Coordinating Power Oscillation Damping Control using Wide Area Measurements

1 Coordinating Power Oscillation Damping Control using Wide Area Measurements E. Johansson, K. Uhlen, Member, IEEE, A. B. Leirbukt, Member, IEEE, P. Korba, Member, IEEE, J. O. Gjerde, Member, IEEE, L. K. Vormedal, Member, IEEE, Abstract In this paper, the potential benefits of a Wide Area Control System for coordinated power oscillation damping control is investigated for the Nordic power system, with an overall motivation to facilitate increased power transfer limits. Several approaches to the design of power system stabilizers making use of phasor measurements from a wide area monitoring system are presented and compared with conventional stabilizers using locally measured control feedback signals. Linear analysis and time domain simulations illustrate the performance of these PSS designs when applied to selected SVCs in the Norwegian power transmission grid. Utilizing remote signals available recently through wide area monitoring systems enables selection of the best feedback control signal with highest modal observability of the modes of interest. Preliminary conclusions indicate that this leads to a higher performance and robustness of the power system stabilizer control. Index Terms--Controllability, linear analysis, observability, power system stabilizers, static VAR compensation, wide area measurements H I. INTRODUCTION igher demand on inter-area power transfer results as new and larger power markets evolve and as a general consequence of load increase. Congestion issues leading to unwanted price differences in the markets are getting more and more attention. Transfer limits are determined from power system security considerations and are either constrained by thermal, voltage or stability limits [1]. Improved damping of inter-area power oscillations is identified as one of the means to increase stability limits on transmission corridors. This can be achieved by installations and proper tuning of Power System Stabilizers (PSSs) in the system. Many studies are performed with the goal to increase damping of low damped oscillations, often with focus on generator applications, FACTS equipment using local measurements, and damping of local modes [2] - [5]. In this study the focus is on damping improvements through SVCs. This area is of high interest since SVCs are E. Johansson is with SINTEF, Trondheim, Norway, (emil.johansson@sintef.no) K. Uhlen is with Sintef, Trondheim, Norway, (kjetil.uhlen@sintef.no) A. B. Leirbukt is with ABB, Oslo, Norway (albert.leirbukt@no.abb.com) P. Korba is with ABB, Baden, Switzerland (petr.korba@ch.abb.com) J. O. Gjerde is with Statnett, Oslo, Norway (jan.gjerde@statnett.no) L. K. Vormedal is with Statnett, Oslo, Norway (lars.vormedal@statnett.no) 978-1-4244-3811-2/09/$25.00 2009 IEEE fully controlled by the Transmission System Operators (TSOs). The availability of generators and their PSSs, on the other hand, depends on the market settlements at any given hour. The main objective of the work has been to investigate the design and tuning of SVC stabilizers with the motivation to increase transfer limits on Power Transmission Corridors (PTCs) between main areas. More specifically the goal has been to investigate coordinated stabilizer design to fully exploit the potential of SVCs in providing damping of inter-area power oscillations, with the possibility to use input signals from wide area measurements utilizing Phasor Measurement Units (PMUs), and finally to identify potential benefits and challenges of using wide area measurements compared to traditional local power measurements. This paper presents a computer simulation study on the use of WAMS measurements in a damping control scheme, exploring the effectiveness of using voltage phasor angle difference as input signals to PSS controllers. Several SVCs are installed in the Norwegian power transmission grid shown in Fig. 3. Some of these SVCs are also equipped with PSS to damp power oscillations. These PSSs are fed with local branch power measurements as input signals, and have successfully been in operation for decades. Local measurements for damping control purposes are convenient since there is no need to communicate a real time control signal over a large geographic distance. However, local measurements are not always favorable when inter-area modes are concerned [6] & [7]. Phasor measurements provided by WAMS provide new opportunities for designing damping control schemes. With high performance communication infrastructure, WAMS provides the opportunity to select a control feedback signal from any PMU-equipped substation in the transmission grid. Voltage phasor angle signals are well suited to measure power system oscillations, since there is no need to monitor a particular branch or multiple branches. In the power transmission community, it is therefore widely expected that WAMS can enhance the damping performance of PSSs, however little field experience exists to support this.

