隨著台灣高科技產業逐年的發展，穩定的電力供給是維持廠區運行的一大要求，其中地下電纜在台灣電網的輸配電系統中扮演重要的角色，檢測高壓系統故障必然是重要的方向，能即早維護或更換，而局部放電(Partial Discharge, PD)是國際專家認為具有潛力發展，也是系統設備故障的指標現象之一。
本研究採用深度學習方法中的卷積神經網路(Convolutional Neural Networks, CNN)建立局部放電診斷程式，利用圖像人工智慧去分析局部放電訊號，除了辨識局部放電類型外，最大的特點是利用CNN方法判定破壞路徑的結果後，額外進行兩種結果分支的劣化模型訓練，利用此特點來進而提高絕緣劣化狀態模型準確率。
本文針對地下電纜直線接頭製作24個實驗樣本，其中22組訓練資料及2組測試資料，分為瑕疵A(絕緣層有空隙)、瑕疵B(絕緣層有空洞)與瑕疵C(外半導層有尖端)。應用卷積神經網路分析三種瑕疵絕緣劣化的數據，以離線訓練程式之模型能即時診斷電纜狀態，主要功能為評估電纜劣化狀態，以瑕疵辨識、破壞路徑識別為輔助，可以向現場做人員提供簡易且明確的指示。結果顯示皆可事先在被試物絕緣破壞前進行危險預測，且警告的時段有充裕的時間給予工作人員做電纜維護或更換。

The stable power supply is a significant requirement for maintaining the operation of the factory with the development of Taiwan’s high-tech industries year by year. Among them, underground cables play an important role in the transmission and distribution system of Taiwan’s power grid. The detection of high-voltage system failures must be an important direction, which can be maintained or replaced as soon as possible. Partial Discharge (PD) is considered by international experts to have the potential for development and is also one of the indicator phenomena of system equipment failures.
The study uses Convolutional Neural Networks (CNN) in Deep Learning to establish a diagnostic program for partial discharge and uses image artificial intelligence to analyze partial discharge signals. In addition to recognizing the type of partial discharge, the most significant characteristic is that after used the CNN method to recognize the result of the damage track, the additional deterioration model training of two branches of the result is performed. And this characteristic is used to further improve the accuracy of the insulation deterioration state model.
This paper made 24 experimental samples for underground cable joints, including 22 training data and 2 test data, which are divided into defect A (void in the insulation layer), defect B (void in the insulation layer), defect C (tip in the outer semi-conductive layer). Convolutional Neural Network is used to analyze the data of insulation deterioration of three types of defects. The model of the off-line training program can diagnose the cable status in real-time. The main function is to evaluate the cable deterioration status, and the defect recognition and damage track recognition are auxiliary functions, which can be provided to the field staff with simple and clear instructions. The results show that this program can predict the danger before the insulation of the test object is damaged, and there is ample time for the staff to do cable maintenance or replacement during the warning period.

[1] 行政院，「815停電事故行政調查專案報告」，民國一零六年。
[2] 經濟部，「108年度物資經濟動員準備演習綜合檢討報告」，民國一零八年。
[3] B. Kruizinga, P. A. A. F. Wouters and E. F. Steennis, “Fault Development upon Water Ingress in Damaged Low Voltage Underground Power Cables with Polymer Insulation” IEEE Transactions on Dielectrics and Electrical Insulation, vol.9, pp. 808-816, April 2017.
[4] 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，民國一百年。
[5] F. P. Mohamed, W. H. Siew, J. J. Soraghan, S. M. Strachan and J. McWilliam, “Partial Discharge Location in Power Cables Using a Double Ended Method Based on Time Triggering with GPS” IEEE Transactions on Dielectrics and Electrical Insulation, vol.10, pp.2212-2221, December 2013.
[6] Shady S. Refaat and Mohammed A. Shams, “A Review of Partial Discharge Detection, Diagnosis Techniques in High Voltage Power Cables” 2018 IEEE 12th International Conference on Compatibility, Power Electronics and Power Engineering, pp. 1-5, Doha, Qatar, April 2018.
