本論文之主旨在於將支撐向量機應用於電力運轉資料之鑑別，並將其應用於IEEE 30-BUS系統之模擬測試，以判斷本文所提方法應用於實際電力系統之可行性。而由於每個匯流排所測量到的電力運轉資料不同，因此各匯流排支向機於訓練時須各自調整數個參數，方期能擁有較好的資料鑑別能力。因此，本文即針對參數資料之選擇，首先利用格點搜尋法於可能的參數空間取樣評估，以判斷各支向機最佳解可能位置，再依評估結果定義出較小區域，續利用進化演算法於劃定之小區域中，執行各支向機最佳解之搜尋。此外，為減少驗證之時間，本文並採用增量學習法訓練支向機，輔以利用減量移除法，執行LOO交叉驗證，以有效評估支向機之資料鑑別能力。經由本文之測試結果，本論文所提之方法於電力系統之資料鑑別上，其計算效能確有大幅提昇，應有助於電力系統之運轉與控制。

　　In this thesis, the method of Support Vector Machines (SVM) is proposed to identify the operation state of a 30-bus power system. In view of difference among measured data found at each bus, an integrated approach has been proposed for the training of each SVM distributed at each bus as well as the tuning of corresponding parameters, anticipating that the data identification performance can be largely improved. In such a method, it begins with the employment of grid search such that each model formulated through the collected data and related parameters can be better evaluated, hence justifying the near-optimal solution and determining an appropriate area for a further search. This is followed by the application of evolutionary algorithms for the search of SVM within the aforementioned area. Meanwhile, in order to decrease the training time, an incremental learning method was utilized to facilitate the training process, while a decremental unlearning was employed to perform the leave-one-out (LOO) cross-validation. Through the results obtained from the tested made in this thesis, the outcome is found to demonstrate a highly improved performance that is deemed beneficial for power system operation and control.

[1]V. Vapnik, Estimation of Dependences Based on Empirical Data, New York: Springer-Verlag, 1982.
[2]V. Vapnik, The Nature of Statistical Learning Theory, New York: Springer-Verlag, 1995.
[3]N. Cristianini and J. S. Taylor, An Introduction to Support Vector Machines, Cambridge University Press, Cambridge, United Kingdom, 2000.
[4]M. A. Hearst, S. T. Dumais, E. Osman, J. Platt and B. Scholkopf, “Support Vector Machines,” IEEE Intelligent Systems, Vol. 13, No. 4, July-Aug. 1998, pp. 18-28.
[5]K. R. Muller, S. Mika, G. Ratsch, K. Tsuda and B. Scholkopf, “An Introduction to Kernel-Based Learning Algorithms,” IEEE Transactions on Neural Networks, Vol. 12, No. 2, March 2001, pp. 181-201.
[6]E. Osuna, R. Freund and F. Girosi, “An Improved Training Algorithm for Support Vector Machines,” Proceedings of the 1997 IEEE Workshop on Neural Networks for Signal Processing, Florida, USA, Sept. 1997, pp. 276-285.
[7]G. Cauwenberghs and T. Poggio, “Incremental and Decremental Support Vector Machine Learning,” in Advances in Neural Information Processing Systems, Vol. 13, 2001.
[8]M. Pontil and A.Verri, “Properties of Support Vector Machines,” Neural Computation, Vol. 10, 1997, pp. 955-974.
[9]J. K. Martin and D. S. Hirschberg, “Small Sample Statistics for Classification Error Rates I: Error Rate Measurements,” Department of Information and Computer Science, UC Irvine, Tech. Rep. 96-21, 1996.
[10]V. Vapnik and O. Chapelle, “Bounds on Error Expectation for Support Vector Machines,” Neural Computation, Vol. 12, No. 9, Sept. 2000, pp. 2013-2036.
[11]C. M. Rocco S and J. A. Moreno, “System Reliability Evaluation Using Monte Carlo & Support Vector Machine,” Proceedings of the 2003 Annual Reliability and Maintainability Symposium, California, USA, Jan. 2003, pp. 482-486.
[12]G. Cauwenberghs, “Kernel Machine Learning: A Systems Perspective,” IEEE International Symposium on Circuits and Systems, Tutorial Guide: ISCAS 2001, Sydney, Australia, May 2001, pp. 4.1.1-4.1.7.
