AN ADAPTIVE APPROACH TO ON-LINE IDENTIFICATION OF DYNAMIC SYSTEM BASED ON RBF NEURAL NETWORKS

In this paper, a neural network based on line adaptive identification model of dynamic system is developed. In order to improve the performance of model, an on line adaptive correcting method of model parameters is proposed based on RBF network's linear outputs. A great amount of simulations in this investigation demonstrates that this identification model and on line adaptive correcting method are of generalization and applicability.