Intelligent bearing fault diagnosis using PCA–DBN framework

Abstract

This paper studies the fault diagnosis problem for rolling element bearings. By casting the bearing fault diagnosis as a class of pattern classification problem, we propose a novel intelligent fault diagnosis approach based on principal component analysis (PCA) and deep belief network (DBN). The dimension of raw bearing vibration signals is reduced by adopting PCA method, which consequently extracts the fault signatures in terms of primary eigenvalues and eigenvectors. The modified samples are subsequently trained and tested by the DBN for fault classification and diagnosis. The distinctive feature of our approach is that it requires no complex signal processing on raw vibration data, rendering it easily achievable and widely applicable. The experimental results indicate the effectiveness of the proposed PCA–DBN fault diagnosis scheme compared with other methods.