基于小波分形的异步电动机匝间短路故障诊断  

Interturn short circuit fault diagnosis of induction motor based on wavelet fractal

作　　者：郭西进[1] 孔利利[1] 孔令烨[1] 张博洋[1] 许允之[1] 

机构地区：[1]中国矿业大学信息与电气工程学院,江苏徐州221008

出　　处：《矿山机械》2013年第2期100-103,共4页Mining & Processing Equipment

基　　金：国家自然科学基金资助项目(60804022)

摘　　要：针对匝间短路故障的定子电流仿真信号,采用Db5小波包函数进行三层分解。试验表明,利用小波包变换法可有效确定故障发生的时间。基于小波变换与分形理论提出的新型匝间短路故障特征提取方法,对输入的定子电流信号进行降噪处理与多分辨分解,并选择适合的参数,计算分形维数,将其作为支持向量机的输入向量进行故障分类。The Db5 wavelet packet function was applied to conduct three-layer decomposition of simulation signal of the stator current resulting from the intertum short circuit fault. The test showed that it was effective to determine the failure occurrence time. In addition, a new feature extraction method of interturn short circuit was proposed based on wavelet transform and fractal theory, so as to carry out the denoising process and the multi-resolution decomposition of the input stator current signal. Moreover, appropriate parameters was used to calculate the fractal dimensions at all levels, which were taken as input vectors of the support vector machine for fault classification.

关 键 词：定子匝间短路 小波包分解 分形维数 故障特征 支持向量机 

分 类 号：TM307[电气工程—电机] TP277[自动化与计算机技术—检测技术与自动化装置]