均匀搜索粒子群算法

摘要

针对基本粒子群优化算法容易陷入局部最优解的问题,本文定义了PSO粒子搜索中心的概念,并对其随机状态下粒子搜索中心在全局最优解与局部最优解之间的概率密度进行了计算,在此基础上提出了粒子搜索中心在两个最优解之间均匀分布的均匀搜索粒子群算法,并通过7个Benchmark函数与基本PSO算法进行了对比实验及算法分析,实验分析结果表明,均匀搜索粒子群算法在函数优化尤其非均匀多峰值函数优化中具有更好的收敛速度及稳定性.%It is well known that the Particle Swarm Optimization (PSO) algorithm easily falls into the local optimal solution.In this paper,we defined a concept of PSO particle-search center,and analyzed the probability density of the center between global and local optimal solutions in random state. A uniform searching particle swarm optimization (UPSO) algorithm whose particlesearch center uniformly distributed between local and global optimal solutions is proposed based on that analysis. By analyzing the comparative experiments between UPSO and PSO algorithm with seven benchmark functions, we found that the UPSO and its improved algorithms are more stable and can improve the convergence efficiency in function optimization,especially in non-uniformly multimodal function optimization.