基于局部均值分解与形态学分形维数的滚动轴承故障诊断方法

Roller bearing fault diagnosis based on local mean decomposition and morphological fractal dimension

针对滚动轴承振动信号通常具有非线性与低信噪比特点,提出基于局部均值分解(Local Mean Decomposition,LMD)与形态学分形维数的滚动轴承故障诊断方法。采用LMD将滚动轴承振动信号分解为若干个乘积函数(Product Function,PF)分量,计算包含有滚动轴承故障特征的PF分量形态学分形维数,并将其用作特征量判断滚动轴承工作状态及故障类型。实验分析结果表明,该方法能有效用于滚动轴承的故障诊断。

The vibration signals of roller bearings usually appear with the specialities of nonlinearity and low signal to noise ratio. In view of this problem, a roller bearing fault diagnosis method based on local mean decomposition (LMD) and morphological fractal dimension was proposed. In the method, the roller bearing vibration signal was decomposed into a set of product functions (PFs) by LMD, and then the morphological fractal dimension of PF which contain the roller bearing fault characteristics was calculated, and adopted as a characteristic parameter to judge the roller bearing working conditions and fault types. The experimental results indicate that the method can be applied to the roller bearing fault diagnosis effectively.