A New RBF Neural Network Training Algorithm Based on PSO

摘要

Based on the study of Radial Basis Function (RBF) neural network training algorithm and Particle Swarm Optimization (PSO) algorithm, a new RBF neural network training algorithm with modified PSO algorithm is formulated, in which a control gene is introduced into basis PSO algorithm. The algorithm can determine network structure and parameters, such as centers and widths of hidden units by combining with least square method. The new training algorithm is applied to the nonlinear system identification problem, comparing with hierachical genetic algorithm and orthogonal least squares algorithm (OLS), the simulation results illustrate its efficiency.