2 The paper is organized as follows: In Section II, a brief description of the studied model and employed analyses methods are given. Section III presents the case study, with linear analysis, model development, and time domain simulations. A summary of result and findings are given in Section IV. II. METHOD A. General 1) Stabilizer Design and Control Structure Two different approaches to stabilizer design are investigated. In the first approach (called single-mode regulator) individual SVCs are selected to focus on damping of one mode. In the second approach (called multi-mode regulator) each SVC is tuned to provide damping of several modes. See Fig. 1 and Fig. 2 for block diagrams illustrating the single- and multi-mode regulators, respectively. Fig. 1. Single-mode regulators, indicating that each SVC is equipped with a stabilizer, possibly with multiple inputs, that is tuned to damp one particular mode. Coordination is achieved by tuning stabilizers on several SVCs to damp different modes. 2) Linear Study Methods Identification of oscillatory modes and studies of the system s sensitivity to these modes are often made using linear analysis. Control design and tuning can be performed through analysis of parameters identified in the linear analysis, where coordinated tuning and minimization of potential adverse regulator effects are important considerations. In this study, the following linear analysis techniques are used: Eigenvalue analysis used for identification of oscillatory modes in the system and transfer function zeroes. Controllability analysis used for identification of optimal equipment for stabilizing purposes. Controllability levels, of for example SVC voltage controller set-points, are analyzed using controllability factors (or alternatively Transfer Function Residues when measurement signals are identified). Observability analysis used for classification of modes (local, inter-area, etc.), as well as identification of optimal selection of measurement signals. Observability levels, of for example voltage angles, are analyzed using mode shape plots. Frequency responses of selected transfer functions used for controller design and tuning. Root-locus plots used for coordinated regulator design, tuning of controller gain and analysis of adverse regulator effects. Transfer function residues are often used for identification of optimal placement for PSS with local measurement [4], [9], [10]. In a wide area control system, however, the observability and controllability factors can be studied independently to identify and pair optimal measurements with the units that provide best controllability in the system. Fig. 2. Multi-mode regulator, indicating that each SVC is equipped with a more complex stabilizer, possibly with multiple inputs, that is tuned to damp more than one mode. Coordination is achieved by careful tuning and selection of inputs to avoid adverse interactions. The working hypothesis has been, first, that a careful choice of phasor measurements as input signal provides for better observability of selected modes (also being less disturbed by other local modes of less interest), and second, that selecting SVCs to focus on damping of particular modes provides for a coordinated design and maximum controllability. Together this should have the potential to improve performance compared to individual tuning of local stabilizers. 3) Time Domain Simulations Time domain simulations of minor events, such as for example drop of loads, are investigated to illustrate the PSS performance in the small signal stability range. In addition, severe disturbances, typically 3-phase faults with or without line clearing, are also analyzed to verify robust performance when controller saturation effects can be expected. B. Study Model 1) Nordic System The study has been carried out using a detailed model of the Nordic transmission system, with approximately 3000 buses, 4000 branches and 1100 generators. The modeled system corresponds to a high load scenario, with a total of 60 GW load, and a surplus generation in Norway, resulting in 1800MW export to Sweden across the main power transmission corridor between Sweden and southern Norway (Hasle PTC).

