基于频谱包络改进的EWT方法的滚动轴承故障诊断研究  

Research on fault diagnosis of rolling bearings based on improved EWT method of spectrum envelope

作　　者：唐诗尧 何佳 任锦胜 吴仕明 柳亦兵[3] Tang Shiyao;He Jia;Ren Jinsheng;Wu Shiming;Liu Yibing(China Resources Power Technology Research Institute Company,Shenzhen 518000,China;Beijing Envada Electric Power Engineering Technology Co.,Ltd.,Beijing 100022,China;Key Laboratory of Condition Monitoring and Control for Power Plant Equipment of Ministry of Education,North China Electric Power University,Beijing 102206,China)

机构地区：[1]华润电力技术研究院有限公司,广东深圳518000 [2]北京英华达电力电子工程科技有限公司,北京100022 [3]华北电力大学电站设备状态监测与控制教育部重点实验室,北京102206

出　　处：《可再生能源》2022年第1期60-66,共7页Renewable Energy Resources

基　　金：国家重点基础研究发展计划项目(973计划)(2017YFC0805905)。

摘　　要：经验小波变换方法被证明是一种有效的滚动轴承故障诊断方法,但该方法的分析精度依赖于频谱的合理分割。因此,文章提出了一种基于改进经验小波变换的故障特征提取方法。首先,通过连接若干个信号频谱的局部极大值来获取频谱包络线;然后,设定阈值,消除噪声干扰;最后,根据频谱包络线的局部极小值来自适应地确定频谱分割边界。工程实例分析表明,基于改进经验小波变换的故障特征提取方法,提高了对故障特征频带的分离精度,在滚动轴承故障特征提取方面表现出一定的优越性。The empirical wavelet transform method has proven to be an effective method for rolling bearing fault diagnosis, but the analysis accuracy of this method is dependent on the reasonable segmentation of the frequency spectrum. Therefore, this paper proposes a fault feature extraction method based on the improved empirical wavelet transform. First, the envelope of the frequency spectrum is obtained by linking several local maximum in the frequency spectrum. Second, a threshold is appointed to eliminate noise interference. Finally, the boundaries for spectrum division is adaptively determined based on the envelope of the spectrum. The analysis of engineering examples shows that the improved empirical wavelet transform method can obviously improve the separation accuracy of the fault characteristic frequency band by improving the frequency spectrum segmentation and it shows advantages compared with the original EWT method. The analysis of engineering examples shows that the improved empirical wavelet transform method can improve the separation accuracy of the fault characteristic frequency band by improving the frequency spectrum segmentation of the original EWT method.

关 键 词：风力发电机 滚动轴承 经验小波变换 频谱划分 特征提取 

分 类 号：TK83[动力工程及工程热物理—流体机械及工程]