Wavelet Neural Network Based on Modified PSO and Its Application in Fault Diagnosis of Gearbox

摘要

A modified particle swarm optimization -- compound model PSO with stochastic inertia weigh is put forward and used to optimize the parameters of wavelet neural network. The trained wavelet neural-network is applied to the fault diagnosis experiment of gearbox. The experimental result indicates that the wavelet neural-network training method based on the modified PSO is effective. This is an available approach to solve the problems on condition monitoring and fault diagnosis.