CONTINGENCY RANKING IN POWER SYSTEMS EMPLOYING FUZZY BASED ANALYSIS

Transcription

1 CONTINGENCY RANKING IN POWER SYSTEMS EMPLOYING FUZZY BASED ANALYSIS MANJULA B. G. 1, SUMA A. P. 2, A. D. KULKARNI 2, NARENDRA KUMAR 1 AND T. ANANTHAPADMANABHA 2 1 Dept of Electrical Engineering, DTU, New Delhi, India 2 Dept of Electrical Engineering, The National Institute of Engineering, Mysore, India Abstract: In deregulated operating regime power system security is an issue that needs due consideration from researchers. Real power & voltage contingency ranking is an integral part of security assessment. The objective of contingency screening & ranking is to quickly & accurately shortlist critical contingencies from a large list of credible contingencies & rank them according to their severity for further rigorous analysis. A performance index (PI) is computed for each single line contingency using both conventional & Fuzzy based approach. To obtain the magnitudes of various parameters, a computer aided power system study software package which employ iterative methods are used. This paper presents an approach using fuzzy logic to evaluate the degree of severity of the conventional contingency & to eliminate the masking effect in the technique. Index Items: System Security, Real power, Voltage, Performance index (PI), Contingency ranking, and Fuzzy logic (FL) approach, SCADA I. INTRODUCTION The effect of the line outage when rest of system is stable is called contingency study. The study of contingency is an essential activity in planning operation & control of power systems. Contingency analysis allows systems to be operated defensively. The operator cannot take action fast enough when many of the problems that occur on a power system that causes serious trouble within fraction of time period leading to cascading failures. Because of this aspect of system operation, modern operations computers & SCADA systems are equipped with contingency analysis programs that model possible system troubles before they arise. The outage or change in the independent parameters of the power systems gives rises to transient phenomena in the electrical and electromechanical states of those power systems * manjulavittal@gmail.com [1]. The main thrust of contingency studies carried out in power system control centers is to determine the steady state effects of outages [2]. Large power systems require the analysis of all the credible contingence within a very short time so as to exercise the control in the short time available for corrective action [3]. In performance index method with high exponent, the resultant performance index value will depend heavily on loading of the particular line which is loaded closest to its limit, i.e. 90% to 95% of the rated capacity, other lines which are less heavily loaded i.e. 80% to 85% of the rated capacity though of large in number, will have relatively small weightage on the performance index value [4]. Ranking all possible contingencies based on their impact on the system voltage profile will help the operators in choosing the most suitable remedial actions before the system moves toward voltage collapse [15]. In [16], surveying possible contingencies with ranking according to line I J E E E S, 4(1) June

2 Manjula B. G., Suma A. P., A. D. Kulkarni, Narendra Kumar and T. Ananthapadmanabha FVSI indicator is carried out. The method of ranking the possible contingency based on right eigenvector and branch parameter specially in [17] is given. Appearing the artificial intelligence, possible contingency ranking is done based on neural networks [18]-[20], fuzzy logic [21], [22] and genetic algorithm [23]. In this paper we have concentrated on the line outages as the contingencies. To decide about the security level of the system every bus voltage & line power flow must be compared to its corresponding maximum & minimum tolerable values. Transmission line failures cause changes in the flows & voltages on the transmission equipment remaining connected to the system. Therefore the analysis of transmission failure requires methods to predict these flows & voltages so as to be sure they are within their respective limits. A scalar function called performance index (PI) which measures system stress is used in the calculation of the contingency ranking. A real power & voltage performance index is calculated which evaluates the severity of contingency derived from the current overload of lines. The operating state of the system is a function of time. It keeps on changing due to variation in load level at various buses or due to rescheduling of generation. To keep the system secure, it is imperative to know the impact of unplanned outages in advance so that suitable preventive / control measure scan be taken if necessary. II. EXPERIMENT The solution algorithm for real power & voltage contingency ranking is as follows: (1) Solve a load flow for base case. (2) Remove single line. (3) Calculate the numerical value of performance index by considering real power flow & voltage of the line. Conventional Performance Index 2n nl P IBASE PINEW PI(P) = 1 1 PIBASE 2n nb V IBASE VINEW PI(V) = 2 1 VIBASE Where nl = number of lines. Where nb = number of Buses. (4) Rank the contingencies from their PI values. (5) Repeat step 2 & 3 for different contingencies The PI list is formed & sorted. The Lower ranked contingencies are most severe than the higher one when real power is considered. The contingency ranking obtained by the conventional method is not précised. This is not a foolproof method of ranking contingencies. It is possible that some severe contingencies may be left out & also some not so severe contingencies may be ranked. There may be some other severe conditions are existing between each ranking. So this unmasking of severity between the two rankings is obtained with the fuzzy based approach. Fuzzy Performance Index = nl PIBASE PINEW FPI(P) 1 1 PIBASE = nb VIBASE VINEW FPI(V) 2 1 VIBASE Last ten years onwards fuzzy system application have received increasing attention in manufacturing industries, heavy industries, transportation systems, water supply systems. The applications are mainly for controllers, diagnosis, optimization & planning. In fuzzy set, the degree of belief of every fuzzy subset from belong to set to not belong to set are represented in a gradual transition which is called membership function. Membership function is representing the degree of belief of every subset 2 I J E E E S, 4(1) June 2012

