刘航斌,林厚飞,褚静,叶静,林权威
LIU Hangbin,LIN Houfei,CHU Jing,YE Jing,LIN Quanwei

• 1:福州大学电气工程与自动化学院
• 2:国网浙江省电力有限公司平阳县供电公司

摘要(Abstract)：

为解决GIS（气体绝缘封闭组合电器）局部放电图谱识别任务中样本数量少且分布不均衡对深度学习模型性能的制约，提出一种CWGAN-div（带条件约束的Wasserstein生成对抗网络）模型以指导多类别局部放电图谱的生成，克服了原始GAN（生成式对抗网络）训练不稳定的问题，使样本数据得到增强并将平均不平衡率由11.01降至3.03。然后，使用5种分类器进行样本增强前后的对比实验，各分类器的F_(1mean)值在样本增强后均得到3.7个百分点以上的提升。实验中，Mi-CNN（多输入卷积神经网络）模型因能够同时利用特高频法和超声波法的PRPD（局部放电相位分布）图谱而表现最优，其F_(1mean)值达到95.8%。
In order to solve the constraints of the limited number and uneven distribution of samples on the performance of the deep learning model in the identification of partial discharge patterns of GIS(gas-insulated switchgear), a CWGAN-div(conditional Wassertein generative adversarial network-divergence) model is proposed to guide the generation of multi-class partial discharge patterns, which overcomes the instability of the original GAN(generative adversarial network) training, enhances the sample data, and reduces the average imbalance rate from 11.01 to 3.03. Then after using five kinds of classifiers for the comparative experiments before and after sample enhancement, the F1mean value of each classifier has been improved by more than 3.7% after sample enhancement. In the experiment, the Mi-CNN(multi-input-convolutional neural networks) model proposed in this paper can use the PRPD(phase resolved partial discharge) spectrum of ultra-high frequency method and ultrasonic method at the same time, and its final F_(1mean) value reaches 95.8%.

关键词(KeyWords)： 气体绝缘封闭组合电器;局部放电;生成式对抗网络;卷积神经网络
GIS(gas-insulated switchgear);partial discharge;GAN(generative adversarial network);convolution neural network

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 刘航斌,林厚飞,褚静,叶静,林权威
LIU Hangbin,LIN Houfei,CHU Jing,YE Jing,LIN Quanwei

