本文針對四具預置瑕疵之單相氣體絕緣開關作升電壓破壞試驗並量測局部放電。於試驗期間以局部放電設備擷取局部放電訊號，將所得訊號經過雜訊抑制、資料轉換、特徵萃取、化簡轉換成104種特徵參量再使用移動平均法後，試圖從特徵趨勢圖中找出共同的絕緣劣化之規則，找出特徵值正放電區域的放電總和-相位分布之相位重心，具有相同的趨勢產生，重心隨著資料筆數的增加有漸漸往第一象限(0度)靠攏的趨勢。
本研究使用k平均法做群集分析，將群集的中心點(轉折點)筆數定義為絕緣劣化中期與末期兩部分，使完成四類瑕疵決策樹模型。為符合未知瑕疵下的氣體絕緣開關之診斷機制，利用瑕疵四之分割點(相位)所對應的轉折點，重新訓練決策樹成一棵完整的模型。最後共得到6種特徵值:相位解析圖譜之分形維度、負放電區域的放電總和-相位分布之標準差、負放電區域的放電總和-高度分布之標準差、負放電區域的放電總和-高度分布之偏態、負放電區域之起始相位、正放電區域的放電總和-高度分布之標準差。本決策樹模型亦可作為診斷在未知瑕疵下氣體絕緣開關絕緣劣化中期與末期參考之準則，將以上特徵值用以訓練決策樹得到的整體測試成功率為95.8%，使完成評估氣體絕緣開關絕緣狀態之研究。

In this thesis, we conducted partial discharge (PD) tests with single gas insulated switchgear (GIS) apparatus including different four prefabricated defects, trying to find out a common insulation deterioration the rules. During the trial in order to capture the partial discharge signals partial discharge device, we can extract 104 characteristic parameters include the resulting signal after noise suppression, data transformation, feature extraction and using moving average method. We find out that the feature of the sum of discharge from positive region- center of phase distribution has the same phenomenon in all the defects. The center towards to the first quadrant (0 degrees) as the data increased and turned again before breakdown.
Then using k-means to cluster data and he cluster's center point (turning point) items defined them as medium or late term of aging that use them to train the decision trees, so that the completion of four flawed decision tree model. To comply with the unknown defect diagnosis mechanism under the gas insulated switchgear, the use of fourth of the split point defects (phase) corresponding to the turning point, re-train a decision tree to tree complete model. Find out that the 6 characteristic parameters: fractal dimension, the sum of discharge from negative region - center of phase distribution, the sum of discharge from negative region - center of height distribution, the sum of discharge from negative height standard distribution, the area of the discharge from negative initial phase, the sum of discharge from positive region center of height distribution to be used. Use the features above we can get the overall success rate of GIS：95.8%. To finish the development for the insulation status assessment of high voltage gas insulated switchgear.

[1]「供電單位變電設備維護手冊」，台灣電力股份有限公司， 修訂六版：中華民國 103 年 6 月。
[2]T. Hoshino, S. Maruyama, K. Nojima, and M. Hanai, “A unique sensitivity verification combined with real-time PD identification method,” IEEE Transmission on Power Delivery,Vol. 20, NO. 3, pp. 1890–1896, Jul. 2005.
[3]T. Hoshino, H. Koyama, S. Maruyama, and M. Hanai, “Comparison of sensitivity between UHF method and IEC 60270 for on-site calibration in various GIS,” IEEE Transmission on Power Delivery, Vol. 21, NO. 4, pp. 1948–1953, Oct. 2006.
[4]吳瑞南、張建國、鍾易樺、曾俊嘉、張智涵、蔡偉聖、林緯中、郭文慶、王柏崴，GIS局放監測系統研製暨時間時間序列法運用於絕緣狀態估之研究，101年度國科會電力學門專題研究計畫成果發表會報告。
[5]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，民國九十三年。
[6]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，民國一零一年。
[7]王柏崴，「應用決策樹評估高電壓電纜接頭絕緣狀態之研究」，碩士論文，國立台灣科技大學，民國一零三年。
[8]V. K. Agarwal and et al.,“ The My steries of Multifactor Ageing,” IEEE Electrical Insulation. Mag., Vol. 11, No. 3, pp. 37~43, May/ June 1995.
