基于VMD和1.5维Teager能量谱的滚动轴承故障特征提取  

Rolling bearing fault feature extraction based on the VMD and 1.5-dimensional Teager energy spectrum

作　　者：向玲[1] 张力佳[1] 

机构地区：[1]华北电力大学机械工程系,河北保定071003

出　　处：《振动与冲击》2017年第18期98-104,124,共8页Journal of Vibration and Shock

基　　金：国家自然科学基金资助项目(51675178;51475164);河北省自然基金资助项目(E2013502226)

摘　　要：为准确提取非线性、非平稳的滚动轴承故障信号中的故障特征,提出基于变分模式分解(Variational Mode Decomposition,VMD)和1.5维Teager能量谱的滚动轴承故障特征提取方法;变分模式分解(VMD)是一种新的信号自适应分解方法,1.5维Teager能量谱具有1.5维谱良好的降噪效果和Teager能量算子强化信号瞬态冲击的优点。故障特征提取过程:首先,对滚动轴承故障信号进行VMD分解得到一组分量,根据峭度-相关系数准则筛选出2个冲击特征明显分量进行信号重构;再次,对重构信号进行1.5维Teager能量谱分析;最后根据能量谱图的分析,提取出滚动轴承的内圈和滚动体故障特征。仿真信号和试验信号的分析都验证了提出方法的有效性;通过与EEMD分解比较,采用VMD变分模式分解和1.5维Teager能量谱的分析方法更具有区分性,可以有效识别滚动轴承的故障特征。Rolling bearing fault signals are usually nonlinear and non-stationary. A method combining the variational mode decomposition（ VMD） with the 1. 5-dimensional Teager energy spectrum,was proposed for the purpose of rolling bearing fault diagnosis. The VMD is a new method for adaptive signal decomposition,and the 1. 5-dimensional Teager energy spectrum not only has the advantage of 1. 5-dimensional denoising,but also strengthens transient impact feature signals by using the Teager operator. Using the VMD,a rolling bearing fault signal was decomposed into a set of components,among which two components,having obvious impact features,were sieved out for the signal reconstruction according to the kurtosis-correlation coefficient criteria. The reconstructed signal was analyzed using the 1. 5-dimensional Teager energy spectrum. Based on the energy spectrum analysis of the reconstructed signal,the inner ring and rolling element fault features were extracted. The analysis of simulated signals and test signals verifies the effectiveness of the proposed method. Compared with the ensemble empirical mode decomposition, the proposed method is of higher distinctiveness and can effectively identify fault features of rolling bearings.

关 键 词：变分模式分解 1.5维Teager能量谱 特征提取 故障诊断 滚动轴承 

分 类 号：TH133.3[机械工程—机械制造及自动化] TH17