摘要
针对标准粒子群优化算法在处理复杂函数优化问题时容易陷入局部最优、收敛精度低的缺点,提出了一种改进的PSO算法,该算法把生物学中的吸引排斥思想引入到PSO算法中,充分利用粒子间的相互影响,修正了其速度更新公式,从而维持了群体的多样性,增强了粒子跳出局部最优解的能力。实验仿真结果表明,改进的PSO算法提高了进化后期的收敛速度,有效避免了PSO算法的早熟收敛问题,而且具有较高的收敛精度。
Standard Particle Swarm Optimization(PSO) algorithm falls into local optima easily and has low convergence accuracy when it is used to address the problem of complex functions optimization.In order to overcome the shortcomings,an improved PSO algorithm was proposed.The proposed algorithm integrated the attraction-repulsion mechanism in the field of biology into PSO algorithm and took full advantage of the mutual influence between particles to modify velocity updating formula,and thus maintained population d...
出处
《计算机应用》
CSCD
北大核心
2009年第2期542-544,557,共4页
journal of Computer Applications
基金
国家自然科学基金资助项目(60674108
60574075)
关键词
粒子群优化
早熟收敛
吸引排斥机制
复杂函数
particle swarm optimization
premature convergence
attraction-repulsion mechanism
complex functions