3 TABLE I gives an overview over the generation and load in each country, and in Fig. 3 the Nordic grid is shown. TABLE I LOAD FLOW STUDY CASE, NATIONAL OVERVIEW Country P gen Q gen P load Q load (GW) (Gvar) (GW) (Gvar) Finland 12.4 2.1 12.1 2.6 Norway 19.1 3.0 16.6 3.2 Sweden 25.7 3.8 27.6 6.5 The static and dynamic study models used for linear and time domain analyses (with software packages PacDyn [7] and PSS/E [8], respectively) are basically identical, simplifying identification and analysis of the study result. N: Nedre Røssåga PMU R: Rød SVC K: Kristiansand PMU S: Sylling SVC Fig. 3. Nordic power grid with indicated PMU and SVC sites 2) SVC and PSS Model SVCs were modeled using the standard PSS/E static var device model CSVGN5. The model s main functionality is described by the block diagram shown in the upper section of Fig. 4. s are represented by a user defined model implemented in both PSS/E and PacDyn. The input signal consists of the angle difference between two voltage phasors. s are represented using the PSS/E model STBSVC, with local branch electrical power measurement as input signal. Both these models are described by the block diagram shown in the lower section of Fig. 4. Fig. 4. SVC & PSS Model block diagram III. CASE STUDY A. Linear Analysis 1) Eigenvalue analysis PacDyn is able to automatically identify eigenvalues for models with up to 3500 state variables [11]. However, the studied model has almost 10000 state variables and therefore the eigenvalues must be identified by iterative techniques or by model reduction. The characteristic oscillatory modes in the Nordic power system have been studied thoroughly in past publications, [1]. Two low damped inter-area power oscillatory modes exist, with frequencies of approximately 0.3 and 0.5 Hz. These modes have as well been observed in the system on several occasions, [7]-[10]. In this study, the focus of the eigenvalue analysis has therefore been to manually identify low damped modes in the area around 0.5 Hz. In TABLE III five identified low damped modes in the frequency range 0.3-0.8 Hz are listed. These identified modes are labeled after their oscillatory frequencies, and are in this study referred to as the 0.33, 0.48, 0.55, 0.62, 0.76, and 0.80 Hz modes, respectively. 2) Controllability analysis The controllability analysis focused on identifying which of the existing SVCs are best suited to damp each of the above identified modes. The SVC controllers voltage setpoints were used as control variables, and the three SVCs showing highest controllability are listed in TABLE II. The controllability factors are normalized, showing the highest factor equal to 1.0. TABLE II CONTROLLABILITY FACTORS Oscillatory SVC Controllability factor Frequency #1 #2 #3 0.33 Hz Sylling: 1.0 Hasle: 0.9 Rød: 0.8 0.48 Hz Sylling: 1.0 Rød: 0.9 Hasle: 0.7 0.55 Hz Viklandet: 1.0 N. Røssåga: 0.9 Tunsjødal: 0.6 0.62 Hz Kvandal: 1.0 Viklandet: 0.85 Kristiansand: 0.5 0.76 Hz Kristiansand: 1.0 Rød: 0.7 Hasle: 0.3

4 The control design objective was to increase the damping of the 0.33Hz and 0.48Hz inter-area modes by optimal selection of SVC units for PSS implementation. The controllability assessment indicates that damping of the 0.33 and 0.48 Hz modes would be best achieved using the SVC in Sylling, with the SVC in Rød as second choice. 3) Observability analysis In the observability analysis, the identified modes are first classified, and then optimal PSS measurements are identified. Fig. 5 and Fig. 6 shows mode shape plots for the 0.33 and 0.48 Hz modes, respectively. Bus voltage angles were used here as observation variables in both cases. Northern Sweden Central Sweden Southern Sweden Northern Norway Central Norway Southern Norway Finland Eastern Denmark Fig. 5. Observability of 0.33Hz mode (Mode-shapes shown by area). Northern Sweden Central Sweden Southern Sweden Northern Norway Central Norway Southern Norway Finland Eastern Denmark Fig. 6. Observability of 0.48Hz mode (Mode-shapes shown by area). From Fig. 5, the 0.33 Hz mode is