3 in the universe of discourse in a number between 0 & 1. Several membership function shapes can be used to develop fuzzy contingency ranking such as triangular, trapezoidal, sigmoid, bell function. In this case trapezoidal membership function shapes were used & indicated in figure 1 is usually expressed by the characteristic points a, b, c, d such that the fuzzy number under study can assumed any value between a and d, but values within the range b and c are most likely to place. Figure 1: Trapezoidal Membership Function All the values within the range b and c have membership value (µ x = 1) that indicates complete membership for event. However, the values within ranges (a-b) and (c-d) have membership values (0 µ x 1), which indicates partial membership values. Any value outside the range of a and d has membership value ( µ x = 0) indicating non membership for parameter. Thus the uncertainty of the parameter x is conveniently characterized by a trapezoidal fuzzy distribution with suitable left right slopes. A specific relationship for the element x and its degree of membership µ x, for the trapezoidal membership function is presented in the equation Trapezoid (x, a, b, c, d) = max (min (x-a/b-a, 1, d-x/d-c), 0) Fuzzy Representation of Contingency Ranking The four characteristic points a, b, c, d are selected between each two rankings represents performance indices. Membership functions used to evaluate the severity of contingency ranking is divided into five categories using fuzzy set notation. Here the inputs are real power & voltage indices whereas the outputs are severity indices. Fuzzy set notations are Very small Small Medium Large Very Large (VS) (S) (M) (L) (VL) III.TEST SYSTEM STUDIES & RESULTS Results of Five bus & IEEE 14 bus Test systems are discussed below. Lower ranked contingencies are more severe than higher ranked contingencies. Contingency ranking on real power and voltage bases are calculated using both Fuzzy & conventional performance index. Contingency Ranking on Real Power Base Test case-1: Five bus system Table N L N M N S N L M M M Table 1 explains the contingency ranking for five bus system when real power flow in a line is considered by using fuzzy approach. N stands for negative PI and is unsafe. Line 2-5 is more severe & next comes Line 1-2. Table 2 explains contingency ranking for Five Bus system using conventional performance index. In both conventional and fuzzy methods the top 2 severe conditions remain same. I J E E E S, 4(1) June

4 Manjula B. G., Suma A. P., A. D. Kulkarni, Narendra Kumar and T. Ananthapadmanabha Table 2 Line Outage by Conventional Performance Index Index 1 L L L L L L L Test case-2: IEEE 14 bus System Table 3 explains the single line contingency ranking using Fuzzy performance index for IEEE 14 bus. Line 5-6 most severe & next comes Line 7-9. Table 4 explains the single line contingency ranking using conventional performance index for IEEE 14 Bus. Line 5-6 is most severe and next comes line Table 3 1 L 5-6 N M 2 L 7-9 N M 3 L 6-13 N VS 4 L 9-14 N L 5 L 2-3 N S 6 L 4-5 N S 7 L 1-2 N M 8 L 4-7 N L 9 L 9-10 N L 10 L 6-12 N M 11 L 4-9 N VS 12 L 6-11 N L 13 L 7-8 N M 14 L 2-4 N M 15 L N S 16 L N M 17 L 1-5 N VL 18 L M 19 L L 20 L L Table 4 Line Outage by Conventional PI Index 1 L L L L L L L L L L L L L L L L L L L L Contingency Ranking on Voltage Base Test case-1: Five bus system Table 5 1 L S 2 L M 3 L L 4 L M 5 L M 6 L M 7 L M Table 5 explains the contingency ranking using fuzzy performance index for five bus system on voltage base. Line 1-2 has the largest contingency index value and hence more severe. Next comes Line I J E E E S, 4(1) June 2012

