本文針對三台不同瑕疵的氣體絕緣開關做辨識加壓試驗，施以高電壓至50kV，並於試驗期間以局部放電設備擷取局部放電訊號，將所得訊號經過濾波、化簡轉換成104 個特徵參量，接著這些局部放電資料可被區分成不同訓練架構，經由適應性類神經模糊推論系統(adaptive-network-based fuzzy inference system, ANFIS)訓練並測試，計算出各個特徵參量之辨識率，從中挑選出對於辨識瑕疵有實際效用的特徵參量，並驗證是否具有良好的排他性，因此瑕疵一選到第82個特徵參量為負放電區域的放電總合-高度分布之偏態，瑕疵二選到第8個特徵參量是正放電區域的放電次數，而瑕疵三的第16個特徵參量較符合我們的需求，其為正放電區域放電次數-相位分布之標準差，且所選擇到的特徵參量其辨識率皆達95%以上，以此做為往後瑕疵辨識的參考，從而節省萃取不必要特徵參量之時間，期待利用有效的特徵參量便可快速的判斷出瑕疵類型。

In this thesis, the partial discharge (PD) tests are conducted on three gas insulated switchgear (GIS) apparatus with different prefabricated defects. The test voltage is gradually raised to 50kV in order to retrieve partial discharge signals. The resulting signal is filtered and transformed into 104 features. Then, these partial discharge data can be divided into different training frameworks. We calculate recognition rates of the various features via adaptive-network-based fuzzy inference system (ANFIS). From the recognition rates of the various features, appropriate features can be chosen for the identification of defects. After verifying the great exclusive of the features, we choose the eighty-second feature for the defect 1, which is Q-negative-height distribution skewness-sum. The eighth feature for the defect 2 is Q-positive-quantity height-num. The sixteenth feature for the defect 3 meets our need which is Q-positive-phase distribution standard deviation-num. The recognition rates of the chosen features can reach more than 95% and they can be used as references for future identification, saving extraction time of unnecessary features. Using the effective features to determine a defect type quickly is the purpose of the thesis.

[1] Bo Qi, C. Li, B. Geng, and Z. Hao, “Severity diagnosis and assessment of the partial discharge provoked by high-voltage electrode protrusion on GIS insulator surface,” IEEE Transactions on Power Delivery, Vol. 26, pp. 2363-2369, 2011.
[2] 張建國，「高電壓地下電纜接頭絕緣狀態之監測與診斷系統之研究」，博士論文，國立台灣科技大學，2012。
[3] E. Gulski and F. H. Kreuger, “Computer-aided recognition of discharge sources,” IEEE Transactions on Electrical Insulation, Vol. 27, pp. 82-92, 1992.
[4] J. S. R. Jang, “ANFIS: adaptive-network-based fuzzy inference system,” IEEE Transactions on Systems, Man, and Cybernetics, Vol. 23, pp. 665-685, 1993.
[5] K. Dreisbusch, H.-G. Kranz, and A. Schnettler, “Determination of a failure probability prognosis based on PD-diagnostics in GIS,” IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 15, pp. 1707-1714, 2008.
[6] C. J. Jones, “CIGRE working group 13.09-monitoring and diagnostic techniques for switch equipment,” IEEE/PES Transmission and Distribution Conference and Exposition, Vol. 2, pp. 1083-1087, 2001.
[7] 林育勳，「數位局部放電量測應用於模鑄式比流器絕緣狀態之評估」，博士論文，國立台灣科技大學，2004。
[8] R. E. James and B. T. Phung, “Development of computer-based measurements and their application to PD pattern analysis,” IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 2, pp. 838-856, 1995.
[9] IEC60270, “High-voltage test techniques-Partial discharge measurements,” 2000.
[10] 朱德恒、嚴璋，「高電壓絕緣」，清華大學出版社，1990。
[11] M. Fenger, “Sensitivity assessment procedure for partial discharge measurements on transmission class cables,” IEEE International Symposium on Digital Object Identifier, pp. 667-671, 2008.
[12] 李忠、陳杰華、胡迪軍、馮允平，「外部傳感器GIS局部放電超高頻檢測的研究」，高電壓技術第29卷第6期，第35-37頁，2003。
[13] 張宏展，「局部放電應用於模鑄式比流器之絕緣狀態分析」，高壓局部放電及避雷器量測儀診斷技術研討會，2010。
[14] 潘彥竹，「局部放電量測系統之研製」，碩士論文，國立台灣科技大學，2004。
[15] 劉博文、陳春雄，「電力設備」，科友圖書股份有限公司，2006。
[16] 簡允堅、吳天得，「發變電工程」，全華科技圖書股份有限公司，2000。
[17] S. Meijer, J. J. Smit, and A. J. L. M. Kanters, “Experience with PD measurements in GIS maintenance concept,” International Conference on Condition Monitoring and Diagnosis, pp. 551-554, 2008.
[18] K. Dreisbusch, H.-G. Kranz, and A. Schnettler, “Determination of a failure probability prognosis based on PD-diagnostics in GIS,” IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 15, pp. 1707-1714, 2008.
[19] B. B. Mandelbort, “How long is the coast of Britain? Statistical self-similarity and fractional dimension,” Science, New Series, Vol. 156, pp. 636-638, 1967.
[20] N. Sakar and B. B. Chaudhuri, “An efficient differential box-counting approach to compute fractal dimension of image,” IEEE Transactions on Systems, Man and Cybernetics, Vol. 24, pp. 115-120, 1944.
[21] 孫宗瀛、楊英魁，「Fuzzy控制理論、實作與應用」，全華科技圖書股份有限公司，2005。
[22] 羅華強，「類神經網路-MATLAB的應用」，高立圖書有限公司，2005。
[23] 楊宗文，「適應性類神經模糊推論系統在舒適度模型上的應用」，碩士論文，國立台灣大學，2001。
[24] J. S. R. Jang, “Input selection for ANFIS learning,” Proceedings of the Fifth IEEE International Conference on, Vol. 2, pp. 1493-1499, 1996.
[25] Suleyman Guven，「Partial discharges pattern classification in GIS using adative neuro fuzzy inference system」，碩士論文，國立台灣科技大學，2012。