基于MED-EMD和切片双谱的滚动轴承故障特征提取

Fault Feature Extraction of Rolling Bearings Based on MED-EMD and Slice Bispectrum

针对风电齿轮箱中轴承故障信号非线性、非平稳的特点,提出了基于最小熵反褶积(MED)、经验模态分解(EMD)和切片双谱相结合的方法来提取轴承的微弱故障特征。通过MED-EMD将原始信号降噪分解为多个本征模态函数(IMF),对与原始信号相关性强的IMF分量进行切片双谱分析,从而提取微弱故障特征频率。对仿真信号和风电齿轮箱轴承实测信号的分析表明:选取MED作为EMD的前置滤波器能够弥补强背景噪声下EMD分解的不足,切片双谱分析能够抑制高斯噪声,提高信噪比,得到了风电齿轮箱故障产生于中间齿轮轴电动机侧轴承内圈点蚀的正确判断。

The fault signal of wind turbine gearbox bearings has characteristics of nonlinear and non - stationary, the combination method is proposed based on minimum entropy deconvolution （MED）, empirical mode decomposition （EMD） and slice bispeetrum to extract weak fault feature of the bearings. The original signal denoising by MED - EMD is decomposed into several intrinsic mode functions （IMF） , and the IMF components having strong correlation with o- riginal signal are analyzed with slice bispectrum to extract weak fault feature frequency. The analysis of simulation sig- nal and measured signal of wind turbine gearbox bearings shows that the MED is selected as prefilter of EMD, and the shortcomings of EMD decomposition are compensated under strong background noise. The Gaussian noise is suppressed and the signal- to -noise ratio is improved by slice bispectrum analysis. The results verify the correct judgment that the fault of wind turbine gearbox comes from inner ring pitting of the bearings in intermediate gear shaft on motor side.