identified as an inter-area mode with best observability in Finland and southern Norway. Similar results for all the identified modes are summarized in TABLE III. Oscillatory Frequency 0.33 Hz 5.40 % 0.48 Hz 2.48 % 0.55 Hz 3.39 % 0.62 Hz 6.20 % 0.76 Hz 1.48 % TABLE III IDENTIFIED OSCILLATORY MODES Relative Mode Classification & Observability Damping Inter-area mode, observable as power oscillations between Finland and Southern Norway Inter-area mode, observable as power oscillations between Sweden and Southern Norway Local area mode, observable in northern Norway Inter-area mode, observable as power oscillations between Sweden and Central Norway Local area mode, observable in western Norway For a wide area PSS, the optimal PMU measurements are now identified, using the voltage angle observability factors. Kristiansand and Nedre Røssåga are identified as best suited for observation of the 0.33 & 0.48 Hz modes 1. In a similar way, power flow observability factors can be used for identification of optimal measurements for a PSS with local measurements. Analysis of the 420 kv lines connected to the SVCs in Sylling and Rød, identifies the highest observability levels on the 420 kv lines Sylling- Tegneby and Rød-Tveiten, respectively. B. Power System Stabilizer Design 1) Control Loop Several control loops were designed using the most promising SVC and measurement combinations, as shown in TABLE IV. TABLE IV PSS CONTROL LOOPS PSS Controlling SVC Observation measurement Sylling Sylling Voltage angle difference: Kristiansand - Nedre Røssåga Sylling Sylling Power flow: Sylling-Tegneby Rød Rød Voltage angle difference: Kristiansand - Nedre Røssåga Rød Rød Power flow: Rød-Tveiten 2) Controller Design Four single- and dual-mode controllers were analyzed, tuned for the 0.33 & 0.48 Hz modes as: Sylling PSS single-tuned for the 0.33 Hz mode Sylling PSS dual-tuned for the 0.33 & 0.48 Hz modes Rød PSS single-tuned for the 0.48 Hz mode Rød PSS dual-tuned for the 0.33 & 0.48 Hz modes 1 It should be noticed that the optimal PSS measurement points does not correspond with the actual PMU installations in Norway. E.g. using the difference in measurement from one bus in Finland and one in southern Norway would be resulting in a considerably higher observability level for some modes than by only using PMUs in Norway. This study is, however, only focusing on the existing equipment in Norway.

5 Based on computation of transfer function residues, the lead-lag filters (see Fig. 4) were tuned to provide phase shifts corresponding to a left shift of the targeted mode in closed loop. TABLE V PARAMETERS SVC Mode K T M T F T 1 T 2 Filter order Sylling 0.33 Hz 0.41 0.02 3 0.65 0.36 2 Rød 0.48 Hz 0.14 0.02 3 0.71 0.15 2 Sylling Dual 0.05 0.02 3 0.79 0.14 2 0.45 0.02 3 0.71 0.34 2 Rød Dual 0.70 0.02 3 0.55 0.4 2 0.14 0.02 3 0.71 0.15 2 In order to perform a fair comparison of the Wide Areaand local PSSs the nominal gains were selected to give similar gain margins. Chosen control parameters for the wide areaand local PSSs are shown in TABLE V and TABLE VI, respectively. TABLE VI LOCAL PSS PARAMETERS SVC Mode K T M T F T 1 T 2 Filter order Sylling 0.33 Hz 0.07 0.02 3 0.82 0.28 1 Rød 0.48 Hz 0.01 0.02 3 1.30 0.09 1 Sylling Dual 0.07 0.02 3 0.82 0.28 1 0.07 0.02 3 1.66 0.06 1 Rød Dual 0.01 0.02 3 0.00 0.00 1 0.01 0.02 3 1.30 0.09 1 3) Regulator Tuning / Optimization Effects of changes in controller gain for different PSS combinations are initially studied using root-locus plots. The controller gains were varied from zero to the suggested values indicated in TABLE V and TABLE VI, and the results in terms of closed loop eigenvalues are shown in the root-locus plots of Fig. 7 and Fig. 8. In these figures, the start of a trajectory (i.e. controller gain equal to zero) is marked with a square, while the trajectory ending is marked with a circle. The analysis shows that it is possible to damp the 0.33 Hz mode, and in particular the 0.48 Hz modes quite well with all studied PSSs. However, one observation not explicitly shown in