DOI: 10.19585/j.zjdl.202308010

参考文献(References)：

• [1]刘洪正.高压组合电器[M].北京：中国电力出版社，2014.
• [2]杨坤.变电设备局部放电带电检测技术研究[D].济南：山东大学，2014.YANG Kun. Research on live detection technology of partial discharge in substation equipment[D]. Jinan:Shandong University,2014.
• [3]王国利，高超，杨芸，等.GIS绝缘缺陷局部放电检测的有效性研究[J].高压电器，2018,54(11):62-72.WANG Guoli,GAO Chao,YANG Yun,et al.Partial discharge detection effectiveness of typical insulation defects in GIS[J].High Voltage Apparatus,2018,54(11):62-72.
• [4]贾嵘，赵佳佳，武桦，等.相关概率小波变换在局部放电检测中的应用[J].高电压技术，2017,43(9):2896-2902.JIA Rong,ZHAO Jiajia,WU Hua,et al.Application of correlated probabilistic wavelet transform in partial discharge detection[J]. High Voltage Engineering,2017,43(9):2896-2902.
• [5]宋辉，代杰杰，李喆，等.运行条件下GIS局部放电严重程度评估方法[J].中国电机工程学报，2019,39(4):1231-1241.SONG Hui,DAI Jiejie,LI Zhe,et al. An assessment method of partial discharge severity for GIS in service[J].Proceedings of the CSEE,2019,39(4):1231-1241.
• [6]王彦博，朱明晓，邵先军，等.气体绝缘组合电器中局部放电特高频信号S参数特性仿真与实验研究[J].高电压技术，2018,44(1):234-240.WANG Yanbo,ZHU Mingxiao,SHAO Xianjun,et al.Simulation and experimental research on S parameter characteristics of partial discharge UHF signals in GIS[J].High Voltage Engineering,2018,44(1):234-240.
• [7] WANG Y X,YAN J,YANG Z,et al. Partial discharge pattern recognition of gas-insulated switchgear via a lightscale convolutional neural network[J].Energies,2019,12(24):4674.
• [8] SONG H,DAI J J,SHENG G H,et al.GIS partial discharge pattern recognition via deep convolutional neural network under complex data source[J]. IEEE Transactions on Dielectrics and Electrical Insulation,2018,25(2):678-685.
• [9] LI G Y,WANG X H,LI X,et al.Partial discharge recognition with a multi-resolution convolutional neural network[J].Sensors,2018,18(10):3512.
• [10]贾勇勇，邓敏，李玉杰，等.基于深度残差网络的GIS局部放电模式识别技术研究[J].高压电器，2018,54(11):123-129.JIA Yongyong,DENG Min,LI Yujie,et al.Research on GIS partial discharge pattern recognition based on deep residual network[J]. High Voltage Apparatus,2018,54(11):123-129.
• [11] NGUYEN M T,NGUYEN V H,YUN S J,et al.Recurrent neural network for partial discharge diagnosis in gasinsulated switchgear[J].Energies,2018,11(5):1202.
• [12]黄建明，李晓明，瞿合祚，等.考虑小波奇异信息与不平衡数据集的输电线路故障识别方法[J].中国电机工程学报，2017,37(11):3099-3107.HUANG Jianming,LI Xiaoming,QU Hezuo,et al.Method for fault type identification of transmission line considering wavelet singular information and unbalanced dataset[J].Proceedings of the CSEE,2017,37(11):3099-3107.
• [13]熊冰妍，王国胤，邓维斌.基于样本权重的不平衡数据欠抽样方法[J].计算机研究与发展，2016,53(11):2613-2622.XIONG Bingyan,WANG Guoyin,DENG Weibin.Undersampling method based on sample weight for imbalanced data[J].Journal of Computer Research and Development,2016,53(11):2613-2622.
• [14]李艳霞，柴毅，胡友强，等.不平衡数据分类方法综述[J].控制与决策，2019,34(4):673-688.LI Yanxia,CHAI Yi,HU Youqiang,et al.Review of imbalanced data classification methods[J].Control and Decision,2019,34(4):673-688.
• [15] GOODFELLOW I J,POUGET-ABADIE J,MIRZA M,et al.Generative adversarial networks[EB/OL].(2014-06-10)[2022-10-03].https：//arxiv.org/abs/1406.2661.
• [16] ARDILA-REY J A,ORTIZ J E,CREIXELL W,et al.Artificial generation of partial discharge sources through an algorithm based on deep convolutional generative adversarial networks[J].IEEE Access,2020,8:24561-24575.
• [17] BERTHELOT D,SCHUMM T,METZ L. BEGAN:boundary equilibrium generative adversarial networks[EB/OL].(2017-03-31)[2022-10-03]. https：//arxiv. org/abs/1703.10717.
• [18]刘建伟，谢浩杰，罗雄麟.生成对抗网络在各领域应用研究进展[J].自动化学报，2020,46(12):2500-2536.LIU Jianwei,XIE Haojie,LUO Xionglin.Research progress on application of generative adversarial networks in various fields[J].Acta Automatica Sinica,2020,46(12):2500-2536.
• [19] WU J Q,HUANG Z W,THOMA J,et al.Wasserstein divergence for GANs[C]//Computer Vision-ECCV 2018.Cham:Springer International Publishing,2018:673-688.
• [20]陈敏，陈隽，刘常颖，等.GIS超声波、超高频局部放电检测方法适用性研究与现场应用[J].高压电器，2015,51(8):186-191.CHEN Min,CHEN Jun,LIU Changying,et al.Applicability research and field application of GIS ultrasonic and UHF partial discharge detection methods[J]. High Voltage Apparatus,2015,51(8):186-191.
• [21] SANDLER M,HOWARD A,ZHU M L,et al.MobileNetV2:inverted residuals and linear bottlenecks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition,June 18-23,2018,Salt Lake City,USA:4510-4520.
• [22]丁登伟，高文胜，刘卫东.采用特高频法的GIS典型缺陷特性分析[J].高电压技术，2011,37(3):706-710.DING Dengwei,GAO Wensheng,LIU Weidong.Analysis on the typical partial discharge using UHF detection method for GIS[J]. High Voltage Engineering,2011,37(3):706-710.
• [23] ZHANG L J,DUAN L X,HONG X C,et al.Imbalanced data enhancement method based on improved DCGAN and its application[J].Journal of Intelligent&Fuzzy Systems,2021,41(2):3485-3498.
• [24]唐佩军.生成对抗网络的可解释性研究[D].成都：电子科技大学，2021.TANG Peijun.Generating a interpretive approach against the network[D]. Chengdu:University of Electronic Science and Technology of China,2021.
• [25]颜贝.基于对抗式网络的图像数据生成技术研究[D].成都：中国科学院大学（中国科学院光电技术研究所），2020.YAN Bei. Research on image data generation technology based on antagonistic network[D].Chengdu:Institute of Optics and Electronics,Chinese Academy of Sciences,2020.
• [26] VERAS R,COLLINS C.Discriminability tests for visualization effectiveness and scalability[J].IEEE Transactions on Visualization and Computer Graphics,2020,26(1):749-758.
• [27]康守强，邹佳悦，王玉静，等.基于无监督特征对齐的变负载下滚动轴承故障诊断方法[J].中国电机工程学报，2020,40(1):274-281.KANG Shouqiang,ZOU Jiayue,WANG Yujing,et al.Fault diagnosis method of a rolling bearing under varying loads based on unsupervised feature alignment[J]. Proceedings of the CSEE,2020,40(1):274-281.
• [28]周志华.机器学习[M].北京：清华大学出版社，2016.