transmission technology. the problem of electromechanical oscillations in the range of 0. 7-2 . Traditionally. the AC transmission alternative is highly attractive for making inexpensive hydroelectric energy available to a rapidly growing area. besides the strategic and political benefits mentioned above. it was verified that the AC solution presented a low frequency (0. 3. Figure 7-2 shows simulation results. simulation software and implementation. To solve the sustained oscillation problem thyristor controlled series compensation (TCSC) was proposed in the interconnection (transmission line Imperatriz–Serra da Mesa).3 Hz). Practical limitations on maximum PSS gain at very low frequencies may reduce the damping of these modified stabilizers. not having any effect on the other modes presented in the system. Power will be transferred from Yimin plant in Mongolia. So if this link is disconnected. to the load centers through a 500-kV parallel transmission line covering a distance of 1300 km. Because of future development of the transmission system. 7. would not always ensure adequate damping for the north–south mode for the various scenarios considered in the study. The modified PSSs. Modified PSSs would be needed in all major power plants of the northeast system. 7-3 . the controller design must be systematic and robust. The main drawbacks of this solution for the north–south interconnection are listed below [71]: 1. Electromechanical oscillations within the northeast system (local modes) and between the north/northeast systems (interarea mode) could have their damping reduced by the action of the modified PSSs. One great advantage of this solution is the fact that the TCSCs are located in the link that introduces the interarea mode and they are tuned only for this mode. 5. assumed to be of fixed structure and fixed parameters. The frequency range of electromechanical oscillations to be damped by the modified PSSs is too wide to yield reliable operation. 4. The TCSCs at each end of the intertie are modulated using local line power measurements. This solution was much more efficient than PSS in providing damping for all possible system scenarios and contingencies. transient and voltage stability.For lower frequency modes (< 0. 2. The theories and the schemes must be applied to the real engineering project and must be easy to manipulate. Commissioning tests verified the powerful damping performance of the TCSCs [7-2]. • It’s required to increase dynamic. effective damping is a difficult task. control algorithms. The paper is motivated by the real engineering project and presents on-going research for TCSC models.2 Analysis and control of Yimin–Fengtun 500-kV TCSC system References 7-3–5 present research done for the thyristor controlled series compensation (TCSC) to be situated on the main corridor of the 500-kV transmission system of northeast China. the TCSCs together with the interarea mode cease to exist. The stability of the two isolated systems (north and south) in this case is guaranteed by PSSs exactly as before the advent of the interconnection. however. The Yimin–Fengtun TCSC projects has the following distinctive features: • It’s located on an important corridor of the main grid. with 2200 MW capacity. the research should be broad. It should adapt to not only the variation of operation conditions. ultimately. fuzzy control. The combined effect of electromagnetic and electromechanical transients has been studied. Simulations show the effectiveness of the control schemes. The influence of the voltage protection of the metal oxide varistor (MOV) is included in the control model and design. Thin line without TCSCs. 7-2. 7. state-space identification algorithm to identify a reduced-order small-signal MIMO model of the 7-4 . the system suffers severe transient and dynamic instabilities. increases the transmission capacity. • Because of the project’s importance. and nonlinear adaptive scheme are studied. numerical sub-space. Simulation of fault with line outage in south system [7-1]. but also future changes of grid topology.seconds Fig.MW 1000 800 600 0 5 10 15 Time . Reference 7-6 explores new ways of putting this extended real-time knowledge of the power system behavior into use by means of supplementary feedback loops which improve dynamic and transient system stability and. The first step is built on a powerful pulse response-based.1200 Power . Without a TCSC. and improve transient and dynamic stability. • The controller must be robust. It’s important to have control schemes that require local signals only and are independent of system models. The design of such advanced controllers is based on a two-stage methodology. thick line with TCSCs. The study shows that the TCSC with proper control schemes can schedule power flow flexibly. Autodisturbance rejection control (ADRC).3 Wide-Area Stability Control New distributed measurement technology using the global positioning system and accurate phasor measurements units have developed steadily in recent years to become the most powerful source of wide-area dynamic information. each consisting of a high-order differential filter. the architecture selected comprises several dynamic feedback loops. the local loop alone was sufficient to stabilize the system.000 MW) Fig. Decentralized/hierarchical PSS located at the Duvernay synchronous condensers (SC). The three inputs of the PSS are the following: y = [ f 294 θ 4 − θ 7 θ 8 − θ Ref ] where f 294 is the local bus frequency at Duvernay. a three-loop stabilizer was designed for a major synchronous-condenser station in an actual power system that simultaneously uses two global and one local input signals. Figures 7-4 and 7-5 provide some interesting clues as to what added value should be ascribed to the information exchange paths outlined on Fig. 7-3.θ ref f 294 PSS θ4 . which in some cases could pay for the implementation costs and cover the additional risks inherent in long-distance telemetry.open-loop system. although its positive action provided 5–10 seconds relief before actual breakdown.The target PSS is at the Duvernay synchronous condenser in the reference area. θ8 . it was unable to do the same for the second and third. Area 750 MVA SC Σ + Vref + SCADA Wide-Area Measurements External grid (25. To tackle the most difficult situations. The second step is to select an appropriate control structure. Therefore. information exchange really has some monetary value.θ7 Duvernay Ref. Controller tuning is then performed by minimizing a selective modal performance index in the parameter space. 7-5 . and θ 4 − θ 7 and θ 8 − θ Ref are angle shifts between the subscript areas. 7-3. Based on three typical contingencies. For illustration. The architecture of the Hydro Quebec test system used in reference 7-7 is recalled in Figure 7-3 . Adding stability and robustness constraints greatly improves the engineering significance of the resulting design. On the first contingency. However. and then tune the stabilizer parameters accordingly. Both linear and nonlinear simulation results clearly demonstrate the added value of wide-area information when properly included in power system stabilizer design. This site was chosen because it shows the highest controllability index over the broadest frequency range. (Deg. 7-4.(Deg.t Ref.r.r. First contingency: The local loop alone prevents the system from collapsing.Duverney frequency deviation (mHz) Unstable fault in the western corridor (BUS #783) 0 −100 −200 −300 −400 No PSS LPSS+GPSS LPSS −500 −600 −700 0 5 10 15 20 25 30 35 Angle shift: 8−Ref (Deg) s −80 −100 −120 −140 −160 0 5 10 15 20 25 s Angle Shift of area #8 w. 7-6 . 7-5.) Fig. but doesn’t prevent instability. (a) Unstable fault in the eastern area (Bus #706) −60 −80 −100 −120 −140 −160 0 5 10 15 20 25 Angle Shift of area #8 w.) s (b) Unstable fault in the western area (Bus #780) −60 No PSS LPSS+GPSS LPSS −80 −100 −120 −140 −160 0 5 10 15 20 25 s Fig.t Ref. Second and third contingencies: The local loop alone substantially improves the system performance. corresponding to a 180 degrees phase change. implementation of discontinuous control schemes show good prospects. Although more difficult to design. the grid configuration was changed so the hydro power station had to feed its power through a weak distribution system before connecting to the main grid A way to verify the damping effect of load switching is to change the sign in the controller.7. The angle difference between the external net and the estimated generator internal EMK was used to control the load switching.4 Active-Load Modulation for Stability Control The first subsection describes large-scale load modulation and the second subsection presents field tests at a small hydro station in Sweden. Large-scale load modulation. operating experience. The tests were done to investigate if load switching could be used to damp power oscillations. 7-7 . In developing the case. At a time when the cost effectiveness of power electronic devices for damping interarea oscillations is constantly being questioned. The load used was pure resistance without any dynamics and was dedicated for this purpose. Reference 7-8 presents field tests performed at the hydro power station Hemsjö Övre the night of 24 and 25 September 1996. Reference 7-7 describes how angle stability can be improved by large-scale active-load modulation. To make the generator susceptible to power oscillations. The idea is to switch a resistive load so it counteracts power oscillations. especially for decentralization and robustness against communication delays. The results show that load switching is an excellent method of damping power oscillations. A change of sign in a well-tuned regulator can induce power oscillations. Damping of Power Oscillations by Load Switching—Field Tests at Hemsjö Hydro Power Station. Two measurements were done to compare damping. The first case is without load switching and in the second case is with load switching. and simulation of a large power system is used to demonstrate that active-load modulation can improve system dynamic performance to a large extent. During the first 10 seconds in Figure 7-6 and 7-7 the regulator sign was changed. Analysis. it was found that continuously modulating load stabilizers need global signals for full effectiveness. it’s natural to look to active-load modulation as a potential alternative method of ensuring grid reliability. with just a fraction of the base load available for control. Figure 7-6 with time > 10 second shows the damping without load switching. The figure clearly shows that the load switching builds up a power oscillation with increasing amplitude. 7-7.02 0.4 0.015 0. and in controlling the power of batteries [7-9] and steam-turbine generators in response to these system oscillations. Generator Power in MW P_gen [MW] 0. The ESPSS can be applied on energy storage systems such as superconducting magnetic 7-8 .3 0 5 10 15 Time [s] 20 25 30 25 30 t24. The objective of the ESPSS is to damp low frequency electromechanical oscillations between large interconnected power systems.005 0 0 5 10 15 Time [s] 20 Fig. thereafter no switching (9–14 s).025 P_load [MW] 0.025 P_load [MW] 0.01 0.5 0.015 0.4 0.01 0.t20. Power to Controlled Load in MW 0. Figure 7-7 shows the damping when load switching is used. Power to Controlled Load in MW 0. 