A Physics-Based Deep Learning Approach for Fault Diagnosis of Rotating Machinery

Abstract

During the recent few years, deep learning has been recognized as a useful tool in condition monitoring and fault detection of rotating machinery. Although existing deep learning approaches to fault diagnosis are able to intelligently detect and classify the faults in rotating machinery, most of these approaches rely exclusive on data and thus are purely data-driven, and do not incorporate physical knowledge into the learning and prediction processes. To address this challenge, this study proposes a novel approach, namely physics-based convolutional neural network (PCNN), for fault diagnosis of rotating machinery and targets fault detection of rolling-element bearings as a special application of PCNN. In the proposed approach, an exclusively data-driven deep learning approach, called convolutional neural network, is carefully modified to incorporate useful information from physical knowledge about bearings and their fault characteristics. The performance of the proposed PCNN approach in machinery fault diagnosis is compared with that of traditional machine learning- and deep learning-based approaches reported in the literature.