5 Table 6 Line Outage by Conventional PI Sl No Line Contingency Index Rank 1 L L L L L L L that conventional method has masking effect & can be unmasked by this method. The original conventional ranking changes when fuzzy approach is used. Table 6 explains the contingency ranking for five bus system on voltage base. Line 1-2 has the largest contingency index value and hence more severe. Next comes Line 2-5. Test case-2: IEEE 14 Bus system Table 7 1 L 3-4 N S 2 L N M 3 L S 4 L M 5 L M 6 L S 7 L M 8 L S 9 L S 10 L L 11 L M 12 L VL 13 L VL 14 L M 15 L S 16 L M 17 L M 18 L L 19 L S 20 L M Similarly Table 7 explains the contingency ranking for IEEE 14 Bus system using Fuzzy performance index. Line 5-6 is most severe and next comes Line 7-9. The fuzzy approach justify Figure 2: IEEE 14 Bus Test System Table 8 Line Outage by Conventional PI Sl No Line Contingency Index Rank 1 L L L L L L L L L L L L L L L L L L L L I J E E E S, 4(1) June

6 Manjula B. G., Suma A. P., A. D. Kulkarni, Narendra Kumar and T. Ananthapadmanabha Table 8 above explains the contingency ranking for IEEE 14 Bus system using conventional performance index. Line 7-9 is most severe and next comes line 5-6. IV. CONCLUSIONS The proposed approach can provide the user with those outages that may cause immediate loss of load or islanding at a certain bus. This kind of information is very helpful to system operators. An overall severity index is given for each outage case, which can be used as a guide line for deciding whether corrective control actions should be taken or not. Conventional power flow analysis is carried out for computing bus voltage magnitude, voltage angle, real and reactive power of the systems. The fuzzy system has many advantageous features such as optimized system complexity, control of power flow, control of non linear system etc. The most critical & harmless outage scenarios for five bus & IEEE 14 bus systems are calculated by both conventional and fuzzy based analysis. Most critical contingencies can be identified correctly by using fuzzy based ranking system than conventional system. This method eliminates the masking effect of conventional methods of contingency ranking effectively. ACKNOWLEDGEMENT The author would like to thank Prof Narendra Kumar from Delhi Technological University, Delhi & Prof A D Kulkarni from The National Institute of Engineering, Mysore, Karnataka for their valuable suggestions and comments. REFERENCES [1] F. D. Galiana. Bound Estimates of the Severity of Line Outages. IEEE Transactions on PAS 1984 ; 103; [2] G. D. Irisarri and A. M. Sasson, An Automatic contingencyselection Method for On-line Security Analysis IEEE Trans. PAS, 100, pp , [3] Repo, S.; Jarentausta, P. Contingency Analysis for a Large Number of Voltage Stability Studies Electric Power Engineering, PowerTech Budapest 99. International Conference. [4] S N R K Srinivas et al., Application of Fuzzy Logic Approach for Obtaining Composite criteria Based Network Contingency Ranking for Practical Power System Networks International Journal of Computer Science & Communication, 1(1), 2010, [5] S. Chauhan. Department of Electrical Engineering, National Institute of Technology, Hamirpur , INDIA Fast Real Power Contingency Ranking using Counter Propagation Network: Feature Selection by Neuro-fuzzy model Received 24 March 2004; received in revised form 9 August 2004; accepted 9 August [6] JATIT C.V et al., Fuzzy Load Modeling & Load Flow study Using Radial Basis Function, Journal of Theoretical & Applied Information Technology [7] Ismail Musirin & Titik Khawa Abdul Rahman Simulation Technique for voltage Collapse Prediction & Contingency Ranking in Power System 2002 Student Conference on Research & Development proceedings, Shah Alam, Malaysia. [8] A Mohamed, G B Jasmon, Voltage Contingency Selection Technique for Security Assessment IEE Proceedings, 136, PT.C.No.1 Jan [9] Mario A. Albuquerque & Carlos A Castro A Contingency Ranking Method for Voltage Stability in Real Time Operation of Power Systems IEEE paper Accepted for Presentation at 2003 IEEE Bologna power tech. conference June 23-26, Bologna, Italy. [10] M. A. Kamarposti, Contingencies Ranking for Voltage Stability Analysis using Continuation Power Flow Method ISSN Electronics & Electrical Engineering No. 3(99). T190-Electrical engineering. [11] V. H.Ortiz M, G. Barajas R, J. A.Gutierrez Management of Contingencies Analysis at Mexican Power System R.2006 IEEE PES Transmission & Distribution conference & Exposition Latin America Venezuela. [12] Anant Oonsivilai & Renedy A.Greyson. Power System Contingency Analysis using Multiagent Systems World Academy of Science, Engineering & Technology [13] F.Fatehi, M. Rashidinejad & A. A. Gharaveisi Contingency Ranking Based on Voltage Stability Criteria Index IEEE, ; [14] A. A. Chowdhury, D. O. Koval Deregulated Transmission System Reliability Planning Criteria based on Historical Equipment Performance Data. Canada IEEE. [15] Majid Poshtan, Parviz Rastgoufard, and Brij Singh, Contingency Ranking for Voltage Stability Analysis of Large Scale Power Systems, Proceeding of IEEE / PES Power System Conference and Exposition, pp , Oct [16] Musirin and T. Kh. A. Rahnian, Fast Automatic Contingency Analysis and Ranking Technique for Power System Security Assessment, Student Conrerence on Research and Development (SCOReD) IEEE Proceedings, Putrajaya, Malaysia, I J E E E S, 4(1) June 2012