7.3 0 5 10 15 Time [s] 20 25 30 25 30 t20. then damping by controlled load switching (14–30 s). t24. It is evident that controlled load switching improves damping considerably.02 0.005 0 0 5 10 15 Time [s] 20 Fig. Excitation with load switching (1-9s) and thereafter no switching.5 Active Power Modulation of Generators and Energy Storage for Oscillatory Instability Control Energy Source Power System Stabilizer (ESPSS). Field tests and monitoring have demonstrated the ESPSS performance in sensing system disturbances. Excitation by load switching (1–9 s). 7-6. Generator Power in MW P_gen [MW] 0.5 0. Note that the power of the switched load is only a fraction of the oscillation amplitude. which controls the real power output to counteract these oscillations. on the tightness of voltage control. and the effectiveness depends on the location and characteristics of load. It can be effective and robust in damping low frequency modes present in the speed signal. The state-of-the-art PCS using Gate-Turn-Off (GTO) thyristors are very fast acting and have the capability to accept both MVAr and the MW power orders. The ESPSS. In damping interarea modes conventional PSS essentially modulates voltage-sensitive load. The ESPSS controls the MW output only.2–0. with less dependence on variable network and load characteristics. In the case of a battery energy storage system (BESS). At Alamitos Generating Station in California. These factors affect the component of electrical torque in phase with (modal) speed that produces damping. The power import capability and the reliability can be increased significantly by damping the interarea power system oscillations that often limit such imports. The ESPSS can be applied on the battery or superconducting energy storage by controlling the power conditioning system (PCS) which converts power between AC and DC. Since the aim is to provide damping torques to generators. modulation of 5 percent of the turbine power 7-9 . By modulating two valves. Tests were conducted by injecting the modulating signal in one and two different valves of these eight valves. Modulating two valves gave almost twice the modulated power output change compared to one valve. For generators. the input signal can be derived from speed/frequency and electrical power measurements. The opening and closing of the valves are controlled to obtain maximum operating efficiency and control. a much larger BESS or SMES is required to effectively damp and stabilize the system. and on the mode shape [7-11]. ESPSS installation on electric power generators.8 Hz range). The ESPSS concept is to produce damping more directly by modulating the mechanical input power instead of generator voltage and reactive power. Thus the ESPSS concept can be extended to the other steam-turbine governors. The steam control is obtained from eight valves. generators 5 and 6 steam turbines are crosscompound units and the steam flow is controlled on the high-pressure side. Field tests conducted on a turbo-generator with a state-of-the-art governor showed that steam-turbine governors can respond fast enough to provide damping of low frequencies oscillations (0.energy storage (SMES) or battery energy storage systems. This is by adding a speed or frequency deviation based signal into the governor valve controls. the most effective location of energy storage is close to generators participating in low frequency oscillations. the ESPSS differs from conventional excitation equipment PSS in that it acts on the mechanical input power of the generator. However. Appendix J further describes mechanical versus electrical side damping. ESPSS processes the frequency deviation signal or similar signal to control the megawatt output or input of the batteries. With a large power source it would be possible to damp the oscillations with even a small change (5 percent) of the generator output. and generator loading. can provide effective damping. Similar to PSS. Tests conducted at the 10 MW Chino battery energy storage system [7-10] demonstrated damping capability with measurable results. It cuts off the excitation system PSS system when it operates as shown in Figure 7-10. The ESPSS acts only for large system disturbances. The phase lag increases as the frequency of the modulation signal increases. the gain also drops rapidly making this control loop ineffective at these higher frequencies. 7-10 . This phase shift includes delays in the steam circuit such as the steam chest. Figure 7-9 shows the response curve for the excitation system of a similar machine. The advanced state-of-the-art governors and the lower interarea oscillation frequencies have made this modulation feasible. The modulation input to the valve is dependent on the frequency excursion from 60 Hz and can be adjusted by changing the gain of the ESPSS. For damping control design. Governor frequency response with signal input into two valves. the phase shift between the injected input signal and the power output increases to about 100 degrees. Although efforts to implement these controls were made in the past. The phase shift in this case is increases much more rapidly. the frequencies that were attempted were mostly local mode oscillations and were in the range of 1. Two ESPSS have been developed and installed at Alamitos generating units 5 and 6. it had not been feasible because the governors were generally slow. However.(about 24 MW) can be achieved. Also.7 Hz.0 Hz. the transfer function between the valve input and mechanical power is required. The modulation control includes phase compensation of the steam circuit lag so that the change in mechanical power is closely in phase with generator speed changes for oscillation frequencies of interest. and this can be computed from measurements of electrical power and speed. At 1 Hz. 7-8.0–3. Figure 7-8 shows the gain and the phase relationship of the governor loop measured by changing the input into the governor control board and monitoring the megawatt change in the machine output. Fig. increasing to about 180 degrees at about 0. With this constant air flow rate. In so-called “industrial” combustion turbines. the turbine compressor and power turbine are mechanically coupled to the synchronous generator and thus turn at a speed that is constantly proportional to synchronous speed. Rapid modulation of fuel flow in combustion turbines. 7-9. 7-10. This arrangement. the turbine power changes within milliseconds of changes in fuel flow into the combustors allowing the rapid power generation change that may be used to improve transient and oscillatory instability. maintains constant air flow through the entire unit.Fig. also called “single-shaft” combustion turbines. Fig. Functional block diagram for an integrated power system stabilizer. Excitation system frequency response for Alamitos generating unit. it’s possible for such a unit to shift from synchronous 7-11 . For instance. 2. pp. Liang.Theory and Field Test. 7-5 X. 7-3 X. 146. Kamwa. J. pp. P. Transm. 7-12 .operation at minimum generation. Liang. 7-4 X. The ancillary services could include system stability (damping). 7-8 O.” IEE Proc.” IEEE Transactions on Power Systems. Samuelsson and M. Grondin.-Gener. Zhou and J. Zhou. Ricardo..” Electric Power Systems Research. 46.. Gérin-Lajoie. to full nameplate power in less than a second by rapidly increasing fuel flow. June 1998. Tenório. Fraga. “On-Off Control of an Active Load for Power System Damping . Edmonton. Asber. approximately 30 % of nameplate power. Gingras. and G. Leoni. Distrib. No. 2. “Nonlinear Adaptive Control of TCSC to Improve the Performance of Power Systems. M. Trudel. The drawback to this process change lies in the fact that the power turbine temperatures also change rapidly with changes in firing rate. L. Gama. Utilities have experimented with these concepts and with rapidly increasing firing rate during emergencies such as loss of large blocks of generation. 146. 157–168. B. et al. J. Gribel. Ping. pp. Trudel. paper 14-101. Eiras. May 1999.” CIGRÉ. 7-7 I.” presented at American Control Conference. R. Transm. A. Akke. No. No. “Brazilian North-South Interconnection — Control Application and Operating Experience with a TCSC. May 1999. and R. For twosided modulation (increase and decrease of power). modulation of gas turbines should be possible without damaging temperature excursions. March 1999. below maximum power.” IEE Proc.” Proceedings of IEEE/PES 1999 Summer Meeting. Vol. 301–305. Gama. and increased reactive power production or reactive power reserve. the gas turbine generator would have to be operated at lower efficiency. 582–590. Cavalcanti. This might require an ancillary service arrangement. Similar to the above descriptions of small (5%) modulation of steam turbines. No. 1103–1108. 14. Vol. Vol. “Analysis and Control of Yimin–Fengtun 500 kV TCSC System. R.” IEEE Transactions on Power Systems. Distrib. W. pp.-Gener. “Large-Scale Active-Load Modulation for Angle Stability Improvement. 7-6 I. 125–130. D. 3. Kamwa. pp. “Overview of Control Schemes for TCSC to Enhance the Stability of Power Systems. and G.. 1998. and the erosion rate of power turbine blades is seriously increased with large. 7-2 C. primary and secondary spinning reserve. 1998. but we know of no in-service applications. J. rapid changes in firing rate and temperature. J. References 7-1 C. 18–22 July 1999. Zhou and J. R. “Brazilian North–South Interconnection — Application of Thyristor Controlled Series Compensation (TCSC) to Damp InterArea Oscillation Mode. “Multi-Loop Power System Stabilizers Using Wide-Area Synchronous Phasor Measurement. with compensation for the lost power sales and lower efficiency. Vol. Germany. 7-9 B. 1997. No. 608-613. 2. pp. 26. Edmonton. July 20-24. 63–72. pp. Walker. 7-13 . 7-10 L. No. “10-MW GTO Converter for Battery Peaking Service.” Proceedings of IEEE/PES 1999 Summer Meeting. “Effective Use of Power System Stabilizers for Enhancement of Power System Reliability.Vol.” presented at IEEE Summer Meeting in Berlin. Bhargava and G. Dishaw.” IEEE Transactions on Industry Applications. 18–22 July 1999. 7-11 P. May 1999. 96– 103. “Energy Source Power System Stabilizer Installation on the 10 MW Battery Energy Storage System at Chino Substation. 1. January/ February 1990. Kundur. H. 14. pp. The recent abundance of natural gas and the rapid progress of combustion turbine and combined-cycle technology has drastically changed the economics of generation.1. or gridlock at toll plazas. transmission. The dynamic performance of power systems. Unfavorable hydrological conditions . So while competition will force the evolution of the most economic generation additions. The restructuring of the power industry requires establishing requirements for new generation equipment and controls. is one such issue. and loads. In Brazil for example. and control additions was well structured. there will still be some aspects of dynamic characteristics requiring cooperation dictated by the effects on overall system performance. Inadequacies in the generation/transmission plant can result in system collapse with unacceptable consequences. that is the ability of maintaining reliable and stable supply within tolerable limits of voltage and frequency. distribution. 