一种改进竞争机制的大规模多目标粒子群优化算法  

A Large-scale Multi-objective Particle Swarm Optimization Algorithm with Improved Competitive Mechanism

作　　者：杨五四 李艳艳 张珏[3] 唐存东 李红 王艺 YANG Wusi;LI Yanyan;ZHANG Jue;TANG Cundong;LI Hong;WANG Yi(School of Computer Sciece,Xianyang Normal University,Xianyang 712000,Shaanxi,China;Freshman College,Xi’an Technohgical University,Xi’an 710021,Shaanxi,China;School of Information Engineering,Yulin University,Yulin 719000,Shaanxi,China;School of Information Science and Technology,Northwest University,Xi’an 710127,Shaanxi,China)

机构地区：[1]咸阳师范学院计算机学院,陕西咸阳712000 [2]西安工业大学新生院,陕西西安710021 [3]榆林学院信息工程学院,陕西榆林719000 [4]西北大学信息科学与技术学院,陕西西安710127

出　　处：《咸阳师范学院学报》2025年第2期13-20,共8页Journal of Xianyang Normal University

基　　金：国家自然科学基金项目(62266047)及子项目(ZKT01);咸阳师范学院重点培育基金项目(XSYK22024);咸阳师范学院大学生创新创业训练计划项目(xysfxy2024143);陕西省自然科学基金项目(2022JM-361,2023-JC-YB-524)。

摘　　要：大规模多目标优化问题难点在于随着问题决策变量的增加,使得求解问题的难度呈指数级增长。实验已证明竞争机制的粒子群优化算法能够有效处理大规模优化问题,但为了更好处理大规模多目标优化问题,文中提出了一种改进的竞争机制大规模多目标粒子群优化算法。该算法对竞争粒子群优化算子进行了改进,以增强算法在大规模多目标优化问题上的全局搜索能力,并采取不同的环境选择策略。文中算法与其他4种大规模多目标优化算法的实验对比结果表明,提出的算法在大多数测试问题上的表现优于对比算法,具有较好的性能。The difficulty of large-scale multi-objective optimization problems lies in the fact that as the decision variables of the problem increase,the difficulty of solving the problem increases exponentially.Research has demonstrated that the particle swarm optimization algorithm with competitive mechanism can effectively deal with large-scale optimization problems,but to better address large-scale multi-objective optimization problems,this paper proposes an improved large-scale multi-objective particle swarm optimization algorithm with competitive mechanism.The algorithm improves the competitive particle swarm optimization operator to enhance the global search ability of the algorithm on large-scale multi-objective optimization problems with different environment selection strategies.The experimental comparison results of this paper's algorithm with four other large-scale multi-objective optimization algorithms show that the proposed algorithm outperforms the comparison algorithms on most of the tested problems and has a better performance.

关 键 词：竞争机制 大规模优化 粒子群优化 多目标优化 移位密度估计 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]