The research of fuzzy neural network based on particle swarm optimization

With the science and technology develop constantly, the research of swarm intelligence has aroused great concern of many scholars, including particle swarm optimization(PSO) algorithm for its advantages of fast convergence, easy to implement, and only a few parameters need to be adjusted are widely used. In this paper, the fuzzy neural network(FNN) based on PSO algorithm have been studied and tested, and applied it to diagnosis the model for fault of gearbox, which is seven inputs and six outputs, and compared the results of diagnosis in the way of fuzzy neural network based on particle swarm optimization(PSO-FNN) with the way of FNN. Obtaining the conclusion that PSO-FNN has better training performance, fast convergence and fewer iterations, better accuracy, good rate of fault identification.