8. Unlike the case of other industries such as communications and transportation where overloads result merely in telephone busy signals. This was due to the high relative cost of generation. with mere estimates on transmission feasibility. control and protection.Chapter 8 STABILITY CONTROLS WITH INDUSTRY RESTRUCTURING Industry restructuring from a highly centralized hierarchical and possibly state-owned system to a new model characterized by competition in generation with guaranteed access to transmission has many impacts on power system stability. the worldwide trend to deregulation has opened the power generation industry to independent producers. and requires administering the required ancillary services in the new operating environment. Transmission planning then proceeded to accommodate an already established generation master plan. is a function of the joint characteristics of generation. Power must be supplied the moment a switch is turned on. power consumption is instantaneous. the plans of generation additions were established almost independently.1 Some Examples of New Scenarios 8. In the traditional approach of integrated planning of bulk electric systems by a centralized company or agency. Stability control. transmission. including the equitable allocation of associated costs.1 The Brazilian electric system The predominantly hydro Brazilian system spans large geographical distances and has most of its generation remote from load centers. the decision process on generation. Concurrently. The aim was an optimum allocation of investment in the various segments so as to achieve a prescribed level of reliability at minimum cost. • Large hydroelectric dams.000 MW which is predominantly hydro (95%). whose decisions are mandatory. • Frequent operating conditions with heavy energy transfers. even during light load. has 150. Up to now. even during light load conditions. generation dropping. Munhoz (418 MW).5 million being residential consumers. and automatic switching of shunt compensation. and associated costs for stability controls in a competitive environment. Other Brazilian system characteristics are: • Large capacity hydro power plants remote from the load centers. the national coordinating pools for planning and operation of the Brazilian power system. The energy consumption per capita is 1. There is therefore need to investigate these aspects in order to establish guidelines. G. Stability problems are therefore naturally aggravated during these conditions. it’s common to have reversals in the power flow of some transmission lines and transformers. with 97% being from hydro. due to hydroelectric generation coordination for optimal water usage. Attributing responsibility for a given stability problem and distributing the costs of candidate solutions are very complex issues with opposing opinions. responsibilities. The energy production is of the order of 309 TWh. during 1996/1997). the transfer of large blocks of energy between generating subsystems having hydrological diversity is carried out mainly during light load conditions. • High load growth (6% per year. dynamic voltage controls. During unfavorable hydrological conditions. HVDC controls. The Brazilian electric system has an installed capacity of 56. etc. In the old system structure stability problems were detected and resolved by GCOI/GCPS. having up to 5 years storage capacity for good regulation of variable inflows. with consequent need for urgent generation expansion. Itumbiara (380 MW). In some parts of the system.000 km of transmission lines of voltage levels from 138-kV to 765-kV.frequently call for high power transfers between regions. and an ISO (Independent System Operator) changes this picture. The introduction of IPPs. • Long transmission lines. • Hydro units of large capacity: Itaipu (700 MW). cogeneration. The stability controls considered include control of system oscillations (PSS). • Delays in construction of high capital investment power plants.954 kWh/year for residential consumers. the allocation of costs of stability controls has been decided jointly by the GCPS and GCOI. underfrequency load shedding. There are about 40 million consumers. controlled islanding. In a few cases involving highly unfavorable hydrological conditions. sometimes presenting bottlenecks in some transmission corridors. the system has operated with violations of the 8-2 . with 32. New scenarios for the Brazilian electric system. • The technology development of combustion-turbine driven power plants makes the combined-cycle power plant one of the most efficient forms of power generation. mainly in the northeastern part of the country. released every year by GCPS. These power plants cause low environmental impact. Note that in these cases the system must still meet the criteria for small-signal stability to avoid spontaneous oscillations. initially with Argentina. the natural gas has very low levels of sulfur. The last Ten-Year Plan. biomass (sugarcane leftovers). it’s possible to envision the following scenario: • Significant increase in thermal generation. The above scenario is very likely because: • The country needs to increase its generation capacity in the immediate and near future. which are composed of managers from Eletrobras and all the other Brazilian utilities—the GCPS (Coordinating Group of System Planning) and GCOI (Coordinating Group for Interconnected Operation). These two groups perform stability studies. The gas supply to these power plants is guaranteed by Petrobras (newly-discovered 8-3 . through long distance or back-to-back HVDC links. offering very-competitive energy prices. estimates that the rise in electrical energy demand in the period 1997– 2006 will call for the installation of an additional 3200 MW of generation every year. • Operation of the existing system closer to its maximum limits. • Implementation of several international interconnections. and the burners can meet the most severe environmental legislation. Another advantage is that gas turbines can be located close to the load centers. The turbine has acoustic insulation. with negligible levels of audible noise. • Utilization of alternative energy sources: Wind power and solar generation. atmospheric pollution and emission of liquid or solid waste. so as to prevent severe power shortages. • Significant rise in distributed generation. there are two coordinating bodies for the expansion planning and operation of the interconnected systems. therefore minimizing high investments in long distance transmission. • The newly implemented legislation regarding IPPs and the open access created favorable conditions for these new agents. the government stimulus to private investors. Two immediate questions appear: a) How will the transmission system evolve? and b) What will be the expansion process for this additionally needed generation? Taking into account the ongoing restructuring process of the Brazilian electrical industry. mainly gas turbines.existing criteria of transient stability. and establish recommendations concerning the required control actions for system stabilization. The reduced construction period for gas turbine power stations is ideal for rapidly commissioning the needed additional generation. Uruguay and Bolivia. and the highly developed technology for combustion turbines. As stated before. It’s expected that in the next ten years gas turbines will represent 10% of the total installed capacity. A significant amount of gas-fired generation will be operating close to the major load centers. Considering the various oil fields and refineries owned by Petrobras where the gas is currently being burned. and part of Denmark. there are the effects of major changes in the generation scenarios over the next two to three years. and private agents prefer investments that can be recovered in shorter periods. Sweden. commissioned February 1999. this interconnection has TCSCs for enhancing stability [8-6].gas fields as well as imported gas). There will also be gas pipelines within the Brazil to distribute gas. While Norway has almost 100% hydro generation. Undoubtedly. In addition to the above factors and uncertainties that have a major impact on the overall system dynamic performance. In this case. This requires. • The north–south interconnection. has strong connections to Nordel through several HVDC links. 1. long distance transmission with the need for stabilization actions will be required. It’s a 500-kV circuit. the solution for this problem is to form partnerships between private investors and Eletrobras. Finland. This interconnection together with those with Argentina will cause some areas of the system to operate close to their maximum transmission capacity.000 MW can be generated. which will bring a gain of 600 MW of guaranteed energy by making optimal use of the hydrological diversity between the river basins involved.1. however. it’s estimated that as much as 10. Finland has mainly thermal generation and Sweden has an even mix of thermal and hydro generation. • Another important factor is the co-generator with Petrobras as the biggest. interconnecting the north/northeast system to the larger south/southeast/centerwest system. The scenario calls for the solution of an important structural problem: How to stimulate private agents to build hydro plants? There is still a considerable hydro potential to be explored in Brazil that is economically feasible. The Danish system is unique with a high penetration of wind energy and co-generation from independent producers (see next section).000 km long. and to describe possible impacts with respect to stability control. together with the gas pipelines Bolivia–Brazil and Argentina–Brazil. The Nordel system has undergone major changes during the last decade due to restructuring. The other part of the Denmark. The Nordic power systems are characterized by a mix of hydro and thermal generation. both first costs and operating costs for controls and how to allocate them among the various parties. so as to ensure that investments that are sound for the whole system will actually be made. The problem in the new competitive generation framework is how to establish costs. an intensive capital investment. All of the above factors point to continued importance of the phenomena of system stability and increasing dependence on control actions. The aim of this section is to give a brief overview of the changes. As described in Chapter 7.2 The Nordel power system The Nordel power system comprises the interconnected power systems of Norway. which is interconnected with the UCPTE system. 8-4 . 8. Restructuring in the Nordic power systems. 8-1 shows the total annual generation in the Nordic countries. in general. The restructuring started in 1991 with deregulation of the Norwegian electricity market. and Finland joined the common Nordic market in 1998. Restructuring. and the transmission capacities between the countries and to the European continent. 8-5 . deals with the following issues: • Unbundling of services. Sweden followed in 1996. Generation and transmission capacities in the Nordic Countries.50 MW Russia 100 MW 740 MW 900 MW NORDEL 700 MW 120 100 80 60 20 200 MW TWh 80 TWh 80 60 40 20 500 MW 1100 MW 60 40 20 0 TWh 20 0 0 Finland Norway 500 MW Sweden 1600 MW Hydro Nuclear Fossil 1050 MW 1800 MW (future) 60 50 40 30 20 10 0 630 MW TWh 670 MW Denmark 1400 MW 600 MW 1000 MW 600 MW 600 MW (future) Germany Poland Holland UCPTE Fig. • Deregulation within trade of electrical energy. 