一种用于求解多目标优化问题的改进MBO算法  

An Improved MBO Algorithm for Solving Multi-Objective Optimization Problems

作　　者：蒙丽萍 姜思佳[2] 韦量 王勇[4] MENG Liping;JIANG Sijia;WEI Liang;WANG Yong(West of Guangxi Science and Technology Co.,LTD,Nanning 530009,China;Liuzhou City Vocational College,Liuzhou 545036,China;Guangxi Financial Vocational College,Nanning 530007,China;College of Artificial Intelligence,Guangxi Minzu University,Nanning 530006,China)

机构地区：[1]广西西子科技有限责任公司,广西南宁530009 [2]柳州城市职业学院,广西柳州545036 [3]广西金融职业技术学院,广西南宁530007 [4]广西民族大学人工智能学院,广西南宁530006

出　　处：《广西民族大学学报（自然科学版）》2022年第2期74-80,共7页Journal of Guangxi Minzu University ：Natural Science Edition

基　　金：国家自然科学基金项目(61662005);广西高校中青年教师基础能力提升项目(2019KY1607)。

摘　　要：大红斑蝶优化算法(MonarchButterflyOptimization,MBO)是近年提出的一种新的随机搜索算法。然而,MBO算法作为一种优化方法仍存在早熟、收敛速度缓慢和全局搜索能力不足的现象,也还没有将MBO算法应用于求解多目标优化问题的相关研究成果。针对这一问题,笔者提出一种新的可用于求解多目标优化问题的改进MBO算法,即MOIMBO算法。选取3个经典的多目标优化问题作为算法优化性能测试,实验结果表明:改进的算法具有较强的全局搜索能力,在一定程度上避免了早熟收敛问题,可用于求解多目标优化问题。Monarch butterfly optimization is a new random search algorithm proposed in recent years.However,as an optimization method,MBO still has the phenomena of prematurity,slow convergence rate and insufficient global search ability.There are no relevant research results on the application of MBO in solving the multi⁃objective optimization problems.Aim at this problem,a new improved MBO algorithm which can be used to solve the multi⁃objective optimization problems is proposed in this paper,called MOIMBO algorithm.Experiments were done on three typical multi⁃objective optimization problems as the performance tests of algorithm optimization.The experimental results show that the proposed algorithm has strong global search ability and avoids premature convergence to some extent,and can be used to solve the multi⁃objective optimization problems.

关 键 词：MBO算法 智能优化算法 多目标优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]