ELM Neural Network-based Fault Diagnosis Method for Mechanical Equipment

摘要

To improve the safety of industrial machinery and equipment, a fault diagnosis method of mechanical equipment based on ELM neural network is proposed, which is a learning machine with many merits such as fast learning and strongly generalizing ability, and it can be well used in fault classification and diagnosis. This paper presents the weights training method of ELM neural network and the steps for fault diagnosis. Besides, the classification effects of ELM neural network and traditional neural network based on the gradient descending algorithm are compared. The simulation experiment shows that the ELM neural network has the advantage of fast learning speed and high studying accuracy. It can diagnose the fault category quickly and accurately.