基于局部均值分解多尺度模糊熵和灰色相似关联度的滚动轴承故障诊断

Rolling Bearing Fault Diagnosis Based on Local Mean Decomposition Multi-scale Fuzzy Entropy and Grey Similar Incidence

提出了一种基于局部均值分解多尺度模糊熵和灰色相似关联度相结合的滚动轴承故障诊断方法。该方法将故障信号自适应地分解为若干乘积函数,并从中选取包含主要故障信息的PF分量计算多尺度模糊熵作为特征向量,通过计算待识别样本与标准故障模式的灰色相似关联度,对滚动轴承故障类型和损伤程度进行判断。将该方法与LMD模糊熵和灰色相似关联度相结合的方法进行了对比,实验表明,基于LMD多尺度模糊熵和灰色相似关联度的滚动轴承故障诊断方法,能够有效地识别滚动轴承运行状态,实现对滚动轴承的故障诊断。

A rolling bearing fault diagnosis method based on local mean decomposition（LMD） of multi-scale fuzzy entropy and grey similar incidence is discussed. In this method, the fault signal is decomposed into several product functions （PF） adaptively, and the multi-scale fuzzy entropies of the PF components covering contain main fault information, which is calculated to get the fault feature vectors. By calculating the grey similar incidence of the sample to be identified and the standard fault pattern, it is realized that the judgement of rolling bearing fault types and damage degree. Compared with the method based on LMD fuzzy entropy and grey similar incidence, the experimental results show that the method based LMD multi-scale fuzzy entropy and grey similar incidence can identify rolling bearing running state efficiently and realize rolling bearing fault diagnosis.