[9]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.
[10] M. D. Judd, O. Farish, J. S. Pearson and B. F. Hampton, "Dielectric windows for UHF partial discharge detection, " IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 8, No. 6, pp. 953-958, 2001.
[11]D. Pommerenke, T. Strehl, R. Heinrich, W. Kalkner, F. Schmidt, and W.Weinenberg, “Discrimination between Internal PD and Other Pulses Using Directional Coupling Sensors on HV Cable Systems,” IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 6,NO.6, pp. 814-824, Dec. 1999.
[12]“High-voltage Test Techniques Measurement of Partial Discharge by Electromagnetic and Acoustic methods,” IEC Standard 62478, Ed 1.0 (in preparation).
[13]ANSI/IEEE Std C57.104-1991, IEEE Guide for The Interpretation of Gases Generated in Oil-Immersed Transformers, IEEE Power Engineering Society, 1992.
[14]郭文慶，「用於15kV氣體絕緣開關瑕疵辨識之特徵參量選取研究」，碩士論文，國立台灣科技大學，民國一零二年。
[15]Y. H. Lin, R. N. Wu, and I. H. Chung, “Novel trend of "l" shape in PD pattern to judge the appropriate crucial moment of replacing cast-resin current transformer,”IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 15,No. 1, pp. 292-301,2008.
[16]X. Zhongrong, T. Ju, and S. Caixin, “Application of complex wavelet transform to suppress white noise in GIS UHF PD signals,” IEEE Transactions on Power Delivery, Vol. 22,No.3,pp. 1498-1504, 2007.
[17]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，民國九十三年。
[18]E. Gulski and F. H. Kreuger,”Computer-aided recognition of discharge sources,” IEEE Transactions on Electrical Insulation, Vol. 27, No. 1, pp. 82-92, 1992.
[19]鍾易樺，「應用適應性類神經網模糊推論系統於氣體絕緣開關瑕疵辨識之研究」，博士論文，國立台灣科技大學，民國一零二年。
[20]N. C. Sahoo, M. M. A. Salama, and R. Bartnikas, "Trends in partial discharge pattern classification: a survey," IEEE Transactions on Dielectrics and Electrical Insulation, Vol. 12, No.2, pp. 248-264, 2005.
[21]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.
[22]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,No. 1, pp. 115-120, 1944.
[23]K. Falconer, Fractalgeometry: mathematical foundations and application, Wiley, 1990.
[24]Wensheng Gao, Dengwei Ding, and Weidong Liu, “Research on the Typical Partial Discharge Using the UHF Detection Method for GIS,”IEEE Transactions on power delivery, Vol.26, No.4, October 2011.
[25]施雅月、賴錦慧，資料探勘，新北市：培生教育出版社，2008
[26]魏志平、鄭滄祥，資料探勘 (原著Introduction to business data mining)，麥格羅．希爾(Mc Graw Hill)，台北市，2008。
[27]Hirose, H.; Hikita, M.; Ohtsuka, S.; Tsuru, S.-i.; Ichimaru, J., "Diagnosis of Electric Power Apparatus using the Decision Tree Method," Dielectrics and Electrical Insulation,” IEEE Transactions on Dielectrics and Electrical Insulation, Vol.15, NO.5, pp.1252-1260, October 2008.
[28]T. K. Abdel-Galil, R. M. Sharkawy, M. M. A. Salama and R. Bartnikas, “Partial discharge pattern classification using the fuzzy decision tree approach,”IEEE Transactions on Instrument and Measurement, Vol. 54, No. 6, pp. 2258–2263, Dec. 2005.
[29]曾憲雄、蔡秀滿、蘇東興、曾秋蓉、王慶堯，資料探勘Data mining，旗標出版股份有限公司，台北市，2005。
[30]王派洲譯，資料探勘:概念與方法(第二版)，滄海書局，2008
[31]C. E. Shannon, “A mathematical theory of communication,” Bell System Technical Journal 27, 379-423, 1948.