采用固定点算法的局部放电特高频信号盲源分离方法  

Blind source separation method of partial discharge extra-high frequencysignal using fixed point algorithm

作　　者：马鹏墀 沈盼 王致杰[1] MA Pengchi;SHEN Pan;WANG Zhijie(School of Electrical Engineering,Shanghai Dianji University,Shanghai 201306,China;Operationand Maintenance Department,Jiangsu Hongyuan Electric Co.,Ltd.,Nanjing 211100,Jiangsu,China)

机构地区：[1]上海电机学院电气学院,上海201306 [2]江苏宏源电气有限责任公司运维检修部,江苏南京211100

出　　处：《上海电机学院学报》2020年第6期323-327,332,共6页Journal of Shanghai Dianji University

基　　金：上海市自然科学基金资助项目(15ZR1417300)。

摘　　要：为了更好地发现现场的局部放电(PD),需要对现场采集的PD信号进行干扰抑制,提出一种采用双树复小波变换(DTCWT)和固定点算法的PD盲源分离方法。首先通过S变换估计源信号的数量,再通过DTCWT对特高频(UHF)PD信号进行分解和重构,最后利用固定点算法对信号进行盲源分离,获取不含噪声的PD信号。利用实验室制作的4种典型放电模型获取PD信号,并人为加入白噪声和周期性窄带干扰信号进行试验,最后对现场实测信号进行盲源分离,实验室和现场的结果均验证了该方法的有效性和实用性。In order to better discover the partial discharge on site,it is necessary to suppress the interference of the partial discharge(PD)signals collected on site.The method of the PD blind source separation using the dual-tree complex wavelet transform(DTCWT)and the fixed point algorithm is proposed.First,the number of the source signals is estimated by the S transform,and then the extra-high frequency(UHF)PD signal is decomposed and reconstructed by the DTCWT.Finally,the fixed point algorithm is used to the blind source separation of the signal to obtain the noise-free PD signal.Four typical discharge models made in the laboratory are used to obtain the PD signals,and the white noise and periodic narrow-band interference signals are artificially added for experiments.Finally,the blind source separation is carried out for the field measured signals.The effectiveness and practicability of the present method are verified by the results of laboratory and field.

关 键 词：局部放电 去噪 盲源分离 双树复小波变换 固定点算法 

分 类 号：TM835[电气工程—高电压与绝缘技术]