Regrouping Particle Swarm Optimization-Based Neural Network for Bearing Fault Diagnosis

Abstract

This paper proposes a regrouping particle swarm optimization-based neural network (RegPSONN) for rolling bearing fault diagnosis. The proposed method applied neural network for rolling bearing conditions classification, and regrouping particle swarm optimization (RegPSO) is utilized for network training, and ten time-domain feature parameters are selected to establish the input vector. To evaluate the performance of RegPSONN, bearing vibration data are used for verification. In addition, the back propagation neural network (BPNN), genetic algorithmbased neural network (GANN) and particle swarm optimization neural network (PSONN) are used to classify the bearing data for algorithm comparison. Experimental results demonstrated that the proposed method was superior to other methods considering the classification rate.