8-1. Fig. These companies are also the main transmission grid owners in their respective countries. distributors. The main grid company. The Regulator grants regional concessions and concessions for trade in electrical energy and has an important role in supervision of the monopoly operations in transmission and distribution. Network Owners. See Nord Pool’s web site [8-17]. and traders/brokers. The Network Owners have by regulation been given the responsibility for generating and distributing metering and settlement data. and include generators. Market Participants. The Regulator of the Norwegian power industry is the governmental body “The Norwegian Water Resources and Energy Directorate” (NVE). • Create incentives for cost reduction. and keeping continuous track of the information so that equal opportunities are given to all the competitors. Changes in system operation from restructuring. These relate to changes in 8-6 . • Achieve reasonable geographical variations. Statnett. Svenska Kraftnät and Fingrid. provides further information. power markets and retail sales. totally independent of which network owner (distribution grid) they are connected to. Statnett SF. The Market Participants are buyers and sellers in the market. issued by SINTEF Energy Research. Economically and functionally separate units are established within power generation. industry. Nord Pool and the Norwegian Electric Federation. transmission and distribution. The operator of the common Norwegian/Swedish/ Finish market is Nord Pool. There are some major changes from system restructuring that affect system operation and control.• Dis-aggregation of utilities. • Create equity among consumers. Retail Sales. Implementation and Experiences 1991–1997 [8-18]. The report Deregulation of the Nordic Power Market. has the system operator responsibility in Norway. In the Norwegian case. the major arguments for restructuring of the electricity market have been to: • Avoid excessive investment. respectively [8-17]. System Operator. Nord Pool is also open to market participants without physical access to the Nordel grid. The main actors in the Norwegian (and Nordic) power market are: Regulator. Market Operator. but is only indirectly related to the power exchange. The Market Operator is responsible for the market clearing process in what is called the organised markets. • Improve selection of investment projects. Retail sales are yet another service made possible through deregulation. Similarly. Retail sales mean that the individual electricity consumers are free to choose from which power company they buy their energy. there are independent system operators in Sweden and Finland. Monitoring and controlling system stability. Larger and more frequent changes in power flow patterns will increase the need for coordinated and more robust control solutions. system reserves.. as contracted services (bilateral contracts between the system operator and a power producer) or as market-based services. Ancillary services are fundamental services needed in order to maintain acceptable power quality and power system security. such as stability control. etc. as well as to new services and ways to operate the system. Changes in responsibilities and ownership. Increasing focus on cost efficiency. but organized in the Scandinavian countries through separate markets. responsibilities and ownership.objectives. System responsibilities. provision of active and reactive reserves and system protection (load shedding or generator tripping) schemes. is to a large extent based on ancillary services. Deregulation of energy markets and increasing competition among the power producers lead to larger and more frequent changes in power flow patterns.. Secondary controls for congestion management or power balancing may also be defined as ancillary services. and thus the existing systems will be operated closer to their capacity limits. New services are introduced to deal with the changes discussed above. which are mainly economically motivated. There is a mix of power producers. and reactive reserves/voltage control become secondary objectives. Thus the need for stability controls will also increase. including stabilizing control. Increased utilization means less reserves and more transmission congestion. possibly with fixed economic compensation. A result is that fewer new lines are being built. transfer limits. Well-functioning system (or ancillary) services are crucial in the restructured environment. Their main control objectives relate to control and optimization of their own energy resources and market obligations.g. The system operator will normally contract or require the individual power producers to provide some system services. contribution to active reserves/frequency control. This has to do with the unbundling of services that defines the responsibilities and tasks of the different entities. Impact on system stability and control. which are the main responsibilities of the system operator. The changes from restructuring may impact power system stability and system controls. Changes in operating patterns. New controls rather than new transmission lines will increasingly solve transmission congestion. In order to become less dependent on ancillary services provided by 8-7 . This relates to both operation and to changing attitudes toward investments in new generation and transmission capacity. The services may range from primary frequency and voltage control. Experience indicates that deregulation has caused decreasing investments in new transmission and generation capacity. Ancillary services can be organized as firm requirements (e. and the systems are operated closer to their capacity limits. primary controls). there is a need for improved EMS tools at the system -control centers. The probability that all the wind production will change rapidly in an equal way is smaller the larger the system is. For this type of power variations. Both the desire for larger power transfers and the increased uncertainty of the power changes create a need for advanced angle stability control. wind power and hydro power combines very well.1. and a free exchange of power between hydro areas and wind areas is desirable even when located far from each other. Power system security and power quality may be regarded as collective benefits. In Denmark 800 MW of wind power was installed in 1997. out of a total capacity of 10. 8-8 . On-line tools for voltage stability and transient stability assessments will become increasingly important. Another part of the power variations will come from uncontrolled power plants such as wind generation. Having one such independent entity may also prove advantageous regarding the technical possibilities of providing coordinated controls.000 MW. it is likely that system operators will show increasing interest in deploying power electronic devices for congestion management (power flow control) and stability control. Besides.3 The Danish electric system Large variations in power transfers in transmission networks of large interconnected systems must be expected in the future. In summary. Development and application of new energy storage devices for fast-acting reserves may also become more attractive in the future. By allowing the power variations to spread freely over the entire system less demands will be put on the control of the controllable power plants in order to maintain a local power balance.generators. In deregulated systems the system operator is given the overall responsibility for maintaining the security and quality criteria. In order to monitor and coordinate the increasing control applications. the system can at least have a short time warning. Another way of handling transmission congestion is to rely more heavily on special protection schemes. It will be most easy to include a large amount of wind power in a system if the natural changes in the power production is allowed to spread freely over a large area. and the system operation can be adjusted to be able to handle the power transfer. This dependence on both existing and partly new control devices will require sophisticated design as well as improved tools for on-line system operation. the major impact on the technical side from system restructuring is an increasing dependence on controls in order to cope with the increasing competition among power producers and the increasing utilization of existing transmission grids. Thus there is a need for robust and coordinated design in order to avoid adverse interaction between protection systems and other controls. One part of the power variations will come from controlled power plants delivering power to remote consumers or power companies on short-term conditions. 8. Members of interconnected systems owning both transmission and generation follow voluntarily the guidelines set by coordinating councils (e.1 Assuring compatibility of equivalent dynamic characteristics In the traditional vertically-integrated power company. in the WSCC every unit over 75 MVA is to be equipped with a PSS). organizations such as NERC. NPCC. Where necessary. In this scenario the cooperative approach to accepting one’s share of investment. ERCOT etc. Examples are in the area of primary frequency control (droop settings and spinning reserve) and automatic generation control (area control error reversals per hour etc. It’s not merely stability that must be addressed.).8. could lead to very high overvoltages and widespread damage to system and consumer equipment. There would be no tendency to under invest in one segment (generation. the overall system reliability is the responsibility of one entity. in the USA. Reference 8-4 describes how the evolution of system structures can affect the necessity for stabilization and its location.. We raise issues for discussion without advocating particular administrative and financial approaches. For interconnected systems using long distance transmission. This dilemma extends to other system reliability aspects such as transient stability. 8. This solution is usually borne by the generation segment. Since this problem of damping can also be abated by adding transmission. and generator dropping. In this environment the acceptable characteristics of generation. where independent producers have no perceived stake in transmission.2 Coordinated Planning and Operation in a Competitive Environment Organizational and administrative issues under the new competitive environment can only be resolved successfully following recognition of technical factors that make interconnected operation possible. Loading. with its effect on angle separation and relative inertia between sending and receiving areas. control and protection evolved naturally to fairly uniform patterns among various utilities. The techno-economic solution is to distribute control effort (PSS in this context) over most generators. dictated by dynamic considerations. Load rejection and system separation.g. load shedding. with overspeeding generators connected to excessive line charging. Distributed generation in systems linked by EHV and UHV transmission can present major challenges in system and protection design.2. or controls) causing a disproportionate impact on reliability. was natural for mutually beneficial interconnected operation. UNIPED in Europe and GCOI/GCPS in Brazil issued recommendations on practices to be followed by all members of such power pools. transmission. the problem of poor damping of inter-machine and inter-area electromechanical oscillations presents a serious reliability problem. play an important role. whether state or investor owned. as dictated by technoeconomic considerations. one can appreciate the problem of enforcement of the most techno-economic solution. transmission. The entire system design must be by a highly trained team considering all relevant parts of the system regardless of ownership. WSCC. 8-9 . • Improvements in voltage stability.2 Detrimental aspects IPPs. • In Brazil.3. 