基于AlexNet深度学习网络的串联故障电弧检测方法
|
摘要
在家用交流供配电系统中,接触松动等原因可能会导致故障电弧的发生,威胁用电系统的安全。线路发生串联故障电弧时的电流基本与正常运行时的电流大小一致,具有很强的隐蔽性。对此,首次提出用深度学习检测电流信号的方法来检测串联故障电弧,该方法只需将电流信号输入深度学习网络,由网络自主挖掘隐含在电流信号数据背后的特征,实现对串联故障电弧的识别。搭建实验平台,并用开关模拟发生正常电弧,分别采集电阻性负载、电感性负载和阻感性负载正常运行和发生串联故障电弧时的电流数据共7 200组。构建AlexNet卷积神经网络并做相应改进,用采集到的数据训练网络并测试,结果显示辨识平均准确率在85%以上,表明该方法能够较好的实现对串联故障电弧的检测。
In domestic alternating current power supply and distribution system, arc fault is caused by loose connect and other reasons and endangers the power system. The current that series arc fault generates is basically the same as the normal operation, so it is concealed. For this reason, an arc fault detection method using deep learning to detect current data is firstly proposed. This method only needs to input the current signal data to the deep learning network, the feature hidden behind the current signal data autonomously extracted by the network and the series arc fault is identified. A platform is set up and a key is used to simulate the normal arc. Normal current signals and series arc fault current signals of resistive load, inductance load and resistive-inductive load totally 7 200 groups are collected using the platform AlexNet convolutional neural network is constructed and improved accordingly, and trained and tested using the collected data. Experiments show that the average accuracy of identification is above 85%, which indicates that the series arc fault can be better detected.
In domestic alternating current power supply and distribution system, arc fault is caused by loose connect and other reasons and endangers the power system. The current that series arc fault generates is basically the same as the normal operation, so it is concealed. For this reason, an arc fault detection method using deep learning to detect current data is firstly proposed. This method only needs to input the current signal data to the deep learning network, the feature hidden behind the current signal data autonomously extracted by the network and the series arc fault is identified. A platform is set up and a key is used to simulate the normal arc. Normal current signals and series arc fault current signals of resistive load, inductance load and resistive-inductive load totally 7 200 groups are collected using the platform AlexNet convolutional neural network is constructed and improved accordingly, and trained and tested using the collected data. Experiments show that the average accuracy of identification is above 85%, which indicates that the series arc fault can be better detected.
引文
[1] 余琼芳.基于小波分析及数据融合的电气火灾预报系统及应用研究[D].秦皇岛:燕山大学,2013.YU Q F.Research on electrical fire forecast system and its application based on wavelet analysis and data fusion [D].Qinghuangdao:Yanshan University,2013.
[2] 杨艺,董爱华,付永丽.低压故障电弧检测概述[J].低压电器,2009,27(5):1-4.YANG Y,DONG AI H,FU Y L.Overview of low voltage fault arc detection [J].Low Voltage Apparatus,2009,27(5):1-4.
[3] 刘丽,董文忠.低压串联电弧故障特征与检测方法分析[J].电气开关,2015(3):12-14.LIU L,DONG W Z.Low voltage series arc fault characteristics and test method analysis [J].Electric Switchgear,2015 (3):12-14.
[4] 杨艺.楼宇电气火灾预警系统中故障电弧检测的研究[D].焦作:河南理工大学,2007.YANG Y.Research on detecting of fault arc in building electrical fire pre-warning system [D].Jiaozuo:Henan Polytechnic University,2007.
[5] YU Q F,ZHENG D Z,YANG Y,et al.An arc fault detection method based on wavelet feature extraction and the design & realization by labwindows/CVI[J].Journal of Computers(Finland),2013,8(2):417-424.
[6] LEZAMA J,SCHWEITZER P,WEBER S,et al.Arc fault detection based on temporal analysis[C].2014 IEEE 60th Holm Conference on Electrical Contacts (Holm),2014:1-5.