[13]S. V. N. Vishwanathan and M. Narasimha Murty, “SSVM: A Simple SVM Algorithm,” Proceedings of the 2002 International Joint Conference on Neural Networks, Hawaii, USA, Vol. 3, May 2002, pp. 2393-2398.
[14]T. Onoda, H. Murata, G. Ratsch and K. R. Muller, “Experimental Analysis of Support Vector Machines with Different Kernels Based on Non-Intrusive Monitoring Data,” Proceedings of the 2002 International Joint Conference on Neural Networks, Hawaii, USA, Vol. 3, May 2002, pp. 2186-2191.
[15]X. Yao, “Global Optimisation by Evolutionary Algorithms,” Proceedings of the 1997 Parallel Algorithms/Architecture Synthesis, Second Aizu International Symposium, Aizu, Japan, March 1997, pp. 282-291.
[16]J. H. Holland, Adaptation in Natural and Artificial Systems, Cambridge, MA: The MIT Press, 1992.
[17]K. S. Tang, K. F. Man, S. Kwong and Q. He, “Genetic Algorithms and Their Applications,” IEEE Signal Processing Magazine, Vol. 13, No. 6, Nov. 1996, pp. 22-37.
[18]G. Lin and X. Yao, "Analysing Crossover Operators by Search Step Size,” IEEE International Conference on Evolutionary Computation, Indiana, USA, April 1997, pp. 107-110.
[19]L. J. Fogel, A. J. Owens and M. J. Walsh, Artificial Intelligence Through Simulated Evolution, New York: Wiley, 1966.
[20]T. Back and H. P. Schwefel, “Evolutionary Computation: An Overview,” Proceedings of IEEE International Conference on Evolutionary Computation, Nagoya, Japan, May 1996, pp. 20-29.
[21]X. Yao, Y. Liu and G. Lin, “Evolutionary Programming Made Faster,” IEEE Transactions on Evolutionary Computation, Vol. 3, No. 2, July 1999. pp. 82-102.
[22]H. K. Birru, K. Chellapilla and S. S. Rao, “Local Search Operators in Fast Evolutionary Programming,” Proceedings of the 1999 Congress on Evolutionary Computation, Washington DC, USA, Vol. 2, July 1999, pp. 1506-1513.
[23]D. B. Fogel, “An Introduction to Simulated Evolutionary Optimization,” IEEE Transactions on Neural Networks, Vol. 5, No. 1, Jan. 1994. pp. 3-14.
[24]Y. Liu and X. Yao, “How to Control Search Step Size in Fast Evolutionary Programming,” Proceedings of the 2002 Congress on Evolutionary Computation, Hawaii, USA, Vol. 1, May 2002, pp. 652-656.
[25]K. Chellapilla, “Combining Mutation Operators in Evolutionary Programming,” IEEE Transactions on Evolutionary Computation, Vol. 2, No. 3, Sept. 1998, pp. 91-96.
[26]H. Ke and X. Zhang, “Editing Support Vector Machines,” Proceedings of the International Joint Conference on Neural Networks, Washington DC, USA, Vol. 2, July 2001, pp. 1464-1467.
[27]P. Xu and A. K. Chan, “An Efficient Algorithm on Multi-Class Support Vector Machine Model Selection,” Proceedings of the 2003 International Joint Conference on Neural Networks, Oregon, USA, Vol. 4, July 2003, pp. 3229-3232.
[28]L. Xuefeng and L. Fang “Choosing Multiple Parameters for SVM Based on Genetic Algorithm,” Processings of the 6th International Conference on Signal Processing, Beijing, China, Vol. 1, Aug. 2002, pp. 117-119.
[29]A. T. Quang, Q. Zhang and X. Li, “Evolving Support Vector Machine Parameters,” Proceedings of the First International Conference on Machine Learning and Cybernetics, Beijing, China, Vol. 1, Nov. 2002, pp. 548-551.
[30]L. S. Moulin, A. P. A. da Silva, M. A. El-Sharkawi and R. J. Marks II, “Support Vector Machines for Transient Stability Analysis of Large-Scale Power Systems,” IEEE Transactions on Power Systems, Vol. 19, No. 2, May 2004, pp. 818-825.
[31]J. J. Grainger and W. D. Stevenson, Power System Analysis, McGraw-Hill, Inc. Press 1994.
[32]李仁豪，有界支向機使用RBF核心之模型選取，碩士論文，國立台灣大學，中華民國九十年六月。
[33]江其峰，支持向量機的特性篩選方法，碩士論文，東海大學，中華民國九十一年六月。