8. • The better dynamic voltage control will yield a more reliable operation of transmission line distance protection.3. with a significant reduction in undesired tripping. • Improvements in electromechanical stability. but also a function of the PSS and other excitation equipment control tuning. The damping of interarea oscillations will also tend to improve as a function of the smaller phase angle differences. frequency decay. The effectiveness of PSS in providing damping is not only a function of their application on generating units. System-wide dependence on PSS for adequate damping performance will require more formal inspections and testing by the regional transmission organization (ISO or independent transmission company) of the restructured power industry. 8. Their motto is to maximize power production and reduce their costs.1 Beneficial aspects of IPPs New thermal-based IPPs will bring many benefits to the interconnected system: • Being close to the load centers they will bring better voltage control and smaller loading of transmission lines. • Extra flexibility in planning equipment and transmission line outages.In systems with widespread transmission and significant interchange over long distance. What could be the consequences? Some of the functions carried out for free in today’s environment (voltage support.3 The Impact of IPP Thermal Generation on System Dynamic Performance 8. due to smaller line loading and added dynamic voltage support. units that normally are not participating in oscillation damping action can become important. frequency 8-10 . because of a larger reactive power support near the load centers. with consequent reduction in transmission system losses. the availability of more thermal generation will allow an effective hydrothermal coordination. • Alleviation of the problem of ever-increasing transmission distances to bring power to the load centers. when compared with state owned or regulated generating companies. the problem of oscillatory instability can dictate the need for stabilizing action under normal operating conditions. Since the nature of the system structure following multiple contingencies is almost unpredictable. In other systems the problem arises only following contingencies. This decreases the risk of system separation. and load shedding. are oriented towards a higher and faster return on investment. 4. primary speed control. The challenge is to develop an organizational structure to execute the necessary system studies and enforce the design requirements among the separate parties.1 is the need for assuring redundancy in the case of unplanned outages of such facilities. which would have to be borne by all consumers. transient overload capability. Offsetting some of the positive aspects of IPPs listed in §8. Little has been done so far to develop such methodology. 8-11 . Some of these aspects are further elaborated in tables in section 8. These issues generally involve dynamic aspects of the plant interacting with the power system. In the restructured environment the technical approach should be the same since the mere fact of separate ownership of generation versus transmission does not change the underlying laws of physics which govern the reliability of overall system performance. transmission. As competitors. dynamic response.3. whether they involve generation. which should include allocation of costs to those agents not contributing their share of ancillary services.3 Problem issues with new IPPs Tables 8-1 and 8-2 list system design and operation considerations for the restructured industry. transmission.regulation. IPP additions may impact the adequacy of transmission networks. etc. Reliability criteria are followed and the design process considers all logical cost-effective alternatives. control or protection.” whose provision will have an associated cost. IPPs are concerned with the generation process and normally would not have the expertise to determine complex control and protection requirements dictated by the overall system. The resulting additional reinforcements needed in transmission would be reflected in transmission costs. the planning and design process is usually undertaken by owner representatives participating in joint interconnected system studies with access to the entire database. the parties have a natural tendency to hold back on free exchange of information. but also of licensing future system additions in generation. 8. These include providing ancillary services— for instance reactive power support.3. protection or control. Such outages result in loss of both power production and voltage support—which must be provided by alternate facilities.) are classified in the new environment as “ancillary service. Issues listed in the tables show the need to establish methods and procedures for requiring certain design features in IPP installations. supplementary damping control (PSS) etc. In the vertically-integrated traditional utility (or power pool made up of such utilities). If this is not done in the planning process. The foregoing considerations point to the logic of a strong independent and competent organization to not only be in charge of system operation.3. Short-time overload capability of generator excitation equipment Exciters are able to produce up to 200% of rated reactive power for approximately 20 seconds.Concerned only with plant equipment security. 8-12 . to maintain short-circuit level and to avoid transmission line opening to mitigate sustained over voltage during light load. System Voltage/Frequency Control. plant tripping can aggravate the voltage profile. Participation in Special Protection Schemes (SPS) The design and implementation of Special Protection Schemes (Emergency Control Schemes) are analyzed by all parties involved. .For disturbances that cause large absorption of reactive power by the generators controlling voltage profile. the tendency would be to order less costly. . In extreme cases it could lead to a system collapse. being considered the most appropriate and economic means.Table 8-1. i. Operation of generators as synchronous condensers This characteristic is used during light load conditions in order to provide better voltage profile control.For disturbances that cause generation deficits. Generator power factor Generators with low rated power factor (0. With increasing numbers of IPP plants. This can lead to voltage control problems and even collapse. This expedient would require installation of clutches representing additional costs. . The deficiency of reactive power reserve could require move expensive alternative equipment in transmission. This non-acceptance may jeopardize system reliability and require system reinforcements. The SPS is installed considering the best location. higher rated power factor machines. This can avoid network reinforcements.Lower capacity excitation systems and conservative setting of limiters. this can lead to larger system overvoltage and equipment tripping (or damage). power system stabilizers and governors Are fully utilized to improve the power system dynamic performance. This improves system dynamic performance. This reduces IPP costs.Larger possibility of plant tripping during disturbances. settings of minimum excitation limits can cause plant tripping. In this case it will be necessary to increase the total load shedding. Excitation equipment. The limiter actuation might be increased to protect excitation and generator windings against failure due to high voltage stress. . Protection. and Stability Aspects Issue Protection settings Traditional Approaches Problem issues with new IPPs Settings take into account the plant equipment’s and power system requirements . increasing the chances of reaching transmission lines protection settings.System requirements could be enhanced with greater MVAr reserves in generators.9) can be used.e. it can be installed in any plant.For disturbances that cause overvoltage. plant tripping will increase the magnitude of frequency dips. The tripping of transmission lines could lead to a system collapse . Application of higher cost machines with improved excitation systems (high initial response yielding move effective action from PSS) would not be normally adopted without some hard rules to define compensation of costs.. Without consideration of system requirements during contingencies. The IPP may not accept to participate in any SPS. . IPP could consider that they have no obligation to do that. very little has been done concerning control action cost allocation. -Every subsystem has at least one power station with black-start capability. Generating unit operation with minor failures When minor failures occur. So far. may rely on remote power station cranking rather than install black-start capability.IPPs. the overall system reliability could be somewhat degraded in the future. can have detrimental impacts to system dynamic performance. Black-start capability The operational planning of the interconnected system determines the restoration planning with its various parallel subsystems. and control and protection.To provide a more reliable and secure operation.3. however. . Information exchange and data availability . Although transmission network investments will rise. and maintain the number of machines in operation in order to obtain the maximum productivity of the plant. The proliferation of IPPs. 8-13 .. including special requirements in equipment. The eventual lack of control actions and the consequent rise in transmission prices can result in loss of economic efficiency in power production. abundant information is made available on current limitations/ unavailability of equipment and power flow constraints. . . the utilities may agree to keep the generating units in operation until system conditions evolve to level at which unit disconnection will not jeopardize the overall system reliability. etc. for cost reduction. which may not be fully compensated in the transmission network at reasonable cost. with their rapid installation cycle. Careful planning and design studies should establish the proper integration of IPPs into the interconnected system.Table 8-2. 8.This intentional withholding of information is detrimental to overall system reliability.The IPP could have inadequate data acquisition and recording equipment. Operating Aspects Issue Minimum number of units in operation Traditional Approaches Problem issues with new IPPs The number of units in each plant is determined to guarantee a minimum value of system inertia and reserve. the main objective being to protect their own equipment. plant operator reports.4 Conclusions related to IPPs 1. the system restoration time may be increased. . The IPP could consider that they have no obligation with system reliability requirements.The IPP could consider having no obligation to inform the others on what is occurring to his plant.) for post operation analysis. 4. . 2. .This could affect voltage control. Many efforts have been noted to develop methods and tools for some of the ancillary services.Traditionally all the data are available including data from disturbances (oscillograms.As a consequence. 3. system stability and increased frequency dips. . This should be considered a priority issue in order to ensure economic efficiency. . 5. The transition period from the cooperative model to the competitive one will cause some additional risks, which are not fully assessed. 