[7] P. Wagenaars, P. A.A.F. Wouters, P. C.J.M. van der Wielen, E. F. Steennis, “Accurate Estimation of the Time-of-arrival of Partial Discharge Pulses in Cable Systems in Service” IEEE Transactions on Dielectrics and Electrical Insulation, vol.10, pp.1190-1199, August 2008.
[8] W. J. K. Raymond, H. A. Illias and A. H. Abu Bakar, "High Noise Tolerance Feature Extraction for Partial Discharge Classification in XLPE Cable Joints," IEEE Transactions on Dielectrics and Electrical Insulation, vol. 24, no. 1, pp. 66-74, February 2017.
[9] Xuefeng Zhao, Lu Pu ; Long Xu, Xing Zhang, Baohong Guo, Aixuan Zhao, Junbo Deng, Guanjun Zhang, Ni Hui and Minghao Fan, “Research on the Defect Development of Cable” International Conference on Electrical Materials and Power Equipment, vol.6, pp457-462, July 2017.
[10] X. Peng et al., "A Convolutional Neural Network-Based Deep Learning Methodology for Recognition of Partial Discharge Patterns from High-Voltage Cables," IEEE Transactions on Power Delivery, vol. 34, no. 4, pp. 1460-1469, August 2019.
[11] 謝沅澔，「電力電纜接頭絕緣狀態智能化診斷規則之研究」，國立台灣科技大學，民國一百零七年。
[12] 劉鑑，「地下電纜接頭絕緣狀態轉變之研究」，碩士論文，國立台灣科技大學，民國一百零八年。
[13] 戴嘉宏，「支持向量機應用於電力電纜接頭絕緣狀態評估之研究」，碩士論文，國立台灣科技大學，民國一百零五年。
[14] 張加明，「電力電纜接頭瑕疵演進之絕緣狀態警示研究」，碩士論文，國立台灣科技大學，民國一百零五年。
[15] 雷宗憲，「暴力演算法應用於地下電纜接頭絕緣狀態評估規則之研究」，國立台灣科技大學，民國一百零六年。
[16] 張瑞村，「數位局部放電量測應用於高壓電纜終端接頭絕緣狀態之評估」，碩士論文，國立台灣科技大學，民國九十四年。
[17] 陳毅哲，「運用支持向量機理論評估電力電纜絕緣劣化狀態」，碩士論文，國立台灣科技大學，民國一百零八年。
[18] A. Eigner and K. Rethmeier, "An Overview on the Current Status of Partial Discharge Measurements on AC High Voltage Cable Accessories," IEEE Electrical Insulation Magazine, vol. 32, no. 2, pp. 48-55, Mar.-April 2016.
[19] “High-Voltage Test Techniques-Partial Discharge Measurements,” IEC60270, 2000.
[20] M. Wu, H. Cao, J. Cao, H. Nguyen, J. B. Gomes and S. P. Krishnaswamy, "An Overview of State-Of-The-Art Partial Discharge Analysis Techniques for Condition Monitoring," IEEE Electrical Insulation Magazine, vol. 31, no. 6, pp. 22-35, November-December 2015.
[21] N. H. Aziz, V. M. Catterson, S. M. Rowland and S. Bahadoorsingh, "Analysis of Partial Discharge Features as Prognostic Indicators of Electrical Treeing," IEEE Transactions on Dielectrics and Electrical Insulation, vol. 24, no. 1, pp. 129-136, February 2017.
[22] Y. Xu, G. Sheng, F. Yang, X. Chen, Y. Qian and X. Jiang, "Classification of Partial Discharge Images within DC XLPE Cables in Contourlet Domain," IEEE Transactions on Dielectrics and Electrical Insulation, vol. 25, no. 2, pp. 486-493, April 2018.
[23] Y. Xia, X. Song, J. He, Z. Jia and X. Wang, "Simulation and Partial Discharge Detection for Typical Defects of 10kV Cable the Joint," The Journal of Engineering, vol. 2019, no. 16, pp. 2856-2859, 3 2019.
