基于MED和自适应VMD的行星齿轮箱故障诊断方法

Fault diagnosis of planetary gearbox based on minimum entropy deconvolution and adaptive variational mode decomposition

为解决变分模态分解(VMD)在行星齿轮箱故障特征频率提取过程出现的鲁棒性低及分解个数不确定的问题,提出一种基于最小熵反褶积(MED)和自适应变分模态分解(AVMD)的齿轮箱故障诊断方法.首先通过MED对信号进行降噪,突出故障信号特征;采用瞬时频率的新定义及变差概念,自适应选择VMD的级数;使用VMD方法将行星齿轮箱的断齿故障信号分解为若干个本征模态函数(IMF)分量;根据相关系数分析选取带有故障信号的IMF分量,对其进行包络谱分析,以提取故障特征频率.仿真信号和试验信号分析结果表明,使用MED去噪后信号的峰值信噪比提高了10%,解决了传统VMD个数经验选择出现的误差问题,从而实现此过程自适应化,解决了VMD在强噪声下针对非线性非平稳信号鲁棒性低的问题,准确提取了风电齿轮箱的故障特征频率.

To solve the problem of low robustness and uncertain decomposition number of variational mode decomposition(VMD)in the fault feature frequency extraction process of planetary gearbox,a gearbox fault diagnosis method based on minimum entropy deconvolution(MED)and adaptive variational mode decomposition(AVMD)was proposed.First,the signal was denoised by the MED to highlight the fault signal characteristics.By using the new definition of the instantaneous frequency and the concept of variation,the series of VMD was adaptively selected.The VMD method was used to decompose the fault signal of the planet gearbox into several intrinsic modal function(IMF)components.According to the analysis of correlation coefficient,the component of IMF with the fault signal was selected to envelope spectrum analysis to extract the fault characteristic frequency.The analysis results of the simulation signal and the experimental signal show that the peak signal-to-noise ratio of the signal is increased by 10%after denoising by using the MED,the error problems of the empirical selection of the number of VMD are solved and the process is self-adaptive.The problem of the low robustness of nonlinear non-stationary signals in the case of the strong noise is solved.The fault characteristic frequency of the wind power gearbox is accurately extracted.