基于SDAE-SVM的高压电缆局部放电类型识别

杨帆; 程琛; 黄乐; 彭小圣

doi:10.13296/j.1001-1609.hva.2026.06.011

您当前的位置：

首页 >

文章列表页 >

基于SDAE-SVM的高压电缆局部放电类型识别

研究与分析 | 更新时间：2026-06-10

- 基于SDAE-SVM的高压电缆局部放电类型识别
- Partial Discharge Pattern Recognition of High Voltage Cables Based on SDAE-SVM
- 高压电器 2026年62卷第6期页码：90-96
- 作者机构：
  
  国网陕西省电力有限公司西安供电公司，西安 710032
  华中科技大学电气与电子工程学院强电磁工程与新技术国家重点实验室，武汉 430074
- 作者简介：
  
  杨帆（1995—），男，硕士，工程师，主要研究方向为电力设备状态监测与新能源功率预测（E-mail：2632460413@qq.com）。
  黄乐（1986—），男，本科，工程师，主要研究方向为电力设备状态监测（通信作者）（E-mail：3230540009@qq.com）。
- 基金信息：
  
  国家自然科学基金资助项目;National Natural Science Foundation of China(51541705)
- DOI：10.13296/j.1001-1609.hva.2026.06.011
  中图分类号：
- 收稿：2025-11-07，
  
  修回：2026-01-25，
  
  纸质出版：2026-06-16
- 稿件说明：
移动端阅览
杨帆, 程琛, 黄乐, 等. 基于SDAE-SVM的高压电缆局部放电类型识别[J]. 高压电器, 2026,62(6):90-96.

YANG Fan, CHENG Chen, HUANG Le, et al. Partial Discharge Pattern Recognition of High Voltage Cables Based on SDAE-SVM[J]. High Voltage Apparatus, 2026, 62(6): 90-96.
杨帆, 程琛, 黄乐, 等. 基于SDAE-SVM的高压电缆局部放电类型识别[J]. 高压电器, 2026,62(6):90-96. DOI： 10.13296/j.1001-1609.hva.2026.06.011.

YANG Fan, CHENG Chen, HUANG Le, et al. Partial Discharge Pattern Recognition of High Voltage Cables Based on SDAE-SVM[J]. High Voltage Apparatus, 2026, 62(6): 90-96. DOI： 10.13296/j.1001-1609.hva.2026.06.011.

摘要

提出一种基于改进堆栈去噪自编码器（SDAE-SVM）的深度学习方法，用于高压电缆不同绝缘缺陷局部放电（PD）信号的模式识别。首先在高压实验室中对5种类型的人工缺陷进行PD测试，并提取3 500组PD瞬时脉冲，构建了34种特征参数。其次，详细介绍了SDAE-SVM的原理和网络架构。然后，使用所提模型识别不同缺陷类型的PD信号，获得了93.56%的识别精度。接着，使用t分布随机邻接嵌入（t-SNE）对SDAE-SVM逐层输出进行了可视化，说明了深度神经网络SDAE-SVM逐层优化的本质。最后，将所提方法与反向传播神经网络（BPNN）、支持向量机（SVM）和堆栈去噪自编码器（SDAE）进行了对比。结果表明，相比BPNN、SVM和SDAE、SDAE-SVM的总体识别精度分别提高了7.46%、6.70%、1.37%，具备较高的工程应用价值。

Abstract

A deep learning method based on an improved stacked denoising autoencoder (SDAE-SVM) is proposed for pattern recognization of partial discharge (PD) signals generated by different insulation defects in high-voltage cables. First

PD tests are conducted on five types of artificial defects in a high-voltage laboratory

and 3500 sets of PD instantaneous pulses are extracted to construct 34 types of characteristic parameters. Then

the principles and network architecture ofSDAE-SVMare introduced in detail. After that

the proposed model is used to recognize the PD signals of different types of defects and the pattern recognition accuracy of 93.56% is obtained. Moreover

the the layer-wise outputs of theSDAE-SVMare visualized using t-distributed stochastic neighbor embedding (t-SNE)

illustrating the essence of layer-wise optimization of the deep neural networkSDAE-SVM. Finally

the proposed method is compared with back propagation neural network (BPNN)

support vector machine (SVM) and stacked denoising autoencoders (SDAE) . The results show that compared with BPNN

SVM

and SDAE

the overall recognition accuracy ofSDAE-SVMhas increased by 7.46%

6.70%

and 1.37%

respectively

demonstraing high engineering application value .

