摘要
作为群体智能的代表性方法之一,粒子群优化算法(PSO)通过粒子间的竞争和协作以实现在复杂搜索空间中寻找全局最优点。提出了一种改进的粒子群优化算法(MPSO),该算法以广泛学习粒子群优化算法(CLPSO)的思想为基础,主要引入了选择墙的概念。同时在参数的设置中结合高斯分布的概念,以提高算法的收敛性。实验结果表明,改进后的粒子群算法防止陷入局部最优的能力有了明显的增强。同时,算法使高维优化问题中全局最优解相对搜索空间位置的鲁棒性得到了明显提高。
As a representative method of swarm intelligence,Particle Swarm Optimization （PSO） is an algorithm for searching the global optimum in the complex space through cooperation and competition among the individuals in a population of particle.A Modified PSO （MPSO） is presented in this paper,This method mainly inducts a new concept called selecting walls on the base of the idea of the CLPSO and combines the Gaussian distribution in controlling the parameters to improve the convergence ability.The experimental result indicates that the modified PSO increases the ability to break away from the local optimum,Simultaneously,the algorithm obtains a robust optimization performance regardless the location of the global optimum in the high dimension problem.
出处
《计算机工程与应用》
CSCD
北大核心
2006年第33期40-42,73,共4页
Computer Engineering and Applications
基金
国家自然科学基金资助项目(60475019)
关键词
粒子群
优化
进化计算
particle swarm
optimization
evolutionary computation
