Networked synchronization control method by the combination of RBF neural network and genetic algorithm

Abstract

Generally, networked synchronization control system is defined as the system which manages and controls the behavior of multi-devices or multi-systems synchronously to realize their synchronous work. However, in the RBF neural network, the center of RBF, the width of RBF, output weight of RBFNN have a great influence on control ability of RBF neural network, so in order to gain RBF neural network with good control ability, the three parameters need to be determined. In the study, genetic algorithm is applied to determine the parameters of RBF neural network. Thus, the combination method of RBF neural network and genetic algorithm is applied to the networked synchronization control. We employ response curve of phase step to testify the synchronization control performance of the combination method of genetic algorithm and RBF neural network. PID controller is used to compare with the proposed genetic algorithm and RBF neural network controller. It is indicated that the networked synchronization control result by GA-RBF neural network controller is better than that by PID controller.