6. One way to minimize the detrimental impacts and additional risks, is ensure that selling ancillary services can be good business. Something should done, so as to make generation fulfill its natural or traditional ancillary functions. Finding other control alternatives in the transmission network (like FACTS) is always more expensive. 7. The Independent System Operator (ISO) concept is good, but that organization should have added functions in long term operational planning and, particularly, licensing and inspection of new facilities to ensure that they meet system requirements. 8.4 Other Issues Related to Power System Performance in the New Utility Environment 8.4.1 Reliability aspects The forces of market deregulation have encouraged a widespread decline in planning resources, and have undercut the planning process itself. Unrealistic models provide a common point of failure for the entire decision making process whereby the power system is planned and operated. Compounding this, the system sometimes operates under conditions that planning cannot anticipate. Market deregulation and utility restructuring are, through a variety of mechanisms, making it impossible to predict system vulnerabilities as accurately or as promptly as the increasingly volatile market demands. Controller-based options for reinforcing the power system can be very attractive. For a control system to be fully competitive in this respect, however, its functional reliability must somehow be established early in the planning process. It’s rarely possible to do this within the conventional framework used for new transmission lines or for new power plants. It’ll always be necessary to trade the benefits promised by a control system against the inevitable risks associated with closing a highpower loop around system dynamics that are not fully understood. If the risks are perceived as too high, or if the functional reliability is perceived as inadequate, then system reinforcements though enhanced control will be displaced by less technically demanding means. Reliability is just one intangible emerging in the new power system. Others include information security, regulatory changes, business survival, and the directions in which a particular regional transmission organization (RTO) evolves. 8.4.2 Implications of equipment ownership As many electric power systems move toward deregulation, there is much focus on the economic issues associated with the new competitive operating environment; details of energy trading and pricing have been in the forefront. However, the ability to operate in such an environment with an acceptable degree of security and reliability, and indeed to 8-14 be economically competitive, requires significant attention to the methods and strategies of power system control. In the new environment, the power system comprises corporate entities having diverse roles, equipment, and business interests. There are independent generating entities, transmission entities, distribution entities and brokering entities. The physical functioning of the integrated power system, however, remains the same as before. Therefore, the responsibility for control of individual equipment should not follow ownership; instead it should be vested with RTO. The specification and design of these controls should be part of overall system planning and design carried out by an independent entity. Otherwise, system security and economy will be sacrificed, defeating the very purpose of restructuring the industry. In particular, it’s essential to recognize the critical role played by generator controls in maintaining system stability and controlling voltages and frequency. It should be mandatory for the generators to be fitted with fast-acting excitation system, AVR, and speed governing systems. In many cases, PSS should be mandatory. The PSS should be designed and tuned so as to contribute to the enhancement of overall system stability, including damping of local as well as interarea modes of oscillation. There should be no difficulty in motivating power plant owners to install controls that enhance the operability and stability of the generators. For those controls that are provided to meet the overall power system requirements there should be proper financial incentives. 8.4.3 AGC in the new environment With deregulation comes the redefinition of system control areas. Both the introduction of new control areas and the consolidation of existing controls areas impact the way traditional control issues are handled. Traditionally, frequency control, achieved through the matching of generation to load, has been one of the functions of control areas using some form of automatic generation control (AGC). Although the extent to which frequency control is required is debatable, some control is required to prevent the instabilities and other adverse effects associated with excessively low or high frequencies. The control of frequency to tight tolerances is arguably associated with improved power quality which may be expected by some customers, but that is not strictly a requirement for successful interconnected operation. In a deregulated environment it’s not clear who will be responsible for any level of frequency control. AGC requires spinning reserve that can be valued as an ancillary service. While it’s possible that certain parties will be prepared to provide such services for a price, what is not clear is the extent to which this will occur. The first issue of maintaining the frequency of a large system within limits required for secure operation is a natural byproduct of near matching of the load and generation which should take place under free energy trading. The second issue of maintaining tight frequency control for power quality concerns should be based on value and price to consumers. 8-15 8.4.4 Modeling/data requirements — a bigger challenge Equally important as the analysis method is the model used to represent the power system. It’s essential the model represent sufficient detail and accuracy to properly reproduce all important system dynamics. While this has led to the use of very large system models (for example, North American Eastern Interconnection is often represented by more than 26,000 buses), analytical tools are available to handle such systems. Good dynamic reduction methods are also available which can be applied to reduce large models to more manageable sizes while retaining the key system dynamics. Perhaps a bigger challenge is the availability of model data for various equipment, including generators and the associated controls, protective systems, and system loads. While phenomenal advancements have been made in terms of analytical techniques and computational tools, data acquisition has not kept pace with the requirements. Many utilities use “typical data” for modeling much of the equipment. For control and protection, the data is often not representative of the actual settings and, in many cases, the condition of the equipment. More effort is needed towards the acquisition and verification of model data. This is being increasingly recognized by the industry, particularly in the aftermath of major system disturbances. For example the two disturbances that occurred in 1996 on the western North American system have motivated the WSCC to mandate field measurement and model derivation for all generation units (unit, exciter, PSS, governor, and protection) greater than 10 MVA. Once good models are obtained (that is, they match the field response), then it’s necessary to use this information to optimally tune the system. Once optimized, it’s essential that field adjustments are not permitted without prior study of the impacts. 8.4.5 On-line dynamic security assessment and real-time monitoring and control In the new power sector the system conditions are extremely unpredictable and the volume of transactions that may have to be examined may be huge. The traditional approach to deploying preventive and emergency controls based on off-line security analysis studies which generate a set of tables indicating stability limits and control measures may not be satisfactory. In the new structure, tools are necessary, such as on-line transient stability assessment and voltage stability assessment. These are described in Chapter 5. In order to make these new tools useful, it’s necessary obtain reliable on-line input data. It’s necessary adopt real-time system monitoring and control. An example of such a scheme is a wide-area measurement system (WAMS) being developed by Bonneville Power Administration western North American. The WAMS use synchronized phasor measurements and portable power system monitors to centralize information at control centers. 8-16 For example. such as IEEE models. Depending upon the type of SPS. varying degrees of functional redundancy may be required to ensure reliable operation. All Type I and Type II DCS (installed for the purpose of mitigating the interarea impact of extreme contingency) are designed so that a critical 8-17 . including protection criteria. This means that vital subsystems should either have a functional redundancy or sufficient selfdiagnostics so that there would be reduced dependency on the DCS in setting transmission system limits. In all cases of SPS. However. for loss of an element without a fault or due to a single line to ground fault cleared in normal time. It’s necessary that any changes to the control parameters be communicated to the ISO. the design and operation must be consistent with all criteria. In addition. In order to ensure the proper modeling of excitation equipment (also other machine and governor parameters). Type I (SPS with potential for interarea impact. the failure of an SPS circuit breaker is considered as part of the normal criteria. Thus there are situations where excess generation may be armed for rejection to ensure that sufficient generation is successfully tripped for a critical fault.3].6 Alternatives for pricing of stability controls in a deregulated industry The shift to a market-based structure necessitates the unbundling of services by stripping out non-energy costs and identifying ancillary services that have costs and value.4. The following is based on procedures of the Northeast Power Coordinating Council in North America [8. With the re-regulation of the electric power industry one important question appears: Will proper market signals in combination with commonly accepted “best practices” foster competition and preserve or even enhance the reliability of the system? This represents a difficult challenge with respect to the interface between the market driven generation sector and the regulated transmission system that may be under the control of an Independent System Operator (ISO).8. The design and operation of the DCS must be approved by the ISO. the ISO could conduct audits (similar to machine parameter measurement R&D projects) as required. DCS are subject to reliability standards that ensure dependability and security. Performance requirements. In addition. the prudent use of Special Protection Systems (SPS) and Dynamic Control Systems (DCS) can play a vital role in enhancing both competition and system reliability provided that proper market signals are implemented. For Type I DCS (whose incorrect operation or failure to operate following a normal criteria contingency would have interarea or interregional consequences). design requirements specify that the DCS should perform its intended function for specified Bulk Power System (BPS) contingencies while itself experiencing a single undetected failure. procedures must ensure that the DCS performs as