局部放电脉冲波形特征提取及分类技术-中国电机工程学报.PDF

第 33 卷 第 28 期 中 国 电 机 工 程 学 报 Vol.33 No.28 Oct.5, 2013 168 2013 年 10 月 5 日 Proceedings of the CSEE ©2013 Chin.Soc.for Elec.Eng. 0258-8013 (2013) 28-0168-08 TM 835 A 470 40 文章编号： 中图分类号： 文献标志码： 学科分类号： ⋅ 局部放电脉冲波形特征提取及分类技术 鲍永胜 ( 国网宁夏电力公司银川供电公司，宁夏回族自治区 银川市 750001) Partial Discharge Pulse Waveform Feature Extraction and Classification Techniques BAO Yongsheng (State Grid Ningxia Yinchuan Electric Power Company, Yinchuan 750001, Ningxia Hui Autonomous Region, China) ABSTRACT: In view of the presence of a variety of 关键词：局部放电；在线监测；特征提取；分类；多局放源； interference and multi partial discharge (PD) around the 等效时频；模糊 C 均值 electrical equipment, the paper proposed multi PD online 0 引言 monitoring and identification method for partial discharge based on a single defect. It was pointed out that one of the key 对电工设备的绝缘情况进行局部放电在线监 technologies of developing multi PD online monitoring and 测时，基于传统的脉冲幅值– 时间序列的宽带检测 identification system was the rapid classification pulse groups 系统在存在随机干扰源或多局放源(2 个或以上)时， using pulse waveform feature extraction and clustering analysis 其获取的幅值– 时间序列将是参杂多种放电信号或 algorithm. With respect to 108 times/s ultra-wideband partial 干扰信号的随机混叠序列，相应产生的用于模式识 discharge pulse waveform-time series, the paper focused on the pulse waveform feature extraction and classification techniques. 别的各种放电谱图也是多种信号随机混叠的[1-5] 。而 The improvements on equivalent frequency algorithm and 用于模式识别的数据库一般都是基于单一人工缺 fuzzy C-means clustering algorithm was proposed.