基于伪故障信号的齿轮箱故障诊断方法

Fault diagnosis method of gearbox based on pseudo-fault signal

针对经验模态分解过程中产生模态混叠导致难以准确提取故障频率问题,提出一种基于伪故障信号(PFS)经验模态分解(EMD)的齿轮箱故障诊断的PFS-EMD方法。该方法先将原始信号进行经验模态分解;再用能量法与互相关准则选取有效IMF(intrinsic mode function)分量;然后从有效IMF分量中选择含有伪故障频率成分的IMF分量(简称PI分量),构建伪故障信号并与PI分量进行重构,对重构后的信号进行EMD分解得到第一阶IMF分量作为最终的IMF分量。对最终的IMF分量进行频谱分析从而隔离和提取故障信息。仿真实验表明,该方法不仅保留了EMD的良好特性,能准确获得信号不同尺度的局部特征信息,而且较好地解决了EMD的模态混叠问题。

To solve the problem of inaccurate extracting of fault frequency because of mode mixing in the process of empirical mode decomposition （ EMD）, a novel fault diagnosis method for gearbox based on PFS-EMD is proposed. First, EMD is applied to the original signal to select the effective intrinsic mode function （IMF） component by using the energy method and the cross correlation criterion. PI component which contains pseudo fault frequency is selected from the effective IMF to construct PFS and reconstruct with the PI component, then the first IMF is obtained by EMD as the final component of IMF. The fault information is isolated and extracted with spectrum analysis of the final IMF. Simulation results show that the proposed method not only retains the good performance of the EMD and can accurately obtain the local feature information of the signal at different scales, but also can solve the mode mixing better.