[7] 张扬,刘艳丽.PSO优化BP神经网络的串联故障电弧识别方法[J].传感器与微系统,2016,35(7):22-25.ZHANG Y,LIU Y L.Series fault arc identification method based on BP neural network optimized by PSO [J].Transducer and Microsystem Technologies,2016,35(7):22-25.
[8] 刘振国,孙鹏,由广祥.数学形态学在串联型故障电弧诊断中的应用[J].高压电器,2016,52(9):190-195.LIU ZH G,SUN P,YOU G X.Application of mathematical morphology in the diagnosis of arc fault in series [J].High Voltage Apparatus,2016,52(9):190-195.
[9] WANG Z,BALOG R S.Arc fault and flash detection in photovoltaic systems using wavelet transform and support vector machines [C].2016 IEEE 43rd Photovoltaic Specialists Coference,2016:3275-3280.
[10] 张俊康,邓启亮,唐金城,等.串联故障电弧的特征分析与建模研究[J].电工技术,2017(1):30-32.ZHANG J K,DENG Q L,TANG J CH,et al.Characteristic analysis and modeling study of series fault arc [J].Electric Engineering,2017 (1):30-32.
[11] 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):70-74.CHEN ZH,LI K,ZHANG Y Z,et al.Arc fault detection based on electromagnetic radiation [J].Advanced Technology of Electrical Engineering and Energy,2017,36(3):70-74.
[12] LIU Y L,GUO F Y,REN Z L,et al.Feature analysis in time-domain and fault diagnosis of series arc fault [C].2017 IEEE Holm Conference on Electrical Contacts,2017:306-311.
[13] GAO H X,WANG X L,NGUYEN T,et al.Research on feature of series arc fault based on improved SVD[C].2017 IEEE Holm Conference on Electrical Contacts,2017:325-331.
[14] 李岚松,周越,熊翔,等.基于LS-SVM的航空故障电弧诊断[J].电器与能效管理技术,2018,59(10):45-49.LI L S,ZHOU Y,XIONG X,et al.Diagnosis of aviation fault arc based on LS-SVM[J].Electrical & Energy Management Technology,2018,59(10):45-49.
[15] 孟丹.基于深度学习的图像分类方法研究[D].上海:华东师范大学,2017.MENG D.Research of image classification methods based on deep learning [D].Shanghai:East China Normal University,2017.
[16] 程曦,吴云志,张友华,等.基于深度卷积神经网络的储粮害虫图像识别[J].中国农学通报,2018,34(1):154-158.CHENG X,WU Y ZH,ZHANG Y H,et al.Image recognition of stored grain pests:based on deep convolutional neural network [J].Chinese Agricultural Science Bulletin,2018,34(1):154-158.
[17] GU S S,DING L,YANG Y,et al.A new deep learning method based on AlexNet model and SSD model for tennis ball recognition [C].2017 IEEE 10th International Workshop on Computational Intelligence and Applications (IWCIA),2017:159-164.
[18] XIAO L S,YAN Q,DENG S Y.Scene classification with improved AlexNet model[C].2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE),2017:1-6.
[19] 陈东敏,姚剑敏.基于优化的并行AlexNet人脸特征点检测算法[J].信息技术与网络安全,2018,37(4):65-70.CHEN D M,YAO J M.Face points detection based on optimized parallel AlexNet model [J].Information Technology and Network Security,2018,37(4):65-70.
[20] 温江涛,闫常弘,孙洁娣,等.基于压缩采集与深度学习的轴承故障诊断方法[J].仪器仪表学报,2018,39 (1):171-179.WEN J T,YAN CH H,SUN J D,et al.Bearing fault diagnosis method based on compressed acquisition and deep learning[J].Chinese Journal of Scientific Instrument,2018,39 (1):171-179.
[21] 杨理践,曹辉.基于深度学习的管道焊缝法兰组件识别方法[J].仪器仪表学报,2018,39(2):193-202.YANG L J,CAO H.Deep learning based weld and flange identification in pipeline[J].Chinese Journal of Scientific Instrument,2018,39(2):193-202.
