基于随机黑洞和逐步淘汰策略的多目标粒子群优化算法.pdfVIP

第28 卷 第11 期 控 制 与 决 策 2013 年 11 月 Vol. 28 No. 11 Control and Decision Nov. 2013 文章编号: 1001-0920 (2013) 11-1729-06 基于随机黑洞和逐步淘汰策略的多目标粒子群优化算法 陈民铀, 程 杉 (重庆大学a. 电气工程学院，b. 输配电装备及系统安全与新技术国家重点实验室，重庆400044) 摘 要: 提出一种基于随机黑洞粒子群算法(RBH-PSO) 和逐步淘汰策略的多目标粒子群优化(MRBHPSO-SE) 算法. 利用RBH-PSO 全局优化能力强和收敛速度快的优点逼近Pareto 最优解; 为了避免拥挤距离排序策略的缺陷, 提出逐 步淘汰策略, 并将其应用到下一代粒子的选择策略中. 同时, 动态选择领导粒子, 运用动态惯性权重系数和变异操作 来增强种群全局寻优能力, 以及避免早熟收敛. 利用具有不同特点的测试函数进行验证, 结果表明, 与同类算法相比, 该算法具有较高的精度并兼顾优化解的多样性. 关键词: 多目标优化；随机黑洞粒子群算法；拥挤距离排序；逐步淘汰 中图分类号: TP18 文献标志码: A Multi - objective particle swarm optimization algorithm based on random black hole mechanism and step-by-step elimination strategy CHEN Min-you, CHENG Shan (a. School of Electrical Engineering，b. State Key Laboratory of Power Transmission Equipment System Security and New Technology，Chongqing University，Chongqing 400044，China ．Correspondent ：CHEN Min-you，E-mail: minyouchen@cqu.edu.cn) Abstract: A multi-objective particle swarm optimization algorithm based on the random black hole particle swarm optimization(RBH-PSO) and step-by-step elimination strategy is proposed. The Pareto optimal solutions are approached by its advantage of speeding up the convergence and improving the performance of global optimizer greatly. To avoid the disadvantage of crowding distance sorting technique, the step-by-step elimination(SE) strategy is proposed, which is used to select the particles from one iteration to another. In addition, dynamic selection of leader particle for each particle, adaptive inertia weight and a special mutation operation are incorporated to enhance the global expl