摘要
粒子群优化算法(particle swarm optimization,PSO)具有实现简单、在演化前期收敛速度快等优点,但在演化后期具有收敛速度慢、容易陷入局部最优以及精度低等不足.针对PSO算法容易陷入局部最优及精度低的不足提出一种带正弦函数因子的粒子群优化算法(TFPSO).该算法在PSO算法的位置更新方程中引入具有周期振荡性的正弦函数因子,使每个粒子位置获得周期振荡性,扩大搜索空间,更容易跳出局部最优,避免算法过早的收敛,找到最优值.实验研究表明,该算法不但实现简单、稳定而且提高了解的精度.
Particle Swarm Optimization{ PSO ） is a simple algorithm with converging very fast in the early stage of evolution process but converging very slow in the later stages. It is also easy to fall into local optimum and causes low accuracy of solution. For easily falling into local optimum and causing low accuracy of solution of PSO algorithm, this paper put forwards an improved PSO algorithm based on sine trigonometric factor （ TFPSO）. The algorithm introduces the periodic oscillations trigonometric factor in the updating e- quation of the position of the PSO algorithm, so that each particle obtains periodic oscillations to expand the search space and more easily escape from local optima and avoid premature convergence of algorithms and find the most excellent value. Experimental studies show that the TFPSO algorithm is not only simple to implement and stability, but also to improve the accuracy of solution.
出处
《小型微型计算机系统》
CSCD
北大核心
2015年第1期156-161,共6页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61070008)资助
中央高校基本科研业务费专项项目(2012211020205)资助
关键词
粒子群算法
三角函数
正弦函数因子
周期振荡性
particle swarm optimization
trigonometric
the trigonometric sine factors
periodic oscillations
