基于小波包与SVM的风电变流器故障诊断  

FAULT DIAGNOSIS OF CONVERTER USED IN WIND POWER GENERATION BASED ON WAVELET PACKET ANALYSIS AND SVM

作　　者：沈艳霞[1] 周文晶[1] 纪志成[1] 吴定会[1] 

机构地区：[1]江南大学电气自动化研究所,无锡214122

出　　处：《太阳能学报》2015年第4期785-791,共7页Acta Energiae Solaris Sinica

基　　金：国家自然科学基金(61104183);教育部新世纪优秀人才支持计划(NCET-10-0437);江苏省自然科学基金(BK2012550)

摘　　要：针对风力发电系统中背靠背式PWM变流器故障诊断问题,以整流状态为例,提出一种基于小波包分析与SVM(支持向量机)分类算法相结合的故障诊断新方法。该方法选取直流侧输出电压信号为研究对象,分析不同开路故障状态下该信号的调制情况,利用小波包分析法提取故障特征样本,最后建立SVM的故障分类器,实现变流器的故障诊断。仿真结果表明该方法可有效实现风力发电系统中变流器的故障诊断。Aiming at the fault diagnosis of back-to-back PWM converter in wind power generation system, taking the rectifier state as an example, a new fault diagnosis method was presented based on wavelet packet analysis and support vector machine （SVM） classification algorithm. DC-side output voltage signal was studied in this method to analyze the modulation degree of signals in the different types of open circuit fault, and fault features were extracted based on wavelet packet analysis, then the SVM-fault classifier was built to implement the fault diagnosis of three-pulse PWM converter. The simulation results show that this method can effectively realize the fault diagnosis of converter used in wind power generation system.

关 键 词：变流器 故障诊断 小波包分析 支持向量机 

分 类 号：TM461[电气工程—电器]