Research On Fault Diagnosis Of Rolling Bearing Based On VMD And Improved VPMCD

The advent of the Industry 4.0 era makes mechanical equipment continuously developing toward the trend of intelligence,integration and precision.Increasing reliability requirements for mechanical equipment promoted the reformation and innovation of diagnosis technology.Rolling bearing as one of the most malfunctioning components in rotating machinery equipment,to monitor its operation status and fault diagnosis is of great significance.How to present bearing fault features and identify fault types effectively under the background of strong noise has always been a difficult problem in this field.This article will investigate bearing fault signal via feature extraction and pattern recognition,so as to realize intelligent diagnosis of rolling bearing fault.The research content of the paper can be summarized as follows:(1)Based on the study of the application mechanism of Variant Mode Decomposition(VMD)method in signal processing,which is used in the analysis of bearing fault signals.Through simulation experiments,the experimental results show that the VMD method can effectively avoid the problems of decomposing instability and end-effects in Empirical Mode Decomposition(EMD),improving the effect of signal decomposition,noise suppression,and noise robustness.(2)A feature extraction method based on VMD and morphological difference filtering is proposed.Firstly,some intrinsic mode functions(IMFs)are decomposed by the VMD method.The kurtosis criterion is used to calculate the IMF with the largest kurtosis.Then,the morphological difference filter is used for analysis to filter out the noise interference;The extracted signal was subjected to Hilbert envelope analysis.The experimental results show that this method can effectively extract the characteristic frequency of the rolling bearing failure.(3)In order to solve the problem of low recognition rate of bearing fault type under compound fault,this paper proposes a VMD-morphological difference filter and an improved Variable Predictive Model based Class Discriminate(VPMCD)bearing,which was applied to the analysis of the measured signal of the bearing,and the practicability of the proposed method was verified.Compared with the VPMCD method,this method effectively improves the bearing fault recognition rate and provides a new fault diagnosis method.