Recognition of Series Arc Fault Based on K-Means Clustering

doi:10.11985/2016.08.004

Abstract

Abstract:

In view of the fire and other accidents caused by series arc fault in a power supply and distribution system, an experimental device of series arc fault is developed and a lots of experiments about typical loads are carried out. Firstly, the first 20 times harmonic content of the current before and after arc fault are extracted by Fourier transform. Secondly, the dimension of sample data is reduced by using principal component analysis(PCA). A few main components are extracted to reflect the fluctuations of current signal. Finally, the main harmonic data are analyzed by K-means clustering to determine whether the fault occurred in original signal. The results show that the series arc fault could be recognized through principal component analysis and K-means clustering.

Key words: Series arc fault, harmonic content, principal component analysis, K-means clustering

CLC Number:

TM501

Changken Chen,Yanli Liu,Ying Li. Recognition of Series Arc Fault Based on K-Means Clustering[J]. Journal of Electrical Engineering, 2016, 11(8): 23-29.

Figures/Tables 12

Fig.1

Fig.2

Tab.1

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Tab.4

Tab.5

References 20

[1]	Peter Muller, Stefan Tenbohlen. Characteristics of series and parallel low current arc fault in the time and frequency domain[C]. The 56th Holm Conference, 2010: 223-229.
[2]	刘晓明, 徐叶飞, 刘婷 , 等. 基于电流信号短时过零率的电弧故障检测[J]. 电工技术学报, 2015,30(13):125-133.
	Liu Xiaoming, Xu Yefei, Liu Ting , et al. The arc fault detection based on the current signal short-time zero crossing rate[J]. Transactions of China Electrotechnical Society, 2015,30(13):125-133.
[3]	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):70-77.
	Yang Kai, Zhang Rencheng, Yang Jianhong , et al. Series arc fault diagnostic method based on fractal dimension and support vector machine[J]. Transactions of China Electrotechnical Society, 2016,31(2):70-77.
[4]	丁浩江, 张认成, 杨凯 , 等. 利用电磁辐射信号的串并联电弧故障检测[J]. 华侨大学学报:自然科学版, 2015,36(2):132-136.
	Ding Haojiang, Zhang Rencheng, Yang Kai , et al. Series and parallel arc fault detection using electromagnetic radiation signal[J]. Journal of Huaqiao University(Natural Science), 2015,36(2):132-136.
[5]	楼嘉宇, 马琪 . 一种基于SVM的电弧故障检测方法[J]. 杭州电子科技大学学报:自然科学版, 2015,35(6):28-31.
	Lou Jiayu, Ma Qi . An arc fault detection approach based on SVM[J]. Journal of Hangzhou Dianzi University(Natural Science), 2015,35(6):28-31.
[6]	Brechtken D. Precentive arc fault protection[C]. IEEE/PES Transmission and Distribution Conference, USA, 2001: 311-316.
[7]	Duan Peiyong, Xu Liping, Ding Xudong, et al. An arc fault diagnostic method for low voltage lines using the difference of wavelet coefficients[C]. The 9th IEEE Conference on Industrial Electronics and Applications, 2014: 401-405.
[8]	Peng Sun, Xiang Gao. Series arc fault diagnosis technology research based on the analysis of current rate[C]. The 1st International Conference on Electric Power Equipment-Switching Technology, 2011: 565-568.
[9]	Bao Jieqiu, Zhang Yi, Zhang Hongqiang. Arc fault identification method based on fractal theory and SVM[C]. The International Conference on Power System Technology, 2014: 1182-1187.
[10]	孙鹏, 郑志成, 高翔 . 基于小波分析的故障电弧检测方法[J]. 高压电器, 2012,48(1):25-34.
	Sun Peng, Zheng Zhicheng, Gao Xiang . Arc fault detection based on wavelet analysis[J]. High Voltage Apparatus, 2012,48(1):25-34.
[11]	廖水容, 张认成, 李夏河 . 低压串联故障电弧谐波含有率分析[J]. 湖南科技大学学报:自然科学版, 2013,28(1):35-38.
	Liao Shuirong, Zhang Rencheng, Li Xiahe . Harmonic ratio analysis for series low voltage arc fault[J]. Journal of Hunan University of Science &Technology(Natural Science Edition), 2013,28(1):35-38.
[12]	马少华, 郭家稳 . 低压串联故障电弧的识别方法[J]. 低压电器, 2013(9):12-16.
	Ma Shaohua, Guo Jiawen. Identification method of low voltage series arc fault[J]. Low Voltage Apparatus, 2013(9):12-16.
[13]	姜斌峰, 王莉 . 低压交流电线故障电弧模型研究[J]. 电力系统及其自动化学报, 2009,21(4):20-24.
	Jiang Binfeng , Wangli. Research on arc model in AC low-voltage electrical wire fault[J]. Proceedings of the CSU-EPSA, 2009,21(4):20-24.
[14]	卢其威, 巫海东, 王肃珂 , 等. 基于差值-均方根法的故障电弧检测的研究[J]. 低压电器, 2013(1):6-13.
	Lu Qiwei, Wu Haidong, Wang Suke , et al. Research on arc-fault detection based on difference-root mean square method[J]. Low Voltage Apparatus, 2013(1):6-13.
[15]	刘艳丽, 郭凤仪, 王智勇 , 等. 基于信息熵的串联型故障电弧频谱特征研究[J]. 电工技术学报, 2015,30(12):488-495.
	Liu Yanli, Guo Fengyi, Wang Zhiyong, et al. Research on the spectral characteristics of series arc fault based on information entropy[J]. Transactions of China Electrotechnical Society, 2015,30(12):488-495.
[16]	杨淑莹 . 模式识别与智能计算——Matlab技术实现[M]. 北京: 电子工业出版社, 2011.
[17]	中国国家标准化管理委员会. 电气火灾监控系统第4部分:故障电弧探测器[S]. 北京: 中国标准出版社, 2014.
[18]	付丽华 . 谐波信号分析与处理[M]. 武汉: 中国地质大学出版社, 2013.
[19]	夏骄雄, 徐俊, 吴耿峰 . 数据库主成分提取方法及其应用[J]. 计算机工程与应用, 2006,42(20):134-137.
	Xia Jiaoxiong, Xu Jun, Wu Gengfeng . Database principal component extraction and its application[J]. Computer Engineering and Applications, 2006,42(20):134-137.
[20]	何晓群 . 多元统计分析[M]. 北京: 中国人民大学出版社, 2015.

