引用本文:纪昌明,马皓宇,李宁宁,等.基于树形结构无界存档的多目标粒子群算法[J].控制与决策,2020,35(11):2675-2686
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览次   下载 本文二维码信息
码上扫一扫!
分享到: 微信 更多
基于树形结构无界存档的多目标粒子群算法
纪昌明,马皓宇,李宁宁,吴嘉杰,彭杨,王丽萍
(华北电力大学可再生能源学院,北京102206)
摘要:
多目标优化算法大多采用基于线性链表结构的有界Pareto存档策略,其存在迭代过程中Pareto前沿震荡衰退等弊端以及相关参数难以预先确定等技术难题.为此,构造一种适用于大规模存档集合的树形结构,并利用其取代线性结构以保证存档维护与管理的高效性,进而提出基于树形结构的无界存档策略.在此基础上,将基于正交设计的种群初始化、基于树形结构的存档更新以及基于树形结构的最优个体选择引入多目标粒子群优化,提出基于树形结构无界存档的多目标粒子群算法.最后,通过测试函数上的仿真实验验证了所提出策略与算法的科学性和有效性.
关键词:  多目标优化  树形结构  无界存档  粒子群优化  正交设计
DOI:10.13195/j.kzyjc.2019.0276
分类号:TP301.6
基金项目:国家自然科学基金项目(51679088,51279062);“十三五”国家重点研发计划课题(2016YFC0402208).
Multi-objective particle swarm optimization algorithm based on tree-structured unbounded archive
JI Chang-ming,MA Hao-yu,LI Ning-ning,WU Jia-jie,PENG Yang,WANG Li-ping
(College of Renewable Energy,North China Electric Power University,Beijing 102206,China)
Abstract:
Most of the current multi-objective optimization algorithms adopt the bounded Pareto archive strategy based on the linear list structure, which has the drawbacks of Pareto fronts' oscillation, shrinking and other technical difficulties such as pre-determining relevant parameters. Therefore, this paper constructs a tree structure suitable for the storage and update of large-scale archive, replacing the linear structure to ensure high efficiency of archive maintenance and management. And a tree-structured unbounded archive strategy is proposed. We introduce population initialization based on orthogonal design, archive updating and optimal individual selection based on the tree structure into multi-objective particle swarm optimization, and propose a multi-objective particle swarm optimization algorithm based on tree-structured unbounded archive. Finally, simulation experiments on test functions verify the feasibility and effectiveness of the improved strategies and algorithm.
Key words:  multi-objective optimization  tree structure  unbounded archive  particle swarm optimization  orthogonal design

用微信扫一扫

用微信扫一扫