Multiple Populations Immune Mechanism Based Particle Swarm Optimizer for Multi-modal Problems

摘要

Particle swarm optimization has been proved efficient in solving optimization problems. When aims at multi-modal problems, it is usually trapped in local optima. In this paper, a novel multiple populations immune mechanism is proposed to guarantee particles escape identical local optima. Particles are grouped into sub-population according to their immune matching value, so similar particles are resemble to local best of own sub-population. Immune mechanisms including matching, selection and clone could improve population diversity and convergence rate as expected. Finally classic multi-modal benchmark problems are list to testify MPI-PSO's performances. The empirical results demonstrate that our approach obtains more global optima per run compared to the well-known PSOs. The experiments also suggest that using multiple models reduces the generations spent to reach convergence.