intended. event reconstruction by simulating actual system events and comparing the results with the actual machine performance could identify units with suspect parameters. initiated by normal conditions) may require two independent protection schemes while a Type III (SPS with potential for local impact only) may require only one set of system protection. Modeling should be consistent with industry standards. Note that the NERC Standards require the generator owners to provide accurate and timely steady state and dynamic data for their generating units [8-8]. Once approved. Payment Schedules for SPS and DCS. For all DCS and those SPS required for stability.) If the system is operating in an insecure state (for either normal or extreme contingency criteria) and the arming of an SPS or DCS would return the system to a more secure state. but will no doubt be complicated by the regulatory process required to approve this methodology. In the future it may be possible to attribute an improved loss of load probability to particular SPS or DCS. The implementation of a SPS or DCS is dependent upon the system conditions that justify their use. owners of DCSs have obligations to perform both maintenance and monitoring functions. a Type I SPS could restore the transfer limit. but there is no special payment made for their use. Future system additions could be addressed through rules such as a requirement that all future generators are required to have high performance excitation systems that include power system stabilizers (PSS). It’s not clear if the Don’t Pay option will cause any degradation in the reliability of the system with respect to the implementation of SPS and DCS. Several options exist for the payment schedules for the arming of existing SPS and DCS. the SPS or DCS becomes essential to maintaining reliability. as well as for the implementation of future devices. but rather defines reliability as the requirement that the implementation of a DCS or SPS cannot reduce the operating limits of the network. the devices are in effect providing greater stability margin to the system for a particular set of contingencies. In this case.failure of the DCS itself does not cause unacceptable BPS behavior. Reliability. In the Don’t Pay scenario. then there are economic penalties. the interface flows may violate the permissible normal criteria transfer limit. For this scenario. The economic use of an SPS or DCS applies when the device is required to increase the normal transfer limit of the system. For example. This could then be weighed against the value that the load places on enhanced reliability of service. In the short term the primary 8-18 . Similar to protection system criteria. An alternate approach to reliability does not account for the additional robustness of the system. it’s assumed that all existing SPS and DCS continue to function in a secure manner. Some options: Don’t Pay. We discuss two main categories of SPS use: reliability and economy. The Don’t Pay method could also require payment from the SPS or DCS providers if they failed to preserve existing system transfer limits. Economy. the use an existing SPS or DCS as well as the planning of a future SPS or DCS would be driven by the transmission tariff structure. Justification for SPS or DCS. (If there is a reduction. It is judged that this will present not only technical challenges. An approach to defining reliability is to recognize that the SPS or DCS is providing greater resiliency to the operation of the network when the device is not required in the setting of normal limits on the system. immediately upon the loss of one or more transmission facilities. A proposal for the payment of the DCS is dependent upon several factors. They are considered as part of the “Voltage Support and Control” Ancillary Service and the payment for this category of DCS is thus highly dependent upon the transmission rate structure. transmission facility. Local arrangements could be made to compensate the involved parties. Therefore.) If this power is from the economy (perhaps the Location Based Marginal Pricing or LBMP) market. Eventually there may be reliability-based rates and the load which is placed at risk might get a discount. the robustness of the network could improve if price signals locate new generation closer to the load and overall system transfers are reduced. the generator. It’s also well recognized by market participants that the great experiment in the deregulation of the electric power industry could come to an abrupt end if there were many interruptions of load. and the load) are the beneficiaries of the reliability aspects of the SPS. SPS is in the same class as underfrequency load shedding. In the long term. then the differences between the economy market and the rejected generator’s price is provided from the TCT (Transmission Cross-Tripping). Embedded Cost. In the event that either Type I or Type II SPS is triggered and works as designed for an actual contingency or has an undesired trip (within reason). For Type III SPS (with potential for local impact only)—the local area is the beneficiary of the SPS. TCT (Transmission Cross-Tripping)—No payment. payment shall be as follows: GR (generation rejection or reduction)—The unit must be made “whole” otherwise the unit would not be willing to provide the extra measure of security. (It’s assumed that only a limited number of false trips due to the SPS would be tolerated. Excitation equipment tuning and supplementary controls. This scenario would of course reduce dependency on SPS and DCS. are essential to the stability performance of the system. back-up power is supplied free of charge to the generator if it is rejected. including whether the control is excitation equipment or governor related. For Type II SPS the system has an extra degree of security against extreme contingencies. or load providing this service is placed at risk. For Type I SPS the ISO and all market participants (the generator. transmission owner. The payment for arming a SPS would include paying for the installation and maintenance of the protection system as part of the Transmission Service Charge (TSC) or Transmission Uplift Charge (TUC). such as power system stabilizers. and the inherent transmission rate structure. We discuss the possible use of an Embedded Cost method. LR (load rejection)—No payment. 8-19 . the generator is not compensated for any additional transmission capability that may be available as the result of arming or installing the SPS. first for SPS and then DCS. The ISO has responsibility for system reliability and the TCT provides an extra level of security. but without necessarily accounting for lost opportunity costs. the type.focus of generator providers will be on issues that are more economically lucrative. This method recognizes the benefits of SPS and DCS and seeks to make the provider cost neutral. However. However. as would be the case for the rejection of generating units. It’s interesting that generators on the downside of. In other instances. Turbine governor DCS fall into several categories. For this scenario. However. the SPS comes at a cost to the owner of SPS. such as replacement with solid state systems and/or the addition of PSS could result in greater system resiliency. This is because the higher limits would result in the generator receiving a lower LBMP and the load paying a higher LBMP price. Alternatively. The following alternative bid based system could be used: GR—Since generation rejection schemes allow for higher economy transfers by placing units at risk. impact the short-term dynamics and can be handled similar to a SPS. the benefiting entities should pay the machines for accepting a possibly 8-20 . the question of how the ISO would arrange for system improvements justified by improved system resiliency needs to be determined based upon the particular ISO definition or reliability. It would be a difficult task to economically quantify the differences between the control performance (AGC) response and the DCS turbine governor response. The proper price signals would establish an economical incentive for providing SPS and DCS services. it may be possible for the SPS holder to utilize a more direct bid based methodology that avoids the complication of a special TCC auction as suggested by some LBMP systems. Those that provide frequency response and regulation services usually impact the long-term dynamics of the network and are commonly addressed by transmission tariffs. particularly with respect to extreme contingencies. The transmission allocation could be defined as the increase in transfer capability across a congested interface. We now discuss possible Market-Based Rate methodologies for SPS and DCS. In either case it’s suggested that the embedded cost method would pay the generator for all or part of the excitation equipment modification. The generator could proceed with the improvement by making financial arrangements with other market participants. the generator is paid for a portion or the full capital and operating cost of the DCS. a congested transmission interface might be reluctant to make system improvements resulting in higher transmission operating limits. For a Type I SPS resulting in higher system transfer levels. However. such as fast valving.Transmission tariffs for Voltage Support and Control are often “embedded” or cost based rates. Market-Based Rates. and loads on the upside of. In this case. It’s recognized that the generator would not realize any additional benefits from increases in transfer limits. An interesting approach to determining the value of and location of the transmission allocation could be the auction of incrementally feasible Transmission Congestion Contracts (TCCs) as suggested in some Locational Based Marginal Pricing (LBMP) methods. Other DCS. possibly including a lost opportunity cost. Excitation equipment improvements. The auction could be conducted similar to proposals for the conversion of “traditional” transmission rights into TCCs. the Transmission Provider could offer a payment based upon a percentage of the expected increases in wheeling revenue. the modification to the excitation system may not be capital intensive and could require simply changing a gain. The SPS holder could theoretically claim an allocation of transmission that could be handled in any number of different ways. there is a cost saving to customers. LR—Load rejection is the dual of generation rejection and could be handled a similar way as follows: The ISO determines transfer limits with and without the SPS activated. Type III SPS is a local issue where it is difficult to generalize reliability versus economy method of compensation. increased transmission capability could be allocated to the owner of the DCS. it’s conceivable that loads may wish to pay for higher levels of reliability. In the future. such as possible loss of life of the machine from additional trips. Here the individual owners of the DCS could come to some business solution. This method is applicable to new or improved DCS as well as for DCS that can be armed or disarmed by operators or a defined set of system conditions. the allocation of the payment would be more complicated and possibly require advanced analyses that determine the individual contributions of the DCS 8-21 . Similar to the methods described in the SPS section. It’s suggested that the following procedure be invoked: The ISO determines transfer limits with and without the SPS activated. a decision could be made on