Fault Diagnosis for Distribution Networks Based on Transfer Learning and Residual Network

ZHOU Tingdong; SHANG Hai; HE Yongtao; LI Panpan; GAN Ping

doi:10.13296/j.1001-1609.hva.2026.04.024

您当前的位置：

首页 >

文章列表页 >

Fault Diagnosis for Distribution Networks Based on Transfer Learning and Residual Network

Research & Analysis | 更新时间：2026-04-08

- Fault Diagnosis for Distribution Networks Based on Transfer Learning and Residual Network
- High Voltage Apparatus Vol. 62, Issue 4, Pages: 208-214(2026)
- 作者机构：
  
  云南电网有限责任公司红河供电局，云南蒙自 661100
  云南电网有限责任公司，昆明 650000
  合肥溢鑫电力科技有限公司，合肥 230000
- 作者简介：
- 基金信息：
- DOI：10.13296/j.1001-1609.hva.2026.04.024
  CLC：
- Received：17 September 2025，
  
  Revised：2025-11-25，
  
  Published：16 April 2026
- 稿件说明：
移动端阅览
ZHOU Tingdong, SHANG Hai, HE Yongtao, et al. Fault Diagnosis for Distribution Networks Based on Transfer Learning and Residual Network[J]. High Voltage Apparatus, 2026, 62(4): 208-214.
DOI：

ZHOU Tingdong, SHANG Hai, HE Yongtao, et al. Fault Diagnosis for Distribution Networks Based on Transfer Learning and Residual Network[J]. High Voltage Apparatus, 2026, 62(4): 208-214. DOI： 10.13296/j.1001-1609.hva.2026.04.024.

摘要

配电网单相接地故障占比较高，存在多种故障类型。如果对不同类型的接地故障进行适当的处理，能有效提高配电网的可靠性和可用性。然而，如何快速区分不同的故障类型，一直以来是一个研究热点。本研究针对高阻接地故障电流和电压特征不明显导致的高阻接地故障检测困难的问题，提出了一种创新的解决方法，即采用迁移学习和深度残差网络对高阻接地故障进行识别。首先，提取单相接地故障后2个周期内的零序电压数据，利用小波变换将零序电压映射为二维时频图；其次，搭建深度残差网络，通过冻结底层参数实现小样本迁移，微调参数对不同的故障样本分类。为了验证算法有效性，搭建了10 kV配电网模型进行仿真。经实验证明，采用该方法可有效检测弧光高阻接地故障信号，并准确识别出这类接地故障。

Abstract

The single-phase grounding faults in the distribution network account for a relatively high proportion

and there are various types of faults. If different types of grounding faults are handled appropriately

both reliability and availability of the distribution network can be effectively improved. However

how to quickly distinguish different types of faults has always been a research hotspot. In this study

an innovative solution is proposed to solve the problem of high resistance grounding fault detection difficulties caused by the unobvious characteristics of high resistance grounding fault current and voltage

that is

the high resistance grounding fault is identified by using transfer learning and deep residual network. First

the zero-sequence voltage data in 2 cycles after the single-phase ground fault is extracted

and the zero-sequence voltage is mapped into the two-dimensional time-frequency map. Then

a deep residual network is built

the small sample migration is achieved by freezing underlying parameters

and different fault samples is classified by fine-tuning parameters.To verify the effectiveness of the algorithm

a 10 kV distribution network simulation model is set up. It is proved by experiment that application of this method can effectively detect arc high resistance grounding fault signals and accurately identify such grounding faults.

关键词

Keywords

references

田君杨,李海勇,蒋连钿,等.含多分布式电源的小电阻接地系统单相接地故障特征分析[J].电力系统保护与控制,2024,52(14):103-110.

TIAN Junyang,LI Haiyong,JIANG Liandian,et al.Analysis of single - phase grounding fault characteristics of low resistance grounding systemwith multi-distributed power supply[J]. Power System Protection and Control,2024,52(14):103-110.

赵福平,范松海,尚豪,等.配电网单相接地故障无源柔性消弧策略[J].电测与仪表,2025,62(8):93-99.

ZHAO Fuping,FAN Songhai,SHANG Hao,et al.Passive flexible arc suppression strategy for single phase grounding faultin distribution network[J]. Electrical Measurement & Instrumentation, 2025,62(8):93-99.

