A Bearing Fault Diagnosis Method Based on VMD-HPE

Since rolling bearings operate in complex and harsh conditions with high speed and heavy load for a long time, their fault signals have the problems of difficulty in feature extraction and low diagnostic accuracy. Therefore, a rolling bearing fault diagnosis method based on variational mode decomposition (VMD) and hierarchical permutation entropy (HPE) is proposed in this paper. Firstly, the fault signals of rolling bearings are decomposed by variational mode decomposition. Secondly, several node signals are obtained after hierarchical decomposition, and the permutation entropy value of the obtained node signals is calculated as the feature vector. Thirdly, a multi-fault classifier based on Bayes is established to realize the fault diagnosis of rolling bearings. Finally, the method is applied to the data of Bearing Center of Case Western Reserve University, and the experimental results verify the effectiveness of the method.