7 [17] Flueck, A. J., and Q. Wei. A New Technique for Evaluating the Severity of Branch Outage Contingencies Based on Two-Parameter Continuation, Proceedings of IEEE PES General Meeting, June 2003, pp [18] Mansour, Y., E. Vahedi, and M. A. El-Sharkawi. Dynamic Security Contingency Screening and Ranking Using Neural Networks. IEEE Transaction on Neural Networks. 8.4 (July 1997): [19] Jain, T., L. Srivastava, and S. N. SINGH. Fast Voltage Contingency Screening Using Radial Basis Function Neural Network. IEEE Transaction on Power Systems (Nov.2003): [20] Srivastava, L, S. N. Singh, and J. Sharma. Knowledge Based Neural Network for Voltage Contingency Selection and Ranking. IEEE Proceedings on Generation, Transmission and Distribution (Nov.1999): [21] Lo, K. L., and A. K. I. Abdelaal, Fuzzy Logic Based Contingency Analysis, Proceedings of Electric Utility De-regulation, Restructing, and Power Technologies, DRPT. London, April, [22] Mori, H., and E. Ando. Two Staged Simplified Fuzzy Inference for Dynamic Contingency Screening in Power systems., IEEE PES Sumer Meeting. 4(July 2000): [23] Sudersan, A., M. Abdelrahman, and G. Radman. Contingency Selection and Static Security Enhancement in Power Systems Using Heuristic- Based and Static Generic Algoritms, Proceedings of Thirty- Sixth Southeastern Symposium on System Theory. 1 (2004): [24] Les Pereira and Don DeBerry, Double Contingency Transmission Outages in a Generation and Reactive Power Deficient Area, IEEE Trans. on Power Systems 15, 2000, [25] K. Vu, M.M. Begovic, D. Novesel and M.M. Saha, Use of local measurement to estimate voltage-stability margin IEEE Transaction on Power Systems, 14(3), 1999, [26] C.W. Taylor, Power System Voltage Stability, McGraw- Hill, [27] F. Milano, Power System Analysis Toolbox, Version 1.3.4, Software and Documentation, July 14, I J E E E S, 4(1) June