Fault Diagnosis of Rolling Bearing Based on BSA Neural Network

Abstract

In the fault diagnosis programs, the intelligent algorithms, such as BP neural network, genetic neural network, ant colony neural network and so on, always suffer from the slow training speed and the local minimums. In this paper, a backtracking search optimization algorithm (BSA) based neural network was put forward, which used the BSA algorithm to train the neural network weights and thresholds. And then it was utilized on the pattern recognition of rolling bearing faults, and the results show that BSA neural network can better solve the problems of slow convergence and local minimums, which has a good application value.