[24] J. Wang, P. E. C. Stone, D. Coats, Y. Shin and R. A. Dougal, "Health Monitoring of Power Cable via Joint Time-Frequency Domain Reflectometry," IEEE Transactions on Instrumentation and Measurement, vol. 60, no. 3, pp. 1047-1053, March 2011.
[25] 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。
[26] C. Mazzetti, F.M.F. Mascioli, F. Baldini, M. Panella, R. Risica and R. Bartnikas, “Partial Discharge Pattern Recognition by Neuro-Fuzzy Networks in Heat-Shrinkable Joints and Terminations of XLPE Insulated Distribution Cables,” IEEE Transactions on Power Delivery, vol.10, pp. 1035-1044, June 2006.
[27] Xiaosheng Peng , Chengke Zhou, Donald M. Hepburn, Martin D. Judd and W. H. Siew,“Application of K-Means Method to Pattern Recognition in On-Line Cable Partial Discharge Monitoring,” IEEE Dielectrics and Electrical Insulation Society, vol.20, pp.754-761, June 2013.
[28] Zhuo Ma, Yang Yang, Martin Kearns, Kevin Cowan, Huajie Yi, Donald M. Hepburn and Chengke Zhou, “Fractal-based Autonomous Partial Discharge Pattern Recognition Method for MV Motors,” High Voltage, vol.11, pp. 103-114, June 2018.
[29] 台灣電力公司業務處，「配電技術手冊(四)地下配電線路設計」，民國八十五年。
[30] 台灣電力公司，「高壓單芯交連PE電纜」(A043)，材料規範，民國一百零二年。
[31] Dongxin He, Xiaoran Wang, Hongshun Liu, Qingquan Li and Gilbert Teyssèdre, “Space Charge Behavior in XLPE Cable Insulation Under AC Stress and its Relation to Thermo-Electrical Aging,” IEEE Transactions on Dielectrics and Electrical Insulation, vol.8, pp. 541-548, April 2018.
[32] K. X. Lai, A. Lohrasby, B.T. Phung and T.R. Blackburnand, “Partial Discharge Characteristics of Electrical Trees prior to Breakdown,” International Symposium on Electrical Insulating Materials, vol.3,pp.649-652, September 2008.
[33] F. Liu, C. Shen, G. Lin, and I. Reid, “Learning Depth from Single monocular images Using Deep Convolutional Neural Fields,” IEEE Transactions on Pattern Analysis and Machine Intelligence., vol. 38, no. 10, pp. 2024–2039, October 2016.
[34] Y. P. Lee, “Multidimensional Hebbian Learning with Temporal Coding in Neocognitron Visual Recognition,” IEEE Transactions on Systems, Man, and Cybernetics: Systems vol. 47, no. 12, pp. 3386–3396, December 2017.
[35] C. Dong, C. C. Loy, K. He, and X. Tang, “Image Super-resolution Using Deep Convolutional Networks,” IEEE Transactions on Pattern Analysis and Machine Intelligence vol. 38, no. 2, pp. 295–307, February 2016.
[36] S. Wang, Y. Jiang, X. Hou, H. Cheng, and S. Du, “Cerebral Micro-Bleed Detection Based on The Convolution Neural Network with Rank Based Average Pooling,” IEEE Access, vol. 5, pp. 16576–16583, August 2017
[37] G. Liang, H. Hong , W. Xie , and L. Zheng, ” Combining Convolutional Neural Network With Recursive Neural Network for Blood Cell Image Classification”, IEEE Access, Vol.6, pp. 36188 – 36197, 03 July 2018.
[38] D. Liu et al., “Learning Temporal Dynamics for Video Super-resolution: A deep Learning Approach,” IEEE Transactions on Image Processing., vol. 27, no. 7, pp. 3432–3435, July 2018
[39] A. Kappeler , S. Yoo, Q. Dai, and A. K. Katsaggelos , ”Video Super-Resolution With Convolutional Neural Networks”, IEEE Transactions on Computational Imaging, Vol.2 , pp.109 - 122, June 2016
[40] 李卓翰，「地下電纜接頭局部放電之差動量測研究」，碩士論文，國立台灣科技大學，民國一百零九年。