this paper, is that stabilizers only tuned for the 0.33 Hz mode have an adverse effect on the 0.48 Hz mode at higher gains. This was noticed in both the local and wide area PSS solutions, but less pronounced in the single-mode wide area case. Best overall damping is obtained in the wide area case, with the two single-mode stabilizers controlling the Sylling and Rød SVCs in parallel. This is mainly due to zero-pole cancellation of the 0.55, 0.62 and 0.76 Hz modes for the local PSSs that is not seen in the wide area case. The zero-pole cancellation has the influence that these modes can neither be observed nor controlled by a local PSS in Sylling or Rød. Adverse effects were also noticed for other modes, which further emphasize the importance of coordinated regulator tuning. 2% Damping Ratio 5% Damping Ratio 0.1 10% Damping Ratio -0.5-0.45-0.4-0.35-0.3-0.25-0.2-0.15-0.1-0.05 0 0 Re [1/s] Fig. 7. Root-locus plot: Sylling & Rød, Single-tuned 2% Damping Ratio 5% Damping Ratio 10% Damping Ratio 0.63 Hz 0.63 Hz 0.56 Hz 0.48 Hz 0.33 Hz 0.56 Hz 0.48 Hz 0.33 Hz 0.76 Hz 0.76 Hz -0.5-0.45-0.4-0.35-0.3-0.25-0.2-0.15-0.1-0.05 0 0 Re [1/s] Fig. 8. Root-locus plot: Sylling & Rød LOCAL PSS, Single-tuned C. Time Domain Simulation Study The two single-mode regulators controlling Sylling & Rød SVCs in parallel were simulated in time domain for the following PSS configurations: in Sylling and Rød Wide area PSS in Sylling and Rød in Sylling and Rød First, a small disturbance in Hasle is simulated, and responses are shown in Fig. 9 - Fig. 11 for the active power flow response on lines between Norway-Sweden, Sweden- Finland, and reactive power response of the Sylling SVC, respectively. The main purpose of these simulations is to verify the control design and results from the linear analysis. 0.8 0.7 0.6 0.5 0.4 0.3 0.2 Im [Hz] 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Im [Hz]

6 Active Power [MW] 1220 1210 1200 1190 1180 1170 1160 1150 0 1 2 3 4 5 6 7 8 9 10 Fig. 9. Small disturbance: Active power flow Norway - Sweden 430 425 The small disturbance simulations show pronounced oscillations with a frequency of approximately 0.5 Hz, implicating that the 0.48 Hz mode is main contributor to the oscillations. A significant increase in damping is shown for both the local and wide area PSS models, relative damping levels are approximated to around 4% (no PSS), 12% (local PSS), and 9% (wide area PSS). These levels correspond quite well with the linear analysis relative damping levels (3.8%, 14.2%, 10.6%). Note that the control action (shown in Fig. 11 for the Sylling SVC) is lower using the wide area PSS than for the local PSS, indicating that there is a potential for even further damping using the wide area PSS. Next, a larger disturbance (a three phase short circuit in Hasle with subsequent line trip Norway-Sweden) is simulated in order to assess the performance of the stabilizer design subject to one of the most critical contingencies at the given operating condition. Results are shown in Fig. 12 - Fig. 14, for the active power flow response on lines between Norway- Sweden (parallel to the tripped line), Sweden-Finland, and voltage response in Rød, respectively. Active Power [MW] 420 415 410 405 400 0 1 2 3 4 5 6 7 8 9 10 Fig. 10. Small disturbance: Active power flow Sweden-Finland 100 Active Power [MW] 2500 2000 1500 1000 0 2 4 6 8 10 12 14 16 18 20 Fig. 12. Large disturbance: Active power flow Norway-Sweden Reactive Power [Mvar] 50 0-50 0 1 2 3 4 5 6 7 8 9 10 Fig. 11. Small disturbance: Reactive power of SVC in Sylling Active Power [MW] 700 600 500 400 300 200 100 0 0 2 4 6 8 10 12 14 16 18 20 Fig. 13. Large disturbance: Active power flow Sweden-Finland

