The recognition of train wheel tread damages based on PSO-RBFNN algorithm

Abstract

In order to the recognition of the train wheel tread damages, the pattern recognition method of the train wheel tread damages based on PSO-RBFNN was developed. The algorithm uses PSO-RBFNN algorithm to optimize center and spread of RBFNN, the connection weight value is sovled by least squares method. Compared with the traditional RBFNN,BP and GA-RBFNN, the experiment results show that the recognition rate of testing samples is higher than the traditional RBFNN, BP and GA-RBFNN, the evolutional generations of PSO-RBFNN algorithm were less than RBFNN, BP and GA-RBFNN.