基于环形邻域拓扑的自适应速度PSO算法.PDFVIP

32 2015 ，51（18） Computer Engineering and Applications 计算机工程与应用 ⦾理论研究、研发设计⦾ 基于环形邻域拓扑的自适应速度PSO 算法 徐 迅，鲁海燕，徐向平 XU Xun, LU Haiyan, XU Xiangping 江南大学 理学院，江苏 无锡 214122 School of Science, Jiangnan University, Wuxi, Jiangsu 214122, China XU Xun, LU Haiyan, XU Xiangping. Self-adaptive velocity PSO algorithm based on ring neighborhood topology. Computer Engineering and Applications, 2015, 51（18）：32-37. Abstract ：Based on the global version of Self-Adaptive Velocity Particle Swarm Optimization（SAVPSO ）algorithm, this paper proposes a local version of SAVPSO using ring neighborhood topology for solving constrained optimization problems in order to counter the disadvantage of the global SAVPSO of easily falling in local optima, and uses Dynamic-Objective Method （DOM）to effectively deal with the constraints. The performance of the proposed algorithm is evaluated on 13 well-known benchmark functions. Experimental results show that the proposed algorithm has stronger ability to find global optimal solutions and to avoid falling in local optima compared with the global SAVPSO, and that the neighborhood size and realization have impact on the performance of the algorithm. Key words ：constrained optimization; particle swarm optimization; Dynamic-Objective Method （DOM）; Self-Adaptive Velocity Particle Swarm Optimization （SAVPSO ）; constraint-handling mechanism; ring neighborhood topology 摘 要：为克服全局粒子群优化算法易陷入局部最优的缺点，基于全局自适应速度粒子群优化（SAVPSO）算法，给出一 种基于环形邻域拓扑的局部SAVPSO 算法来求解约束优化问题，同时采用动态目标方法（DOM）来有效处理约束条件，并 以13 个经典的测试函数为例对算法的性能进行仿真实验研究。测试结果表明，与全局SAVPSO 算法相比，该算法具有 较强的全局寻优能力，可以较好地避免陷入局部最优；另外，粒子的邻域大小及实现形式对算法的性能均有一定的影响。 关键词：约束优化；粒子群优化；动态目标方法（DOM）；自适应速度粒子群优化（SAVPSO）；约束处理机制；环形邻域拓扑 文献标志码：A 中图分类号：TP301.6 doi ：10.3778/j.issn.1002-8331.1406-0377 1 概述 最优位置更新迭代，于是其收敛速度快，但同时也容易 粒子群优化（Particle Swarm Optimization ，PSO）算 早熟，难以找到全局最优解；局部PSO 收敛速度较慢，但 [1-2]