关键词

Keywords

references

周彦,常俊,曹妤婕,等.高压开关柜局部放电多物理信号特性对比研究[J].高压电器,2023,59(8):196-202.

ZHOU Yan, CHANG Jun, CAO Yujie, et al. Comparative study on multi-physical signal characteristics of partial discharge in high voltage switchgear cabinet[J]. High Voltage Apparatus,2023,59 (8):196-202.

祝学海.在线局部放电检测技术在工厂配电设备运行中的应用[J].自动化应用,2023,64(13):153-156.

ZHU Xuehai. Application of online partial discharge detection technology in the operation of power distribution equipment in factories[J]. Automation Application,2023,64(13):153-156.

黄辉,杨智豪,魏建国,等.变压器套管局部放电射频信号在外部空间的分布特征[J].中国电力,2023,56(4):175-183.

HUANG Hui, YANG Zhihao, WEI Jianguo, et al. Distribution char acteristics of partial discharge radio frequency signal in transformer tank and bushing[J]. Electric Power,2023,56(4):175-183.

PENG Xiaosheng, YANG Fan, WANG Ganjun, et al. A convolutional neural network-based deep learning methodology for recognition of partial discharge patterns from high-voltage cables[J]. IEEE Transactions on Power Delivery,2019,34(4):1460-1469.

任保瑞,郑德芳,关鹤,等.高压电力电缆行波传播特性研究[J].西安工程大学学报,2025,39(4):63-72.

REN Baorui, ZHENG Defang, GUAN He, et al.Characterization of traveling wave propagation in high voltage power cables[J]. Journal of Xi'an Polytechnic University,2025,39(4):63-72.

严亚兵,江志文,肖俊先,等.基于高频电流信号的电缆局部放电故障特征分析研究[J].供用电,2024,41(3):96-102.

YAN Yabing, JIANG Zhiwen, XIAO Junxian, et al.Fault characteristics analysis of cable partial discharge based on the highfrequency current signal[J]. Distribution & Utilization,2024, 41(3):96-102.

刘浩,侯春光,高有华.基于多传感信息融合的电缆附件局部放电诊断算法研究[J].电器与能效管理技术,2024(10):36-41.

LIU Hao, HOU Chunguang, GAO Youhua.Research on partial discharge diagnosis algorithm for cable accessories based on multi sensor information fusion[J]. Electrical & Energy Management Technology,2024(10):36-41.

王建伟,郑祥.基于高压电缆局部放电TDR法的便携式转发器研究[J].电器与能效管理技术,2024(7):36-41.

WANG Jianwei, ZHENG Xiang.Research on portable transponder based on partial discharge TDR method of high voltage cable[J]. Electrical & Energy Management Technology,2024(7):36-41.

陈峰,姜伊欣,娄雨靖.基于小波包分解和支持向量机的局部放电识别方法研究[J].山东电力技术,2020,47(6):5-9.

CHEN Feng, JIANG Yixin, LOU Yujing. Partial discharge classification method based on wavelet packet decomposition and support vector machines[J]. Shandong Electric Power,2020,47 (6):5-9.

LI Ping, TANG Ju, LIU Yilu. Partial discharge recognition in gas insulated switchgear based on multi-information fusion[J]. IEEE Transactions on Dielectrics and Electrical Insulation,2015,22(2):1080-1087.

郑业爽,李世春,鲁玲.基于多策略ISOA优化SVM的变压器故障诊断研究[J].智慧电力,2023,51(2):38-44.

