引用本文:全玉生,何秋宇,卢天盛,马彦伟,王辰.电力变压器局放信号抗干扰与模式识别[J].电力自动化设备,2008,(1):57-60,76
.Anti-interference and pattern recognition of power transformer partial discharge signals[J].Electric Power Automation Equipment,2008,(1):57-60,76
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 3912次   下载 1 本文二维码信息
码上扫一扫!
电力变压器局放信号抗干扰与模式识别
全玉生,何秋宇,卢天盛,马彦伟,王辰
作者单位
摘要:
在局部放电信号抗干扰方面,考察了几种较新的阈值除噪方法,提出了将二进小波变换分别与单抽头最小均方LMS(Lease Mear Square)算法滤波器和基于Birge-Massart策略的小波去噪相结合的变压器局部放电信号抗干扰方法.同时针对局部放电信号的模式识别问题,首先分别根据模糊概率理论和波形匹配理论,针对时域单个特高频UHF(Ultra High Frequency)脉冲信号,提出了2种电力变压器特高频局部放电信号的模式识别方法.基于时域单个UHF脉冲信号的信息,提出了用于神经网络输入的8个特征量,分别利用BP网络、Elman回归神经网络和PNN概率神经网络对4种典型的变压器局部放电信号进行了模式识别的尝试,取得了较好的识别效果.
关键词:  电力变压器  局部放电  抗干扰  模式识别  电力变压器  信号抗干扰  模式识别  signals  partial discharge  power transformer  识别效果  概率神经网络  回归神经网络  Elman  利用  特征量  输入  信息  识别方法  脉冲信号  Ultra  High Frequency  特高频  匹配理论
DOI:
分类号:TM41 TM835
基金项目:
Anti-interference and pattern recognition of power transformer partial discharge signals
QUAN Yusheng  HE Qiuyu  LU Tiansheng  MA Yanwei  WANG Chen
Abstract:
Based on several de- noising methods,it is proposed to combine the binary wavelet transform with the one spur track LMS filter and the wavelet de- noising based on Birge- Massart strategy respectively for the anti- interference of power transformer partial discharge signals. For the single UHF pulse signal,two methods of power transformer PD pattern recognition based on fuzzy probability theory and UHF pulse wave matching theory respectively are introduced, Based on the information of single UHF pulse signal in time domain,eight inputs of neural network are proposed,which are applied to the BP network,Elman network and PNN network respectively to carry out the pattern recognition of four typical power transformer PD signals. Better results are obtained.
Key words:  power transformer  partial discharge  anti- interference  pattern recognition

用微信扫一扫

用微信扫一扫