基于MRA的PD信号提取算法与实验研究  

Extracting algorithm and experimental study for PD signals based on MRA

作　　者：赵大伟[1] 郑殿春[1] 白少佐[1] 

机构地区：[1]哈尔滨理工大学电气与电子工程学院,黑龙江哈尔滨150040

出　　处：《哈尔滨工程大学学报》2007年第12期1377-1381,共5页Journal of Harbin Engineering University

基　　金：黑龙江省自然科学基金资助项目(E0303)

摘　　要：小波方法不但具有多尺度分辨和时频局部化特性,而且可以准确地捕捉突变信号的特征,并可以在不同尺度上考察"瞬变"信号特征的演化过程.针对PD信号的瞬变性和随机性,基于小波的MRA法,分析了PD和噪声信号的尺度和小波空间表征,给出了PD信号提取和重构算法,并进行了数学仿真.通过实验模拟验证,结果表明,基于小波MRA的PD信号提取和重构算法快速准确,可以为高压电器绝缘故障诊断、模式辨识提供准确信息源.Theoretical analysis indicates that wavelet transform not only possesses characteristics of multiscale resolution and time-frequency localization, but also exactly captures features of transient signals and reveals the evolving course of instantaneous signals at different scales. In this paper, based on the wavelet MRA method, the scale and wavelet space characterization of PD and noise signals are analyzed and discussed in detail. The extraction and reconstruction methods for PD signals are given based on multi-resolution analysis. Finally, a sults show that this met mathemat hod is not ical simulation and experimental demonstration was carried out. The reonly quick and exact, but also supplies correct signal sources for insulation failure diagnosis and pattern recognition in high voltage equipments.

关 键 词：FOURIER分析 小波变换 局部放电 信号提取 

分 类 号：TM851[电气工程—高电压与绝缘技术]