Fault Diagnosis of Rolling Bearing Based on the Permutation Entropy of VMD and Decision Tree

Abstract

Rolling bearing is the most important component of rotating machinery. The research of the fault diagnosis of rolling bearing can effectively improve the running reliability of rotating machinery, avoid sudden failures and ensure the safety of workers. Because of the characteristics with non-stationary and nonlinear of the rolling bearing vibration signal, a fault diagnosis method for rolling bearing based on permutation entropy of VMD and decision tree is proposed. The original vibration signal is decomposed by VMD to obtain a series of BIMFs. And then, the permutation entropy value of each BIMF is extracted as the fault feature. Finally, the decision tree is used to predict the fault category of rolling bearing. The experimental results show that the propose method can accurately diagnose rolling bearing faults, and it is an effective method for fault diagnosis of rolling bearings.