2023年 03期

基于深度自编码网络的电弧故障检测与负载类型识别

Arc Fault Detection and Load Type Identification Based on Deep Auto-encoding Network


摘要(Abstract):

针对有监督学习的电弧故障检测方法需要大量带标签数据且大多只检测电弧故障而未对负载类型进行识别的问题,提出一种基于深度自编码网络的电弧故障检测与负载类型识别方法;采用小波包分解的节点系数作为自编码网络的无标签输入特征量,并运用逐层训练方法对自编码网络进行预训练;为了使所提出方法的权重系数达到全局最优,采用少量有标签数据对所得权重进行微调,通过Softmax多分类器输出电弧故障检测结果,并根据负载类别最大概率识别电弧故障可能的负载类型。结果表明,所提出的方法对电弧故障检测与负载类型识别准确率达到98.56%,高于相同层数和参数规模的有监督学习网络的准确率。

关键词(KeyWords): 电弧故障;深度自编码网络;无监督学习;故障检测;负载类型

基金项目(Foundation):国家自然科学基金项目(51907047);; 河北省自然科学基金项目(E2020202204);; 特种电机与高压电器教育部重点实验室开放课题项目(KFKT202003);; 浙江省基础公益研究计划项目(LGG20E070002)

作者(Author):王尧,马啸尘,赵宇初,张丹,李奎

DOI: 10.13349/j.cnki.jdxbn.20230314.001

参考文献(References):

[1] 刘义祥,汤昊,金静,等.基于原因分析的典型城市火灾统计[J].消防科学与技术,2017,36(8):1135.

[2] 罗云庆,何泰健.供电线路中电弧性短路检测技术探讨[J].消防科学与技术,2018,37(11):1557.

[3] 谢正荣,单立辉,时庆兵,等.电气火灾早期预警及监测前沿技术探讨[J].建筑电气,2019,38(8):39.

[4] ARTALE G,CATALIOTTI A,NUCCIO V C S,et al.A set of indicators for arc faults detection based on low frequency harmonic analysis[C]//2016 IEEE International Instrumentation and Mea-surement Technology Conference Proceedings.May 23-26,2016,Taipei,China.New York:IEEE,2016:16176609.

[5] QI P,JOVANOVIC S,LEZAMA J,et al.Discrete wavelet transform optimal parameters estimation for arc fault detection in low-voltage residential power networks[J].Electric Power Systems Research,2017,143:130.

[6] 郑昕,许志红.低压串联电弧故障通用诊断方法的研究[J].电子测量与仪器学报,2014,28(9):991.

[7] 陈照,李奎,张洋子,等.基于电弧电磁辐射的故障电弧识别[J].电工电能新技术,2017,36(3):70.

[8] 戴隽文,郝瑞祥,游小杰.一种用于大功率电弧加热器的等离子电弧模型[J].电工技术学报,2011,26(1):129.

[9] 张冠英,张晓亮,刘华,等.低压系统串联故障电弧在线检测方法[J].电工技术学报,2016,31(8):109.

[10] 王尧,韦强强,葛磊蛟,等.基于电弧电流高频分量的串联交流电弧故障检测方法[J].电力自动化设备,2017,37(7):191.

[11] 孙鹏,郑志成,闫荣妮,等.采用小波熵的串联型故障电弧检测方法[J].中国电机工程学报,2010,30(增刊1):232.

[12] 苏晶晶,许志红.基于EMD和PNN的故障电弧多变量判据诊断方法[J].电力自动化设备,2019,39(4):106.

[13] JIANG J,WEN Z,ZHAO M X,et al.Series arc detection and complex load recognition based on principal component analysis and support vector machine[J].IEEE Access,2019,7:47221.

[14] SIEGEL J E,PRATT S,SUN Y B,et al.Real-time deep neural networks for internet-enabled arc-fault detection[J].Engineering Applications of Artificial Intelligence,2018,74:35.

[15] LEE S,KANG S,KIM T,et al.Improving the performance of serial arc detection using VMD and deep neural network[C]//IEEE 62nd International Midwest Symposium on Circuits and Systems (MWSCAS),August 04-07,2019,Dallas,TX,USA.New York:IEEE,2019:19113029.

[16] ABDULRACHMAN M A,PRASETYONO E,ANGG-RIAWAN D O,et al.Smart detection of AC series arc fault on home vol-tage line based on fast Fourier transf orm and artificial neural network[C]//2019 International Electronics Symposium (IES),September 27-28,2019,Surabaya,Indonesia.New York:IEEE,2019:19172753.

[17] 张士文,张峰,王子骏,等.一种基于小波变换能量与神经网络结合的串联型故障电弧辨识方法[J].电工技术学报,2014,29(6):290.

[18] 杨凯,张认成,杨建红,等.基于分形维数和支持向量机的串联电弧故障诊断方法[J].电工技术学报,2016,31(2):70.

[19] 郭凤仪,李坤,陈昌垦,等.基于小波近似熵的串联电弧故障识别方法[J].电工技术学报,2016,31(24):164.

[20] WANG Z,McCONNELL S,BALOG R S,et al.Arc fault signal detection:Fourier transformation vs.wavelet decomposition techniques using synthesized data[C]//2014 IEEE 40th Photovoltaic Specialist Conference (PV-SC),June 08-13,2014,Denver,CO,USA.New York:IEEE,2014:14683858.

[21] 崔江,唐军祥,龚春英,等.一种基于改进堆栈自动编码器的航空发电机旋转整流器故障特征提取方法[J].中国电机工程学报,2017,37(19):5696.

[22] CHO K,COURVILLE A,BENGIO Y.Describing multimedia content using attention-based encoder-decoder networks[J].IEEE Transactions on Multimedia,2015,17(11):1875.

[23] 贺珂珂.基于深度学习理论的电机故障诊断方法研究[D].兰州:兰州理工大学,2019.