基于形态自相关和时频切片分析的轴承故障诊断方法  

Bearing fault diagnosis method based on morphological filtering, time-delayed autocorrelation and time-frequency slice analysis

作　　者：钟先友[1,2] 赵春华[1] 陈保家[1] 曾良才[2] 

机构地区：[1]武汉科技大学冶金装备及其控制教育部重点实验室,武汉430081 [2]三峡大学水电机械设备设计与维护湖北省重点实验室,宜昌443002

出　　处：《振动与冲击》2014年第4期11-16,共6页Journal of Vibration and Shock

基　　金：国家自然科学基金资助项目(51075234,51205230)

摘　　要：频率切片小波变换(Frequency Slice Wavelet Transform,FSWT)是一种新的时频分析方法,信号中的噪声会降低FSWT分析的频率分辨率。为了提高分析精度,提出了基于形态滤波和时延自相关的时频切片分析方法,并成功应用到轴承故障诊断中。该方法首先采用多结构元素差值形态滤波和时延自相关方法对信号进行降噪,采用FSWT分解降噪后的轴承振动信号,然后根据轴承故障特征频率选择时间频率切片区间,进行细化分析来提取故障特征。仿真信号与轴承故障振动信号的分析验证了该方法的有效性。Frequency slice wavelet transformation （FSWT）is a new time-frequency analysis method,and the noise in a signal reduces frequency resolution of FSWT.In order to improve the analysis accuracy,a FSWT method based on morphological filtering and time-delayed autocorrelation was proposed and applied to diagnose bearing faults successfully. With this method,the multi-structure element difference morphological filtering method and the time-delayed autocorrelation method were used to reduce the noise of a bearing vibration signal,and the denoised signal was decomposed by applying FSWT,then time and frequency slice intervals were selected for the detailed analysis to extract fault characteristics according to the bearing fault characteristic frequencies.Simulation signal analysis and bearing fault vibration signals analysis demonstrated the effectiveness of the method.

关 键 词：频率切片小波变换 形态滤波 结构元素 时延自相关 轴承故障诊断 

分 类 号：TH113.1[机械工程—机械设计及理论]