基于短时傅里叶变换及其倒频谱的滚动轴承静电监测  

Electrostatic monitoring of rolling bearings based on short-time Fourier transform and cepstrum

作　　者：刘若晨 顾双双 孙见忠[2] 左洪福[2] 贝绍轶[1] LIU Ruochen;GU Shuangshuang;SUN Jianzhong;ZUO Hongfu;BEI Shaoyi(School of Automobile and Traffic Engineering,Jiangsu University of Technology,Changzhou Jiangsu 213001,China;College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)

机构地区：[1]江苏理工学院汽车与交通工程学院,江苏常州213001 [2]南京航空航天大学民航学院,南京211106

出　　处：《航空动力学报》2024年第9期312-319,共8页Journal of Aerospace Power

基　　金：国家自然科学基金(51705221,91860139,52072176);江苏理工学院研究生实践创新计划项目(XSJCX20_44);“江苏省政府留学奖学金”资助项目。

摘　　要：针对滚动轴承在常规监测方法下耦合多激励源问题,引入静电监测技术,提出了基于短时傅里叶变换及其倒频谱的故障特征提取方法。设计搭建了滚动轴承静电监测试验平台,采集正常和故障情况下滚动轴承的静电、振动信号。从静电信号的时域、频域和时频域角度对比研究,证明了使用时频分析联合倒频谱的方法能准确提取出与实际轴承故障位置相匹配的特征值;对比同步振动信号的故障特征,展现了静电信号低频特征突出、高频衰减快的特点。试验结果表明:滚动轴承发生早期磨损故障后会伴随强烈的静电现象产生。对静电信号进行短时傅里叶变换及倒频谱分析,能有效去除高频激励源,凸显出低频段内轴承故障特征。相比于振动检测,静电检测采集的信号源能较为直接地反映轴承故障信息,为设备故障诊断提供了一种思路。In view of the problem of coupling multiple excitation sources of rolling bearing under conventional monitoring methods,the electrostatic monitoring technology was introduced,and a fault feature extraction method based on short-time Fourier transform and cepstrum was proposed.A test platform for electrostatic monitoring of rolling bearing was designed and built to collect the electrostatic,vibration signals of rolling bearing under normal and fault conditions;from the perspective of time domain,frequency domain and time-frequency domain,it was proved that the method of time-frequency analysis combined with cepstrum can accurately extract the eigenvalues matching with the actual bearing fault location;by comparing with the fault characteristics of synchronous vibration signal,the low-frequency characteristics of electrostatic signals were prominent and the high-frequency attenuation was fast.The test results showed that the early wear failure of rolling bearing could be accompanied by strong electrostatic phenomenon.The short-time Fourier transform and cepstrum analysis of electrostatic signals can effectively remove the high-frequency excitation sources and highlight the bearing fault characteristics in lowfrequency.Compared with vibration detection,the signal source collected by electrostatic detection can reflect the bearing fault information more directly,providing an idea for equipment fault diagnosis.

关 键 词：静电监测 滚动轴承 短时傅里叶变换 倒频谱 试验验证 

分 类 号：V216.3[航空宇航科学与技术—航空宇航推进理论与工程] TH133.33[机械工程—机械制造及自动化]