State Monitoring and Early Fault Diagnosis of Rolling Bearing based on Wavelet Energy Entropy and LS-SVM

Abstract

Rolling bearing is one of the most widely used elements in rotary machines. In this paper, a novel method is proposed to extract early fault features and diagnosis the early fault accurately for rolling bearing. Wavelet Energy Entropy is introduced as a feature parameter for bearing state monitoring and least square support vector machine (LS-SVM) is used for early fault diagnosis. In order to test the effectiveness of the method, a series of bearing whole life cycle test are performed on the accelerated bearing life tester. The result shows that Wavelet Energy Entropy has better performance and can forecast fault development earlier compared to conventional signal features. LS-SVM method can distinguish early bearing fault modes more accurate and faster than classic pattern recognition methods.