基于VMD和MED的滚动轴承微弱故障特征提取

Weak fault feature extration of rolling bearing based on VMD and MED

针对强噪声环境下滚动轴承故障特征信息非常微弱且难以提取的问题,提出基于变分模态分解(Variational Mode Decomposition,VMD)和最小熵解卷积(Minimum Entropy Deconvolution,MED)的滚动轴承微弱故障特征提取方法。基于VMD和MED的滚动轴承微弱故障特征提取方法首先采用VMD对滚动轴承故障信号进行分解,得到多个模态分量,由于噪声的干扰,很难从各个模态分量中提取有效的故障特征信息;然后根据相关系数准则,选取与原始信号相关系数较大的模态分量进行重构,再对重构后的信号进行MED降噪处理;最后对降噪处理后的信号进行Hilbert包络解调,从得出的包络谱中即可准确地提取到故障特征信息。轴承故障实验信号处理结果表明,该方法可以有效地降低噪声的影响,精确地提取滚动轴承微弱的故障特征信息。

In the strong noise environment, the fault feature information of rolling bearing is very weak and difficult to be extrac- ted,propose the Variational Modal Decomposition （VMD） and the Minimum Entropy Deconvolution （MED） method to extract fault feature of rolling bearing. Firstly, VMD was used to decompose original fault signals into several components. Due to influ- ence of noise, it is difficult to extract the effective fault feature information from the component. According to the correlation coeffi- cient criterion, selecting the bigger correlation coefficient of component with the original signal to reconstruct, and then the recon- structed signal is processed by the Minimum Entropy Deconvolution（MED）. Finally, the processed signal is analyzed by Hilbert envelope. The fault characteristic frequency can be extracted accurately from the envelope spectrum through the analysis of the experimental data, the results show that the method can effectively reduce the influence of the noise, and accurately realize the ex- traction of bearing fault feature information.