李奎, 张丹, 王尧.基于卷积自编码网络的故障电弧多分类识别方法[J].河南理工大学学报（自然科学版）,2022,41(4):107-116.

LI K, ZHANG D, WANG Y.Multi-classification recognition method of arc fault based on convolutional autoencoder network[J].Journal of Henan Polytechnic University(Natural Science) ,2022,41(4):107-116.

基于卷积自编码网络的故障电弧多分类识别方法

李奎1,2, 张丹1,2, 王尧1,2

1.河北工业大学 河北省电磁场与电器可靠性重点实验室，天津 300130;2.河北工业大学 省部共建电工装备可靠性与智能化国家重点实验室，天津 300130

摘要:针对非线性负载条件下线路正常工作电流波形与故障电弧波形具有相似特征，容易引起故障电弧保护装置误动作的问题，提出一种基于卷积自编码网络的故障电弧多分类识别方法。采用卷积自动编码器进行故障电弧特征提取，优化设计卷积自动编码器网络模型参数，并利用 Softmax多分类器建立故障电弧多分类识别网络模型。实验结果表明，所提方法的故障电弧识别准确率达到99.31％，相应负载类型识别准确率达到97.94％，满足故障电弧识别要求。

关键词:串联故障电弧;卷积自动编码器;Softmax多分类器;非线性负载;故障电弧识别

doi:10.16186/j.cnki.1673-9787.2021010112

基金项目:国家自然科学基金资助项目（51607055）；河北省自然科学基金资助项目（E2020202204）；特种电机与高压电器教育部重 点实验室（沈阳工业大学）开放课题项目（KFKT202003 ）；浙江省基础公益研究计划项目（LGG20E070002 ）

收稿日期:2021/01/24

修回日期:2021/03/09

出版日期:2022/07/15

Multi-classification recognition method of arc fault based on convolutional autoencoder network

LIKui 1,2, ZHANGDan 1,2, WANGYao 1,2

1.Key Laboratory of Electromagnetic Field and Electrical Apparatus Reliability of Hebei Province， Hebei University of Technology，Tianjin  300130，China;2.State Key Laboratory of Reliability and Intelligence of Electrical Equipment，Hebei University of Technology,Tianjin  300130，China

Abstract: Under non-linear load conditions,the current waveform during normal operation has similar charac- teristics as arc faults. The arc fault protection device is prone to malfunction.A multi-class recognition method of arc fault based on convolutional autoencoder network was proposed. A convolutional autoencoder was used to extract arc fault features,the parameters were optimized,and Softmax multi-classifier was used to build the arc fault multi-classification and recognition network model. Experimental results showed that,the arc fault identification accuracy of the proposed method was 99.31%, the identification accuracy of the corresponding load type reached 97.94%. It met the requirements of arc fault identification.

Key words:series arc fault;convolutional autoencoder;Softmax multi-classifier;non-linear load;arc fault recognition

基于卷积自编码网络的故障电弧多分类识别方法_李奎.pdf