一种信息点多样性的粒子群优化算法

针对PSO在寻优后期尤其在高维搜索空间中无法得到满意结果,提出了一种信息点多样性的改进粒子群优化算法。粒子个体最优位置及全局粒子最优位置是两个有用的精确的信息点,而PSO的信息交互方式正依赖于这两个信息点,从多样性方面考虑,将该有用的信息点增加为粒子个体最优位置附近随机的一点。实验仿真结果表明,新算法的全局搜索能力、收敛速度、精度和稳定性均有了显著提高。

Particle Swarm Optimization Algorithm with Information Point Diversity

A modified particle swarm optimization based on information point diversity is proposed for the fact that no satisfactory results can be reached during later period of PSO optimization,especially,in high-dimensional search space.The best position of individual particle and the global particle are two useful accurate information points.The PSO method of information exchange is dependent on these two information points.From the viewpoint of diversity,the useful information is changed onto a fuzzy point which is randomly generated in the useful range.Experimental simulation shows that the new algorithm global search ability,convergence rate,accuracy and stability are improved significantly.