廖德辉,金鑫,田君杨,等.灵活接地配电网单相接地故障全过程特征分析[J].智慧电力,2024,52(5):90-97.

LIAO Dehui,JIN Xin,TIAN Junyang,et al.Analysis of the entire process characteristics of single phase-to-ground faults in flexible grounded distribution networks[J]. Smart Power ,2024 ,52(5):90-97.

王笑然,范春菊,胡炎,等.含有储能电站的主动配电网单相接地故障选线方法[J].电力科学与技术学报,2025,40(5):59-70.

WANG Xiaoran,FAN Chunju,HU Yan,et al.Single - phase grounding fault line selection method for active distribution network with energy storage power stations[J]. Journal of Electric Power Science and Technology,2025,40(5):59-70.

王宾,崔鑫,董新洲.配电线路弧光高阻故障检测技术综述[J].中国电机工程学报,2020,40(1):96-107.

WANG Bin,CUI Xin,DONG Xinzhou.Overview of arc high impedance grounding fault detectiontechnologies in distribution system[J]. Proceedings of the CSEE,2020,40(1):96-107.

罗国敏,杨雪凤,尚博阳,等.基于改进堆叠降噪自编码器的配电网高阻接地故障检测方法[J].电力系统保护与控制,2024,52 (24):149-160.

LUO Guomin,YANG Xuefeng,SHANG Boyang,et al.High impedance grounding fault detection method of a distribution network based on an improved stacked denoised autoencoder[J]. Power System Protection and Control,2024,52(24):149-160.

白浩,潘姝慧,邵向潮,等.基于小波去噪与随机森林的配电网高阻接地故障半监督识别方法[J].电力系统保护与控制, 2022,50(20):79-87.

BAI Hao,PAN Shuhui,SHAO Xiangchao,et al.A high impedance grounding fault semi-supervised identification method based on wavelet denoising and random forest[J]. Power System Protection and Control,2022,50(20):79-87.

曾晶.配电网高阻接地故障诊断方法研究[D].长沙:湖南大学,2019.

ZENG Jing.Research on high impedance grounding fault detection method of distribution network[D]. Changsha:Hunan University, 2019.

郑星炯.基于支持向量机的配电线路高阻接地故障检测方法[J].电子设计工程,2015,23(14):122-125.

ZHENG Xingjiong.SVM-based method of high impedance fault detection in distribution network[J]. Electronic Design Engineering, 2015,23(14):122-125.

戴栩生.配电网故障检测与保护方法研究[D].长沙:湖南大学,2015.

DAI Xusheng.Research on fault detection and protection methods for distribution networks[D]. Changsha:Hunan University,2015.

许光,刘漫雨,王兴念,等.基于多信息融合的自适应单相接地故障在线定位研究与应用[J].电测与仪表,2019,56(12):64-72.

XU Guang,LIU Manyu,WANG Xingnian,et al. Research and application of adaptive single phase ground fault online locating based on multi-information fusion[J]. Electrical Measurement &Instrumentation,2019,56(12):64-72.

林雨丰,薛毓强,高超.基于多故障特征量提取的配电网单相接地故障类型识别[J].供用电,2020,37(6):40-47.

LIN Yufeng,XUE Yuqiang,GAO Chao.Distribution network singlephase grounding fault type identification based on multi-fault feature extraction[J]. Distribution & Utilization,2020,37(6):40-47.

黄建业,钱健,陈春生,等.配电网单相接地现场试验及结果分析[J].供用电,2018,35(5):31-37.

HUANG Jianye,QIAN Jian,CHEN Chunsheng,et al.Field test and result analysis of single phase grounding for distribution network[J]. Distribution & Utilization,2018,35(5):31-37.

李天友,徐丙垠,薛永端.配电网高阻接地故障保护技术及其发展[J].供用电,2018,35(5):1-6.

LI Tianyou,XU Bingyin,XUE Yongduan. High-impedance fault protection of distribution networks and its developments[J]. Distribution & Utilization,2018,35(5):1-6.

耿建昭,王宾,董新洲,等.中性点有效接地配电网高阻接地故障特征分析及检测[J].电力系统自动化,2013,37(16):85-91.

GENG Jianzhao,WANG Bin,DONG Xinzhou,et al.Analysis and detection of high impedance grounding fault in neutral point effectively grounding distribution network[J]. Automation of Electric Power Systems,2013,37(16):85-91.

