基于粒子多样性研究的改进PSO算法

Improved PSO Algorithm Based on Diversity of Particle Swarm

从研究粒子群多样性影响PSO算法最优适应值进化的角度出发,结合目前已取得的惯性权值非线性动态自适应调节的研究成果,给出了一种带"精英集团"策略和变异操作的改进PSO算法。对几个高维典型函数的最优化解的测试结果表明,改进算法同时具备较强的全局探索能力和局部开发能力,能够在保证算法较快收敛的前提下,有效地提高最优化解的精度。

An improved particle swarm optimization （PSO） algorithm was proposed based on the study of the relation between particles ＇diversity and the best fitness value during the whole process of iterations. A strategy of ＂＇elite group ＇＂ and mutation operation were adopted to improve the optimization algorithm. The fruit of nonlinear inertia weight variation for dynamic adaptation in PSO was used for reference. The results of several simulations for different high dimension benchmark functions illustrate that the proposed algorithm has both better ability of global exploration and local exploitation, and the optimal precision is effectively improved under the condition of guaranteeing the convergence of PSO algorithm.