基于ELMD和灰色相似关联度的滚动轴承故障诊断研究

A Fault Diagnosis Research on Rolling Bearing Based on ELMD and Gray Similarity Degree

针对滚动轴承发生故障时,其故障信号不同频带的能量分布变化与其故障状态之间存在一定关系的情况,提出了一种总体局部均值分解(Ensemble Local Mean Decomposition, ELMD)与灰色相似关联度相结合的滚动轴承故障诊断方法.首先,选取滚动轴承处于不同故障状态的样本信号,利用ELMD对其进行分解并得到若干乘积函数(Product Function, PF);然后,计算每个PF分量的能量分布并构造特征向量;最后,结合灰色相似关联度对滚动轴承故障状态进行分析和识别,并与LMD(Local Mean Decomposition, LMD)和EEMD(Ensemble Empirical Mode Decomposition)方法进行对比,其实验对比分析结果论证了方法的可行性,也为滚动轴承的故障诊断提供了新的解决方案.

Aiming at the certain relationship between energy distribution changes in different frequency bands of fault signals and fault state when failures happen in the rolling bearings,a fault diagnosis method based on ensemble local mean decomposition(ELMD)and grey similarity degree is proposed.First,the sample signals of rolling bearing in different fault states were selected and decomposed into several product functions(PF)through ELMD;then,the energy distribution of each PF component was calculated and feature vectors were constructed;finally,grey similarity degree was combined with ELMD to realize the analysis and recognition of fault status,and ELMD was compared with local mean decomposition(LMD)and ensemble empirical mode decomposition(EEMD).The experimental results show the feasibility of the method and provide a new solution for fault diagnosis of rolling bearings.