负载性质	负载类型	工作状态
负载性质	负载类型	工作电压/V	工作电流/A
阻性负载	灯泡4×200W	220	3.6
	灯泡8×200W	220	7.3
	电水壶	220	7
感性负载	三相电动机2kW	380	3
	三相电动机4kW	380	6
	角磨机	220	3.4
	电钻	220	3.4
非线性负载	1台电脑	220	1.4
非线性负载	2台电脑	220	2.7
混合负载	1台电脑,灯泡1×100W	220	0.58

负载	2次谐波	3次谐波	4次谐波	5次谐波
灯泡	0.123 137 6	0.091 164 7	0.043 988 7	0.110 577 2
电水壶	0.119 004 5	0.271 121 0	0.050 845 5	0.066 910 0
三相电动机	0.042 570 0	0.036 604 3	0.004 813 8	0.027 622 7
角磨机	0.035 302 4	0.337 295 4	0.005 594 8	0.054 893 1
电钻	0.013 499 6	0.100 303 3	0.005 468 1	0.034 650 8
电脑2台	0.020 890 7	0.513 656 6	0.028 494 6	0.303 242 0
混合负载	0.014 181 8	0.270 916 5	0.012 953 7	0.131 382 2

负载	2次谐波	3次谐波	4次谐波	5次谐波
灯泡	0.050 546 8	0.279 943 7	0.066 070 3	0.237 655 523
电水壶	0.155 737 3	0.305 942 7	0.044 926	0.119 718 477
三相电动机	0.222 643 4	0.275 111 3	0.160 381 7	0.064 772 895
角磨机	0.018 807 4	0.378 355 9	0.013 027 3	0.053 417 073
电钻	0.005 564 4	0.098 682 7	0.002 899 7	0.027 169 725
电脑2台	0.045 078 5	0.497 733 7	0.049 022 7	0.294 715 223
混合负载	0.010 226 1	0.247 897 3	0.006 903 1	0.131 327 229

已知结果负载	样本数量	识别率（%）
灯泡8×200W	50	100
电水壶	50	100
三相电动机2kW	50	96
角磨机	50	99
手电钻	50	97
电脑2台	50	99

负载类型	样本数量	识别率（%）
灯泡4×200W	100	96
三相电动机4kW	100	96
电脑1台	100	96
电脑1台和灯泡1×100W	100	92