峭度准则EMD与空域相关结合的滚动轴承故障特征提取  

Feature Extraction of Rolling Bearing Using EMD and Spatial Correlation Based on Kurtosis Criterion

作　　者：艾延廷[1] 方妍[1] 田晶[1] AI Yan-ting;FANG Yan;TIAN Jing(Liaoning Key Laboratory of Advanced Measurement and Test Technology for Aircraft Propulsion Systems,Shenyang Aerospace University,Liaoning Shenyang 110136,China)

机构地区：[1]沈阳航空航天大学辽宁省航空推进系统先进测试技术重点实验室

出　　处：《机械设计与制造》2019年第12期213-216,共4页Machinery Design & Manufacture

基　　金：国家自然科学基金(51675351);中航工业产学研(cxy2012sh17)

摘　　要：针对滚动轴承故障信号非线性、非平稳的特点,提出一种经验模态分解(Empirical Mode Decomposition,EMD)与空域相关相结合的信号特征提取方法。首先,利用EMD方法将振动信号分解成若干个固有模态分量(Intrinsic Mode Function,IMF);然后采用峭度准则选取能够反应故障特征的IMF分量进行重构,再对重构信号运用空域相关法进行降噪;最后将处理后的振动信号进行Hilbert包络谱分析,提取出轴承的故障特征。采用所建立的方法分析轴承外圈故障的实验数据。结果表明,峭度准则EMD与空域相关相结合的方法能够对振动信号进行降噪处理并有效地提取出轴承外圈故障特征频率。Aiming at the non-linear and non-stationary characteristics of rolling bearing fault signal,a method of signal feature extraction based on empirical mode decomposition(EMD)and Spatial correlation is proposed.Firstly,EMD is used to decompose the vibration signal into several Intrinsic Mode Functions(IMF).Then we use the kurtosis criterion to select the IMF component which can reflect the fault feature and reconstruct the signal using the spatial correlation method to reduce the noise.Finally,the vibration signal of the noise reduction is analyzed by Hilbert envelope,and the fault characteristics of the rolling bearing are extracted.The experimental data of the bearing outer ring are analyzed by the established method.The results show that the combination of EMD and Spatial correlation can reduce noise of vibration signal and effectively extract the characteristic frequency of bearing outer ring fault.

关 键 词：经验模态分解 空域相关 峭度 包络谱 故障诊断 

分 类 号：TH16[机械工程—机械制造及自动化] TP206[自动化与计算机技术—检测技术与自动化装置]