7 Voltage [kv] 420 415 410 405 400 395 390 0 2 4 6 8 10 12 14 16 18 20 Fig. 14. Large disturbance: Voltage at 420 kv node Rød The large disturbance simulations show oscillations with a frequency of approximately 0.3 Hz, implicating that the 0.33 Hz mode is the main contributor in this case. Also here an increase in damping is shown for both the local and wide area PSS models, relative damping levels are approximated to around 5% (no PSS), 7% (local PSS), and 10% (wide area PSS)). These levels correspond quite well with the linear analysis relative damping levels (6.3%, 8.5%, 10.0%). Note also that the voltage responses (Fig. 14) are improved with the chosen stabilizers, even at the high voltage nodes where the SVCs are connected. It is further observed that both regulator designs provide increased damping without excessive utilization of the SVCs, for both simulations. IV. DISCUSSION OF RESULTS AND CONCLUSIONS This study has focused on power oscillation damping improvements using SVC units. The objective has been to investigate coordinated stabilizer design to fully exploit the potential of SVCs in providing damping of inter-area power oscillations, with the motivation to increase transfer limits between areas. In particular, the use of remote measurements provided by WAMS as PSS input signals has been explored, and case studies on the Nordic power transmission system have been performed. Utilizing such remote signals enhances the options for the selection of the best feedback control signal which can be seen as an inherent part of the PSS design: selecting the feedback signal with highest modal observability of the mode of interest (among all available, i.e. local and remote ones) increases the performance and robustness of the PSS control. Hence, the additional benefit and potential of the wide area approach is that it is easier to design a set of stabilizers that provide improved damping of several modes, and that this may be a more robust solution in the sense that performance is improved over a larger range of operating conditions and subject various contingencies. For the Nordic power transmission system, PSSs for the Norwegian SVCs in Sylling and Rød with phasor measurements from Nedre Røssåga and Kristiansand were found to be good solutions for stabilizing the two critical inter-area modes: 0.33Hz and 0.48Hz. Analysis of local and wide area solutions with single- and dual-mode regulator designs and different regulator combinations showed that the best overall performance was achieved with Sylling & Rød SVCs regulated independently by single mode regulators in a wide area control system. Significant improvement in damping of the studied modes, as well as of other low damped modes, was observed in both linear analysis and time domain simulations. The results show that it is possible to use existing PMUs and SVCs in Norway to damp critical modes in the Nordic system, and that this can be achieved without excessive utilization of the SVCs. A number of practical considerations concerning the use of wide area measurements have not been addressed in this study, including the availability and robustness of the PMUs providing the wide area input signals, and the issue of communication system time delays. This has to be further assessed, but a robust implementation is expected to include an automatic fall-back to local signals when PMU signals are missing or delayed. Such a solution could even add robustness compared to the use of local measurements only. In particular this could be the case in a contingency situation involving outage of the locally measured line (or a nearby line), and thus making the local input signal useless. Future work to be considered: Optimization of proposed regulator parameters. Detailed study of dynamic influences of proposed WACS control, including effects of communication time delays. Feasibility study of large scale WACS, involving all PMU, FACTS, and HVDC components in the Nordic system, with intent to damp critical modes and to increase power transfer limits between several areas in the system. V. REFERENCES [1] K. Uhlen, M. Pálsson, T. R. Time, and J. O. Gjerde, Raising Stability Limits in the Nordic Power Transmission System, 14 th Power Systems Computation Conference (PSCC 02), Sevilla, 24-28 June 2002 [2] E. V. Larsen, J. J. Sanchez-Gasca, and J. H. Chow, Concepts for Design of FACTS Controllers to Damp Power Swings, IEEE Transactions on Power Systems, vol. 10 pp. 948-956, 1995 [3] J. H. Chow and J. J. Sanchez-Gasca, Pole-placement Designs of Powr System Stabilizers, IEEE Transactions on Power Systems, vol. 4 pp. 271-277, 1989 [4] M. J. Gibbard, D. J. Volwes, and P. Pourbeik, Interaction Between, and Effectiveness of, Power System Stabilizers and FACTS Device Stabilizers in Multimachine Systems, IEEE Transactions on Power Systems, vol. 15 pp. 748-755, 2000

