快速自适应局部均值分解及轴承故障诊断应用

提出了一种新的非平稳信号处理方法——快速自适应局部均值分解(Fast and Adaptive Local Mean Decomposition,FALMD)。采用顺序统计滤波器求取信号上下包络线的均值来获得局部均值函数及包络估计函数,然后将信号分解为若干乘积函数(Product Function,PF)分量及一个残余分量。该算法一方面改变了局部均值分解(Local Mean Decomposition,LMD)严格的终止条件,提高了运算速率,另一方面减少了对极值点的依赖,在一定程度上抑制了端点效应。仿真信号和实验信号分析证明了该方法在非平稳信号自适应分解中的有效性,成功地提取出了滚动轴承的故障特征。

Fast and adaptive local mean decomposition method and its application in rolling bearing fault diagnosis

Local Mean Decomposition(LMD)has been widely used in signal processing and fault diagnosis of mechanical equipment,but there are still some issues that need to be improved,such as endpoint effects,mode mixing,and so on.This study proposes a novel method called Fast and Adaptive Local Mean Decomposition(FALMD),which can decompose the signal into some product function components(PF)and a residual component,just like LMD.This method changes the termination criterion,and the speed of operation is accelerated;order statistics filters are used to weaken the endpoint effect.Using this method,the simulation and experimental signal can be decomposed quickly,effectively and accurately.This method can be applied to the fault diagnosis of bearing.