Abstract
Function optimization problems have been studied and applied in many fields. For getting good solutions of these problems well, many optimization algorithms have been developed. In this paper, we focus on the swarm intelligence methodologies and discuss on the characteristic of them. We make a hybrid algorithm in which there are different swarm models. And we consider a hybrid model of PSO and FA with particles character will change in its search process. Several strategies of character change rules are presented. Then through testing the algorithm with popular benchmark problems, it indicates that the proposed hybrid algorithm has well properties for various function optimization.

This publication has 11 references indexed in Scilit: