Solving Parameter Identification Problem by Hybrid Particle Swarm Optimization

Ordinary differential equations have been a useful tool for describing the behavior of wide variety of dynamic physical systems. In this study, a method for solving parameter identification problem for ordinary second order differential equations using hybrid Nelder-Mead simplex search and particle swarm optimization (NM-PSO) approach is presented. Experiments using two case problems are presented and compared with the best known solutions reported in the literature. The comparison results demonstrate that NM-PSO produced better estimated results with respect to previous findings from particle swarm optimization and genetic algorithm.