采用扰动加速因子的自适应粒子群优化算法

针对粒子群算法容易陷入早熟收敛和搜索效率不高等问题,分析了几个现有的改进粒子群优化算法.在粒子对称分布有利于提高搜索结果的基础上,对粒子群优化算法进行了改进.改进后的算法可以在运行过程中的不同阶段自适应地以余弦函数的变化方式调整惯性权重系数;在加速因子线性变化的基础上,基于一定的条件对加速因子进行扰动;并确定了相应条件参数的参数取值.通过几个经典的函数,对该算法进行了验证,并与相关文献中改进的粒子群优化算法进行了对比.结果表明,新算法不仅显著提高了收敛速度,而且能有效地改善早熟现象.

Adaptive particle swarm optimization via disturbing acceleration coefficents

The Particle Swarm Optimization(PSO) is an evolutionary method which is used to search for the global optimal solution by iteration.However,PSO has the problem that the particle swarm algorithm falls easily into premature convergence and has a low search efficiency.In this paper,after analyzing several existing improved particle swarm algorithms,a new improved particle swarm algorithm is proposed based on the fact that symmetrical particles distribution can enhance the optimation search results.The proposed algorithm can adjust the inertia weight factor adaptively in different phases of the process according to the variation of the cosine function.In addition,the acceleration coefficents based on linear variation are disturbed under a certain condition.Moreover,an appropriate value of the parameter in this condition is determined via experiments.Several classic functions have been used to test this new algorithm and then the results of this new algorithm are analyzed by comparing it with several relevant algorithms in the literature.The results show that this new algorithm can not only improve the convergence speed significantly,but also improve the premature convergence phenomenon.