突然変異を適用した粒子群最適化の探索性能に関する研究

抄録

Particle Swarm Optimization (PSO) is one of the metaheuristic techniques. It is widely used in various fields. It is excellent in convergence of solutions. However it has a tendency to easily drop into local optima and is difficult to get out of them. Mutation operators used in Genetic Algorithm (GA) are effective in avoiding local optima and in maintaining diversity. Recently some researchers try to incorporate mutation operators to PSO to improve its search performance. In this paper, the influence that the amount of displacements caused by mutation operators gives the search performance is investigated. Five test functions are taken up to evaluate the performance.