APPLICATION OF LMD-AR MODEL AND DE-SVM TO DIAGNOSIS ROLLER BEARING FAULT

Abstract

This study investigates a new method for roller bearing fault diagnosis based on product functions (PFs) and Autoregressive (AR) model, together with a Support Vector Machine designed using a Differential Evolution (DE) Algorithm, referred to as a DE-SVM. First, the original acceleration vibration signals of roller bearings are decomposed into PFs by using local mean decomposition (LMD) method. Second, the concept of AR model is introduced. Third, AR model is used to extract PFs into feature vectors and served as input vectors for the support vector machine classifier. Finally, the DE-SVM classifiers are proposed to recognize the faulty roller bearing pattern. The experimental analysis results show that the proposed method can classify working condition of roller bearings with higher classification accuracy and lower cost time compared to the other methods.