[22] KAEWPRATEEP J;PROM-ON S.Evaluation of small-scale deep learning architectures in Thai speech recognition[C].2018 International ECTI Northern Section Conference on Electrical,Electronics,Computer and Telecommunications Engineering (ECTI-NCON),2018:60-64.
[23] YONEL B,MASON E,YAZCI B.Deep learning for waveform estimation in passive synthetic aperture radar[C].2018 IEEE Radar Conference (RadarConf18),2018:1395-1400.
[24] BALOUJI E,GU I Y H,BOLLEN M H J,et al.A LSTM-based deep learning method with application to voltage dip classification[C].2018 18th International Conference on Harmonics and Quality of Power (ICHQP),2018:1-5.
[25] KARADUMAN G,KARAKOSE M,AKIN E.Deep Learning based arc detection in pantograph-catenary systems[C].2017 10th International Conference on Electrical and Electronics Engineering (ELECO),2017:904-908.
[26] KRIZHVESKY A,SUTSKEVER I,HINTON G E.imagenet classification with deep convolutionalneural networks [C].International Conference on Neural Information Processing Systems,2012,60(2):1097-1105.
[27] GB 14287.4-2014.电气火灾监控系统第4部分:故障电弧探测器[S].GB 14287.4-2014.Electrical fire monitoring system-Part 4:Arcing fault detectors[S].
[2] 杨艺,董爱华,付永丽.低压故障电弧检测概述[J].低压电器,2009,27(5):1-4.YANG Y,DONG AI H,FU Y L.Overview of low voltage fault arc detection [J].Low Voltage Apparatus,2009,27(5):1-4.
[3] 刘丽,董文忠.低压串联电弧故障特征与检测方法分析[J].电气开关,2015(3):12-14.LIU L,DONG W Z.Low voltage series arc fault characteristics and test method analysis [J].Electric Switchgear,2015 (3):12-14.
[4] 杨艺.楼宇电气火灾预警系统中故障电弧检测的研究[D].焦作:河南理工大学,2007.YANG Y.Research on detecting of fault arc in building electrical fire pre-warning system [D].Jiaozuo:Henan Polytechnic University,2007.
[5] YU Q F,ZHENG D Z,YANG Y,et al.An arc fault detection method based on wavelet feature extraction and the design & realization by labwindows/CVI[J].Journal of Computers(Finland),2013,8(2):417-424.
[6] LEZAMA J,SCHWEITZER P,WEBER S,et al.Arc fault detection based on temporal analysis[C].2014 IEEE 60th Holm Conference on Electrical Contacts (Holm),2014:1-5.
[7] 张扬,刘艳丽.PSO优化BP神经网络的串联故障电弧识别方法[J].传感器与微系统,2016,35(7):22-25.ZHANG Y,LIU Y L.Series fault arc identification method based on BP neural network optimized by PSO [J].Transducer and Microsystem Technologies,2016,35(7):22-25.
[8] 刘振国,孙鹏,由广祥.数学形态学在串联型故障电弧诊断中的应用[J].高压电器,2016,52(9):190-195.LIU ZH G,SUN P,YOU G X.Application of mathematical morphology in the diagnosis of arc fault in series [J].High Voltage Apparatus,2016,52(9):190-195.
[9] WANG Z,BALOG R S.Arc fault and flash detection in photovoltaic systems using wavelet transform and support vector machines [C].2016 IEEE 43rd Photovoltaic Specialists Coference,2016:3275-3280.
[10] 张俊康,邓启亮,唐金城,等.串联故障电弧的特征分析与建模研究[J].电工技术,2017(1):30-32.ZHANG J K,DENG Q L,TANG J CH,et al.Characteristic analysis and modeling study of series fault arc [J].Electric Engineering,2017 (1):30-32.
[11] 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):70-74.CHEN ZH,LI K,ZHANG Y Z,et al.Arc fault detection based on electromagnetic radiation [J].Advanced Technology of Electrical Engineering and Energy,2017,36(3):70-74.