whether or not to pay the reliability-based rate for the SPS. The generator would need to weigh these costs against possible lost opportunity costs due to the lower transmission interface capability that would result from disarming the SPS. The Load (possibly through a power exchange) accepts or rejects bids from other generating units or load serving entities for the activation of the SPS. At that time a power exchange could be used as a mechanism for bidding for the activation of the SPS. The transmission allocation problem becomes more complex for the case where multiple DCS are coordinated to increase transfer limits. For the case where the transmission system is stability limited. However. TCT—Presumably higher system transfer limits would result in greater transmission revenues. Based upon this economic determination. Other costs to the machine include possible penalties for backup supply (assuming a bilateral contract) and physical costs. The generation unit (through a power exchange) accepts or rejects bids from other generating units or load serving entities for the activation of the SPS. Therefore. the owner of the TCT is reimbursed for the costs of the SPS through the Transmission Uplift Charge. possible based upon techniques that are used for tuning DCS and Dynamic Security Analysis. the application of a single DCS could increase the transfer capability of the system. This could be weighed against the cost of the service interruption. Type I SPS do not increase transfer limits and Type II SPS are reliability based and do not have an economy market at this time. It’s envisioned that alternative approach would be an ISO calculation of the probability of the contingency events necessitating the SPS action. Total payment for DCS used to enhance the reliability of the network would be determined similar to the method used for SPS.lower capacity factor. J. 2. Recife. 8-22 . Donnelly. Cavalcanti. No. D. Trudnowski. Brazil.” IEEE Transactions on Power Systems. marketed as an ancillary service. Vol. Clark. Vol. pp. “Brazilian North–South Interconnection — Application of Thyristor Controlled Series Compensation (TCSC) to Damp InterArea Oscillation Mode.Market Power. M. J. 87. • Large-scale stability control (LSSC) faces many technical challenges that make it very difficult to assure reliable LSSC performance. L. May 1998. This would favor a shift in emphasis. 8-5 M. Leoni. Ricardo. Fink. 8-3 M. E.7 Large scale stability controls and legal liabilities Legal liability with stability controls may be a concern when the nature and role of the RTO (Independent System Operator or Independent Transmission Company) is not well established [8-14–16]. Rogers “Impacts of the Distributed Utility on Transmission System Stability. An RTO should be held harmless for duties performed according to sound engineering practice. January 1997. Aug 21–27. could be a magnet for lawsuits. 8. W. “Some Aspects of Transmission System Planning and Design in Developing Countries. Vojdani. New Hampshire. Henderson. 1998. “Stability Controls in a Restructured Industry.” IEEE Transactions on Power Systems. References and bibliography 8-1 M. Gribel. Adibi and L.11. H. R.” V SEPOPE. 8-2 H. toward greater acceptance of system failures but with LSSC action to make the failures “graceful” and to facilitate prompt restoration of electrical services. Fraga. J.J. 8-7 D. 1. “Power System Restoration Planning. Key points are: • Redefinition of electricity as a market product may well expose all providers to legal liabilities from which they are now immune. PTI Newsletter. A. K.” CIGRÉ. 8-4 F.4. No. 22–28. It will be necessary to prevent anti-competitive actions by generators and loads by constant observation and possible dispute resolution by regulatory authorities. P. Tenório. and R. Only the RTO(s) will have the infrastructure and other assets needed to monitor LSSC performance effectively.9. “An Overview of Ancillary Services. J. Dagle. Issue No. May 1996. M. In all cases where the owner of an SPS or DCS is paid there is the issue of market power. Gama. G. 741–746. LSSC. R. paper 14-101. Eiras. Shirmohammadi and A. Fourth Quarter 1996. de Mello. J. Henniker.” presentation to IEEE/PES Power System Stability Controls Subcommittee. 8-6 C. 1976. February 1994.” Engineering Foundation Conference. Ping. pp. • LSSC actions that are initiated after system failure is clearly underway face less legal exposure. 8-12 L. “Application of Power System Stabilizers for Enhancement of Overall System Stability. May 1989. 1999 (to appear in IEEE PES Review). 8-9 P. Klein.4. “Control Area Trends: Principles and Responses.svk. 8-15 A.” IEEE Transactions on Power Systems. 8-11 L.no/. August 24–28. Fink. H. von Son.” IEEE Transactions on Power Systems. 8-14 J.fingrid. Fleishman.” 1997 IBC Conference on Ensuring Electric Power Reliability in the Competitive Marketplace. Philadelphia. S. J.htm). Hauer and C. http://www. J.se/ and http://www. and M. Canada. Reliability.fi/. Rogers. Zymno. Taylor. No. 8-13 P. K. “Power System Control: Requirements and Trends in the New Utility Environment. J. M. San Francisco.nordpool. 8-16 B. NERC Planning Standards. Roman. Kundur and G. 1997. pp. J. and Control in the New Power System. F. M. SINTEF Energy Research.statnett. 34–39. 8-17 Internet addresses: http://www. “On System Control within a Restructured Industry.sintef. May 1998.com. Morison. No. Nord Pool. W. Vol.13. pp. pp. Vol. July 18–22. 614–626.nerc. Pennsylvania. R.energy. 2.no/publ/rapport/97/tr4602.” IEEE Computer Applications in Power. P.” Bulk Power System Dynamics and Control IV Restructuring. Edmonton. Greece.” Proceedings of 1998 American Control Conference. www. 8-23 . “Legal Responsibility for Reliability in the New Competitive Electricity Markets in Canada: Who Do I Sue if the Lights Go Out?” plenary session of the IEEE/PES Summer Meeting.8-8 North American Electric Reliability Council. Kundur. McGraw-Hill. Power System Stability and Control. Statnett. September 29–30. 8-18 Deregulation of the Nordic Power Market. Kundur. 611–616. April 1995. Santorini. 1994. 8-10 P. Norwegian Electric Federation 1997 (http://www. Implementation and Experiences 1991–1997. September 1997. 2. “Information. G.no/. “Emerging Liability Issues for the New Electric Power Industry. June 24–26. Day. See also: http://www. 1 Conclusions 1. and the various information technologies such as digital sensors and signal processing. The primary stability controls are fast fault clearing and generator excitation control..Chapter 9 Conclusions and Suggested Future Work Angle stability control is an old power system problem. 5. with many effective solutions. Technologies include high voltage power electronics. power transactions may be very different than planned. intelligent controls. Digital controls should not be simple replicas of analog controls.g. control mode shifting. 2. Possibilities for control adaptation. however. presents new challenges. 3. We wish to exploit recent and emerging technologies for the development of cost-effective advanced stability controls. fiber optics communications. Special feedforward controls such as generator tripping for severe disturbances are very effective and are widely used. New long-distance interconnections on several continents present synchronous stability challenges. superior observability) exist. Direct control of rotor angle is not normally appropriate. competitive environment for the generation subsystem. For cost and reliability/complexity reasons. Control and communication technologies allow wide-area control where benefits (e. The present deregulated. Generator excitation control and control of other existing actuators should be fully exploited before considering transmission level mechanically-switched or power electronic controlled equipment. High damping ratio for oscillation damping or “stiff” (high synchronizing power) performance may not be cost-effective. fuzzy logic) are promising? 9. and different control structures should be considered. GPS. . with need to increase stability-related transfer limits. Questions investigated by the task force include: • What is the value and application of wide-area (centralized) stability control? • What is the value and application of direct control of rotor angles? • What are needs for adaptive control? • What new control techniques (examples: robust control theory. The purpose of stability controls is to remove stability-imposed limits on power transfer. and advanced control theory. local control strategies are the first choice. digital communications. digital controls. 4. These are available for special needs. One example is generator automatic voltage regulation. Digital fiber optic communication is rapidly becoming available. Direct load control is facilitated by informationage technology. 7. 9. Simulations must include sensitivity analysis of various operating/ disturbance conditions and other uncertainties. 9-2 . hardware and software algorithm failure modes and frequency should be investigated.6. Simulations should be validated by field tests and system monitoring. Rather. Wide-area control based on new communication technologies. Transmission-level power electronic equipment offers many possibilities for powerful stability control.2 Areas for Future Work 1. 12. 4. 3. Power plants should be able to withstand voltage and frequency excursions associated with islanding and other abnormal conditions. requirements to maintain stability of synchronous generators remain. A particular challenge for on-line interarea stability assessment is state estimation for power systems spanning large portions of a continent. Emerging technologies such as low earth orbit satellites are promising. Overall power system engineering for stability is required. and ongoing development may make the equipment cost-effective for more widespread use. 11. 8. This requires development of accurate models and data sets. Wide-area monitoring of power plants and substations is desirable to support stability control implementation and operation. Synergies are possible between stability control and control center EMS (energy management system) applications. Further exploitation of digital control possibilities that break paradigms established during the decades of analog control development. Modulation of steam and gas turbines mechanical power for damping of low frequency oscillations. along with the consequences of failures. 9. Integration of control center data/application programs with stability controls. or as the database for pattern-recognition based controls. Control reliability should not be based on simple redundancy requirements. Other requirements are suitable for ancillary service arrangements. Stability controls may include load shedding and controlled separation. Time and frequency domain simulations are essential for robust stability control design and for control certification. “Defense-in-depth” and “multiple lines of defense” are essential to minimize catastrophic power system instability and widespread outages because of rare multiple outages and failures. Some system requirements should be mandated. Dynamic security assessment may be used for control arming and adaptation. 10. With independent ownership of generation. 2. This will include better-defined mandatory practices with enforcement. and also ancillary service markets for power system stability enhancing controls.5. Strategies and criteria for stability control in the partially deregulated and restructured electric power industry. 9-3 .