李景丽,袁豪,徐铭铭,等.小电阻接地系统高阻接地故障检测技术综述[J].电测与仪表,2023,60(6):10-18.

LI Jingli,YUAN Hao,XU Mingming,et al. Overview of high impedance fault detection technology in small resistance grounding system[J]. Electrical Measurement & Instrumentation,2023,60(6):10-18.

周封,朱瑞,王晨光,等.一种配电网高阻接地故障在线监测与辨识方法[J].仪器仪表学报,2015,36(3):685-693.

ZHOU Feng,ZHU Rui,WANG Chenguang,et al. Online criterion and identification of single-phase ground fault with high resistence in distribution network[J]. Chinese Journal of Scientific Instrument, 2015,36(3):685-693.

许桉晨,崔双喜,邓顺黎.基于时间序列分析与预测的电—气互联综合能源系统运行风险评估[J]. 智慧电力,2024,52(11):40-47.

XU Anchen,CUI Shuangxi,DENG Shunli.Operational risk assessment of electricity-gas interconnected integrated energy system based on time series analysis and prediction[J]. Smart Power,2024,52(11):40-47.

李嵩,夏成军,刘译夫,等.基于KAN和误差时间序列自相关性特征的多时间尺度光伏功率预测[J].电力系统保护与控制, 2025,53(24):176-187.

LI Song,XIA Chengjun,LIU Yifu,et al.Multi-time scale PV power forecasting based on KAN and error time series autocorrelation features[J]. Power System Protection and Control,2025,53(24):176-187.

陈容均,严宣辉,杨超城.面向时间序列的混合图像化循环胶囊分类网络[J].计算机应用,2023,43(3):692-699.

CHEN Rongjun,YAN Xuanhui,YANG Chaocheng. Fusion imagingbased recurrent capsule classification network for time series[J]. Journal of Computer Applications,2023,43(3):692-699.

袁智勇,白浩,邵向潮,等.基于一维多尺度轻量级DenseNet的配电网高阻接地故障检测方法[J].南方电网技术,2022,16 (6):33-43.

YUAN Zhiyong,BAI Hao,SHAO Xiangchao,et al.Detection method of high impedance grounding fault in distribution network based on one-dimensional multi-scale lightweight DenseNet[J]. Southern Power System Technology,2022,16(6):33-43.

黄宜军,汪金友.小波分析在微弱信号测量中的应用研究[J].计量学报,2007,28(2):163-166.

HUANG Yijun,WANG Jinyou.Application of wavelet threshold denoising to weak signal measurement[J]. Acta Metrologica Sinica, 2007,28(2):163-166.

HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).USA:IEEE, 2016:770-778.

吴忠强,卢雪琴.基于深度迁移学习和LSTM网络的微电网故障诊断[J].计量学报,2023,44(4):582-590.

WU Zhongqiang,LU Xueqin.Microgrid fault diagnosis based on deep transfer learning and LSTM network[J]. Acta Metrologica Sinica,2023,44(4):582-590.

李长鹏,吴小钊,张文凯,等.中压气体绝缘金属封闭开关设备泄压孔设计方法研究[J].高压电器,2016,52(7):155-160.

LI Changpeng,WU Xiaozhao,ZHANG Wenkai,et al.Research on the design method of pressure relief hole of medium voltage gas insulated metal-enclosed switchgear[J]. High Voltage Apparatus, 2016,52(7):155-160.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Combined Method for Traveling Wave Location of Single Phase Grounding Fault in Tree Type Distribution Network

Real-time Parameter Measurement Method Against Ground Based on Hybrid Flexible Arc Suppression

Collaborative Optimization Method for Photovoltaic and Energy Storage System in Distribution Network Based on Improved Circulatory System Algorithm

Adaptive Current Protection Scheme for Distribution Network with Inverted Interfaced Distributed Power Generator

Fault Identification Method for Newly Added Categories in High Voltage Circuit Breakers Based on Improved Transfer Learning

Related Author

ZHANG Cuiling

JIANG Xuechen

NING Yi

GUO Runsheng

WANG Huanming

LI Liang

NIU Wei

HU Fan

Related Institution

Ningxia Institute of Science and Technology

College of Information Science and Engineering,Northeastern University

Shuozhou Power Supply Company, State Grid Shanxi Electric Power Company

Taiyuan Power Supply Company, State Grid Shanxi Electric Power Company

Beijing Information Science and Technology University

⁰