8 [5] N. Martins and L. T. G. Lima,, Determination of suitable locations for power system stabilizers and static var compensators for damping electromechanical oscillations in large scale power systems, IEEE Transactions on Power Systems, vol. 5 pp. 1455-1469, 1990 [6] A. Heniche and I. Kamwa, Assessment of Two Methods to Select Wide- Area signals for Power System Damping Control, IEEE Transactions on Power Systems, vol. 23 pp. 572-581, 2008 [7] A. B. Leirbukt, J. O. Gjerde, P. Korba, K. Uhlen, L. K. Vormedal, and L. Warland, Wide area monitoring experiences in Norway, IEEE/PES Power Systems Conference and Exposition, vol 1-5 pp. 353-360, 2006 [8] K. Uhlen, L. Warland, J. O. Gjerde, Ø. Breidablick, M. Uusitalo, A. B. Leirbukt, and P. Korba, Monitoring Amplitude, Frequency and Damping of Power Systems Oscillations with PMU Measurements, in Proc. 2008 IEEE Power & Engineering Society General Meeting [9] K. Uhlen, N. Martins, O. B. Gjøsæter, and P. E. M. Quitão, Application of small signal stability analysis to the Nordel power system, VII Symposium of specialists in electric operational and expansion planning,, 21-26 May 2000 [10] K. Uhlen, S. Elenius, I. Norheim, J. Jyrinsalo, J. Elovaara, and E. Lakervi, Application of Linear Analysis for Stability Improvements in the Nordic Power Transmission System, in Proc. 2003 IEEE Power Engineering Society General Meeting, vol. 4, pp. 2097-2103. [11] http://www.pacdyn.cepel.br/, 2008-03-01, PacDyn is a software package for the analysis and control of small-signal stability of large power systems [12] http://www.pti-us.com/pti/software/psse/index.cfm, 2008-03-01, PSS E is a software tool used by electrical transmission participants world-wide. Petr Korba received his M.Sc. degree in electrical engineering from the Czech Technical University, Prague (1995) and Ph.D. degree in control engineering from the University of Duisburg, Germany (2000). He then became a member of staff at University of Manchester Institute of Science and Technology (UMIST), Control Systems Centre. In 2001, he joined ABB Switzerland Ltd. where he is currently with ABB Corporate Research as a principal scientist. Jan Ove Gjerde was born in Hareid in 1958. He received a M.Sc. in electrical engineering from The Norwegian Institute of Technology in 1983. He is running his own consultant company and since 2005 he has been a project manager for the Secure Transmission project at Statnett. Lars Vormedal received his M.Sc. degree in electrical engineering from the Norwegian University of Science and Technology (NTNU), Trondheim, Norway in 1976. He is currently Vice President, head of R&D, at Statnett. VI. BIOGRAPHIES Emil Johansson was born in Sweden 1975 and received his M.Sc. in electrical engineering at the Swedish Royal Institute of Technology (KTH), Stockholm, in 2002. Between 2001 and 2007 he worked for ABB with HVDC and harmonic filtering. He is currently with SINTEF Energy Research in Trondheim as a research scientist. Kjetil Uhlen (M 1995) was born in 1961. He received the Sivilingeniør degree from the Norwegian Institute of Technology in 1986 and a Ph.D. degree in control engineering from the same institute in 1994. Since 1987 he has worked at SINTEF Energy Research in Trondheim, presently as team leader and senior research scientist. Albert Leirbukt received his siv.ing (graduate degree) in electrical engineering from Norwegian University of Science and Technology (NTNU), Trondheim, Norway, in 1997. He is currently with ABB Network Management division in Oslo, and is product manager for WAMS.