摘要
针对多模态多目标优化算法难以搜索到多个最优解,提出了结合两种拓扑结构的多模态多目标粒子群优化算法。该算法采用两种不同的拓扑结构,以应对不同的搜索时期。算法初期采用星型拓扑来构造粒子的邻域,让粒子在全局范围内进行信息交换,提高算法勘探的性能;算法中后期采用基于距离的环型拓扑结构,让粒子邻居解在搜索空间上也相邻,增强粒子间局部信息交互能力,提高算法开采的性能。算法还设计一种新的粒子修补策略,增强解的分布性。在15个多模态多目标测试函数上进行仿真,实验结果表明,所提算法相较于经典算法,所得Pareto最优解的分布性更好。
To solve the multi-modal and multi-objective optimization problem that is difficult to find the multiple optimal solutions,a multi-modal multi-objective particle swarm optimization algorithm with two topology structure is proposed.The algorithm adopts two different topological structures to deal with different search periods.At the beginning of the algorithm,star topology is used to construct the neighborhood of particles,which enables the particles to exchange information in the global range,and improve the speed of searching.In the middle and later stages of the algorithm,the ring topology structure based on distance is used to make the neighbor solutions of particles adjacent in the search space,and enhance the ability of local information interaction between particles.A new particle repair strategy is used to enhance the distribution of solutions.The simulation results of 15 multimodal and multi-objective test functions show that the proposed algorithm has better distribution of Pareto optimal solutions than the classical algorithms.
作者
汪慎文
王佳莹
高娜
周瑶
WANG Shenwen;WANG Jiaying;GAO Na;ZHOU Yao(School of Information Engineering,Hebei GEO University,Shijiazhuang 050031,China;Laboratory of Artificial Intelligence and Machine Learning,Hebei GEO University,Shijiazhuang 050031,China)
出处
《南昌工程学院学报》
CAS
2020年第4期66-71,95,共7页
Journal of Nanchang Institute of Technology
基金
国家自然科学基金资助项目(61402481,61773296)
河北青年拔尖人才支持计划(冀字[2013]17号)
河北省科技厅重点项目(19970311D,20373303D)
河北省教育厅自然科学基金重点项目(ZD2018083,ZD2018043,ZD2019134)
河北地质大学博士科研启动基金项目(BQ201322)。
关键词
多模态多目标优化:粒子群优化算法
星型拓扑结构
环型拓扑结构
multi-modal and multi-objective optimization:particle swarm optimization algorithm
star topology structure
ring topology structure
作者简介
汪慎文(1979-),男,博士,教授,硕士生导师,CCF会员,wangshenwen@hgu.edu.cn.