[12] LIU Y L,GUO F Y,REN Z L,et al.Feature analysis in time-domain and fault diagnosis of series arc fault [C].2017 IEEE Holm Conference on Electrical Contacts,2017:306-311.
[13] GAO H X,WANG X L,NGUYEN T,et al.Research on feature of series arc fault based on improved SVD[C].2017 IEEE Holm Conference on Electrical Contacts,2017:325-331.
[14] 李岚松,周越,熊翔,等.基于LS-SVM的航空故障电弧诊断[J].电器与能效管理技术,2018,59(10):45-49.LI L S,ZHOU Y,XIONG X,et al.Diagnosis of aviation fault arc based on LS-SVM[J].Electrical & Energy Management Technology,2018,59(10):45-49.
[15] 孟丹.基于深度学习的图像分类方法研究[D].上海:华东师范大学,2017.MENG D.Research of image classification methods based on deep learning [D].Shanghai:East China Normal University,2017.
[16] 程曦,吴云志,张友华,等.基于深度卷积神经网络的储粮害虫图像识别[J].中国农学通报,2018,34(1):154-158.CHENG X,WU Y ZH,ZHANG Y H,et al.Image recognition of stored grain pests:based on deep convolutional neural network [J].Chinese Agricultural Science Bulletin,2018,34(1):154-158.
[17] GU S S,DING L,YANG Y,et al.A new deep learning method based on AlexNet model and SSD model for tennis ball recognition [C].2017 IEEE 10th International Workshop on Computational Intelligence and Applications (IWCIA),2017:159-164.
[18] XIAO L S,YAN Q,DENG S Y.Scene classification with improved AlexNet model[C].2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE),2017:1-6.
[19] 陈东敏,姚剑敏.基于优化的并行AlexNet人脸特征点检测算法[J].信息技术与网络安全,2018,37(4):65-70.CHEN D M,YAO J M.Face points detection based on optimized parallel AlexNet model [J].Information Technology and Network Security,2018,37(4):65-70.
[20] 温江涛,闫常弘,孙洁娣,等.基于压缩采集与深度学习的轴承故障诊断方法[J].仪器仪表学报,2018,39 (1):171-179.WEN J T,YAN CH H,SUN J D,et al.Bearing fault diagnosis method based on compressed acquisition and deep learning[J].Chinese Journal of Scientific Instrument,2018,39 (1):171-179.
[21] 杨理践,曹辉.基于深度学习的管道焊缝法兰组件识别方法[J].仪器仪表学报,2018,39(2):193-202.YANG L J,CAO H.Deep learning based weld and flange identification in pipeline[J].Chinese Journal of Scientific Instrument,2018,39(2):193-202.
[22] KAEWPRATEEP J;PROM-ON S.Evaluation of small-scale deep learning architectures in Thai speech recognition[C].2018 International ECTI Northern Section Conference on Electrical,Electronics,Computer and Telecommunications Engineering (ECTI-NCON),2018:60-64.
[23] YONEL B,MASON E,YAZCI B.Deep learning for waveform estimation in passive synthetic aperture radar[C].2018 IEEE Radar Conference (RadarConf18),2018:1395-1400.
[24] BALOUJI E,GU I Y H,BOLLEN M H J,et al.A LSTM-based deep learning method with application to voltage dip classification[C].2018 18th International Conference on Harmonics and Quality of Power (ICHQP),2018:1-5.
[25] KARADUMAN G,KARAKOSE M,AKIN E.Deep Learning based arc detection in pantograph-catenary systems[C].2017 10th International Conference on Electrical and Electronics Engineering (ELECO),2017:904-908.
[26] KRIZHVESKY A,SUTSKEVER I,HINTON G E.imagenet classification with deep convolutionalneural networks [C].International Conference on Neural Information Processing Systems,2012,60(2):1097-1105.
[27] GB 14287.4-2014.电气火灾监控系统第4部分:故障电弧探测器[S].GB 14287.4-2014.Electrical fire monitoring system-Part 4:Arcing fault detectors[S].