An improved parallel genetic algorithm based on injection island approach and k<inf>1</inf> triangulation for the optimal design of the flexible multi-body model vehicle suspensions

Abstract

In this paper, we introduced IPGA (Improved Parallel Genetic Algorithm), a new parallel genetic algorithm which is based on k 1 triangulation of fixed point theory and an injection island approach. This approach guarantees that the processors only generate solutions within their assigned region. And what's more, the k 1 triangulation theory is introduced to the genetic algorithm to design convergence criteria objectively and to solve the convergence problem in the later period. The running environment is constituted in this paper and the optimization of multi-body model vehicle suspension is implemented through establishing the interface between ADAMS software and genetic algorithm. The results show that the parallel genetic algorithm developed in this paper is efficient.