ZHENG Yeshuang, LI Shichun, LU Ling. Transformer fault diagnosis based on multi-strategy ISOA optimized SVM[J]. Smart Power,2023,51(2):38-44.

吉兴全,陈金硕,张玉敏,等.基于CNN-SVM的配电网故障分类研究[J].智慧电力,2022,50(1):94-100.

JI Xingquan, CHEN Jinshuo, ZHANG Yumin, et al. Fault classification in distribution network based onCNN - SVM[J]. Smart Power,2022,50(1):94-100.

刘文浩,吴毅江,李文泽,等.基于随机决策森林的高压电缆局部放电模式识别[J].高压电器,2022,58(6):165-170.

LIU Wenhao, WU Yijiang, LI Wenze, et al. Partial discharge pattern recognition of high-voltage cables based on random forest method[J]. High Voltage Apparatus,2022,58(6):165-170.

杨帆,王干军,彭小圣,等.基于卷积神经网络的高压电缆局部放电模式识别[J].电力自动化设备,2018,38(5):123-128.

YANG Fan, WANG Ganjun, PENG Xiaosheng, et al. Partial discharge pattern recognition of high-voltage cables based on convolutional neural network[J]. Electric Power Automation Equipment,2018,38(5):123-128.

黄光磊,李喆,许永鹏,等.基于改进深度信念网络的直流XLPE电缆局部放电模式识别[J].高电压技术,2020,46(1):327-334.

HUANG Guanglei, LI Zhe, XU Yongpeng, et al. Partial discharge pattern recognition of XLPE DC cable based on improved deep belief networks[J]. High Voltage Engineering,2020,46(1):327-334.

韩雪源.基于LSTM算法的高压交联电缆线路振荡波局部放电检测方法[J].电机与控制应用,2022,49(12):41-46.

HAN Xueyuan. Oscillating wave partial discharge detection method of high voltage cross-linked cable line based on LSTM algorithm[J]. Electric Machines & Control Application,2022,49 (12):41-46.

张怡,张恒旭,李常刚,等.深度学习在电力系统频率分析与控制中的应用综述[J].中国电机工程学,2021,41(10):3392-3406.

ZHANG Yi, ZHANG Hengxu, LI Changgang, et al. Review on deep learning applications in power system frequency analysis and control[J]. Proceedings of the CSEE,2021,41(10):3392-3406.

杨帆.基于深度学习的短期风电功率预测技术研究[D].武汉:华中科技大学,2020.

YANG Fan. Research on short-term wind power prediction based on deep learning methodology[D]. Wuhan:Huazhong University of Science and Technology,2020.

PENG Xiaosheng , WEN Jinyu , LI Zhaohui , et al. SDMF based interference rejection and PD interpretation for simulated defects in HV cable diagnostics[J]. IEEE Transactions on Dielectrics and Electrical Insulation,2017,24(1):83-91.

张金水,蒋伟,潘伟杰.基于栈式降噪自编码器的GIS绝缘缺陷识别研究[J].电气自动化,2021,43(4):81-83.

ZHANG Jinshui, JIANG Wei, PAN Weijie. GIS insulation defect recognition based on stacked denoising autoencoder[J]. Electrical Automation,2021,43(4):81-83.

马良玉,孙佳明.基于SDAE预测模型和改进SSA的NOx排放优化[J].中国电机工程学报,2022,42(14):5194-5201.

MA Liangyu, SUN Jiaming. NOx emission optimization based on SDAE prediction model and improved SSA[J]. Proceedings of the CSEE,2022,42(14):5194-5201.

贾庆超.基于稀疏表示和深度学习分类器的SAR目标图像的分割与分类[D].西安:西安电子科技大学,2017.

JIA Qingchao. SAR target image segmentation and classification based on sparse representation and deep learning[D].Xi'an:Xidian University,2017.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于卷积神经网络的变压器局部放电模式识别

基于SURF-BoVF特征与主动学习EQB-LightGBM的发电机局部放电模式识别方法

基于距离的放电模式识别方法

局部放电识别中分维数分布的研究

电力设备局部放电模式识别中分形理论的应用