隨著國內電力需求增加，以及近年再生能源的推動發展，電力系統的強健性將成為重要課題，而預防性的電力設備故障診斷便是其中重要一環。局部放電的檢測和圗譜辨識在近年已有不少發展，然而在局部放電類型辨識上，普遍以特徵提取的方式，對圗譜進行預處理，為解決樣本數不足的問題。
本文主要提出應用於高壓設備故障診斷之局部放電相位分析圖譜(PRPD 圖譜)作為輸入特徵，使用卷積神經網路辨識局部放電類型。並同時使用相同輸入特徵，即局部放電相位分析圖譜(PRPD 圖譜)，通過利用生成對抗網路輸出圖譜，增加卷積神經網路的輸入特徵樣本數，以加強卷積神經網路辨識局部放電類型的可靠度。
　　結果指出，在使用生成對抗網路生成圖譜，從而增加輸入特徵樣本數的方法下，卷積神經網路在辨識局部放電類型的表現上有所提升，得見方法的可行性，可作為日後相關研究提供一個具有價值的參考依據。

With the increase in domestic power demand and the development of renewable energy in recent years, the robustness of the power system will become an important issue, and preventive power equipment fault diagnosis is an important part of it. In recent years, there have been many developments in the detection and identification of partial discharges. However, in the identification of partial discharge types, feature extraction is generally used to preprocess the data to solve the problem of insufficient samples.
This paper mainly proposes the phase resolved partial discharge analysis patterns (PRPD patterns) applied to the fault diagnosis of high-voltage equipment as the input feature, and uses the convolutional neural network to identify the partial discharge type. And use the same input feature at the same time, that is, the phase resolved partial discharge analysis patterns (PRPD patterns), and increase the number of input feature samples of the convolutional neural network by using the generative adversarial network output map to strengthen the convolutional neural network to identify the partial discharge type. reliability.
The results pointed out that under the method of using the generative adversarial network to generate the map, thereby increasing the number of input feature samples, the performance of the convolutional neural network in identifying the type of partial discharge has been improved. Research provides a valuable reference basis.

[1] Y. Lecun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard and L. D.Jackel: Backpropagation Applied to Handwritten Zip Code Recognition, Neural Computation, 1(4):541-551, Winter 1989.
[2] Y. Lecun, L. Bottou, Y. Bengio and P. Haffner, "Gradient-based learning applied to document recognition," in Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, Nov.1998, doi: 10.1109/5.726791.
[3] Krizhevsky, Alex; Sutskever, Ilya; Hinton, Geoffrey E. "ImageNet classification with deep convolutional neural networks" (PDF). Communications of the ACM. 60 (6): 84–90. doi:10.1145/3065386. ISSN 0001-0782. S2CID 19590877, May 24 2017.
[4] M. D. Zeiler, and R. Fergus. "Visualizing and understanding convolutional networks." European conference on computer vision. Springer, Cham, 2014.
[5] K. Simonyan, and A. Zisserman. “Very deep convolutional networks for large-scale image recognition”. arXiv preprint arXiv:1409.1556, 2014.
[6] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, A. Rabinovich, Going deeper with convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1-9), 2015.
[7] K. He, X. Zhang, S. Ren, and J. Sun,” Deep residual learning for image recognition”. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778), 2016.

[8] F. N. Iandola, S. Han, M. W. Moskewicz, K. Ashraf, W. J. Dally, and K. Keutzer, (2016).SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and < 0.5 MB model size. arXiv preprint arXiv:1602.07360.
[9] C. -K. Chang, H. -H. Chang and B. K. Boyanapalli, "Application of Pulse Sequence Partial Discharge Based Convolutional Neural Network in Pattern Recognition for Underground Cable Joints," in IEEE Transactions on Dielectrics and Electrical Insulation, vol. 29, no. 3, pp. 1070-1078, doi: 10.1109/TDEI.2022.3168328, June 2022.
[10] Y. Wang, J. Yan, Q. Sun, J. Li and Z. Yang, "A MobileNets Convolutional Neural Network for GIS Partial Discharge Pattern Recognition in the Ubiquitous Power Internet of Things Context: Optimization, Comparison, and Application," in IEEE Access, vol. 7, pp. 150226-150236, doi: 10.1109/ACCESS.2019.2946662, 2019.
[11] X. Peng et al., "A Convolutional Neural Network-Based Deep Learning Methodology for Recognition of Partial Discharge Patterns from High-Voltage Cables," in IEEE Transactions on Power Delivery, vol. 34, no. 4, pp. 1460-1469, doi: 10.1109/TPWRD.2019.2906086, Aug. 2019.
[12] V. -N. Tuyet-Doan, H. -A. Pho, B. Lee and Y. -H. Kim, "Deep Ensemble Model for Unknown Partial Discharge Diagnosis in Gas-Insulated Switchgears Using Convolutional Neural Networks," in IEEE Access, vol. 9, pp. 80524-80534, doi: 10.1109/ACCESS.2021.3084950, 2021.
[13] T. -D. Do, V. -N. Tuyet-Doan, Y. -S. Cho, J. -H. Sun and Y. -H. Kim, "Convolutional-Neural-Network-Based Partial Discharge Diagnosis for Power Transformer Using UHF Sensor," in IEEE Access, vol. 8, pp. 207377-207388, doi:10.1109/ACCESS.2020.3038386, 2020

[14] Goodfellow, Ian J.; Pouget-Abadie, Jean; Mirza, Mehdi; Xu, Bing; Warde-Farley, David; Ozair, Sherjil; Courville, Aaron; Bengio, Yoshua. Generative Adversarial Networks. arXiv:1406.2661, 2014.
[15] J. A. Ardila-Rey, J. E. Ortiz, W. Creixell, F. Muhammad-Sukki and N. A. Bani, "Artificial Generation of Partial Discharge Sources Through an Algorithm Based on Deep Convolutional Generative Adversarial Networks," in IEEE Access, vol. 8, pp. 24561-24575, doi: 10.1109/ACCESS.2020.2971319, 2020.
[16] “High voltage test techniques – Partial discharge measurement,”
IEC 60270, 2001
[17] Zhicheng Wu,“Source database for Machine-learning-based PD Pattern Recognition in GIS”, Mendeley Data, V1, doi: 10.17632/cz8gwg9d2v.1, 2020
[18] Yeh James,“[資料分析&機器學習] 第5.1講: 卷積神經網絡介紹(Convolutional Neural Network) ”, Dec 25, 2017