基于自适应VMD和MOMEDA的变转速轴承早期故障诊断

Early Fault Diagnosis of Variable Speed Bearing Based on Adaptive VMD and MOMEDA

针对变转速工况滚动轴承早期故障特征微弱且难以提取的问题,提出一种基于自适应变分模态分解(VMD)和多点最优最小熵反褶积修正(MOMEDA)的故障诊断方法.对振动信号进行角度重采样,将信号转化为角域平稳信号;利用中心阶次确定VMD分解层数,对角度重采样信号进行分解;基于相关系数指标选取敏感的本征模态函数(IMF)分量,运用MOMEDA算法对该分量进行特征增强,实现微弱故障的提取.仿真信号和轴承实测信号的分析结果表明:该方法能成功提取变转速下轴承早期故障的微弱特征.

Aiming at the problem of weak early fault characteristics and difficult to extract in variable speed rolling bearings,a fault diagnosis method based on adaptive Variational Mode Decomposition(VMD)and Multipoint Optimal Minimum Entropy Deconvolution Adjusted(MOMEDA)is proposed.Angle resampling is used to transform the vibration signal into angular stationary signal;the VMD decomposition level is determined by the central order,and the angle resampling signal is decomposed;the sensitive Intrinsic Mode Function(IMF)component is selected based on the correlation coefficient index,and the feature enhancement is carried out by using MOMEDA algorithm to extract weak faults.The analysis results of simulation signals and bearing measured signals show that this method can successfully extract the weak features of bearing early fault at variable speed.