求解大规模全局优化问题的新型三层递归差分分组方法  

New three-level recursive differential grouping method for large-scale optimization problems

作　　者：李飞 刘翔 徐洪斌 刘建昌[4] LI Fei;LIU Xiang;XU Hong-bin;LIU Jian-chang(School of Electrical and Information Engineering,Anhui University of Technology,Anhui Ma’anshan 243032,China;Anhui Province Engineering Laboratory of Intelligent Demolition Equipment,Ma’lanshan 243032,China;College of Oceanography and Space Informatics,China University of Petroleum(East China),Qingdao Shandong 266580,China;College of Information Science and Engineering,Northeastern University,Shenyang Liaoning 110004,China)

机构地区：[1]安徽工业大学电气与信息工程学院,安徽马鞍山243032 [2]安徽省智能破拆装备工程实验室,安徽马鞍山243002 [3]中国石油大学(华东)海洋与空间信息学院,山东青岛266580 [4]东北大学信息科学与工程学院,辽宁沈阳110819

出　　处：《控制理论与应用》2024年第4期691-700,共10页Control Theory & Applications

基　　金：国家自然科学基金青年项目(61903003);安徽省自然科学基金青年项目(2008085QE227);安徽省高校自然科学重点研究项目(KJ2019A0051);安徽省智能破拆装备工程实验室开放基金资助项目(APELIDE2022A007)资助.

摘　　要：协同进化算法在求解大规模全局优化问题上具有较好的效果,其核心思想是利用分而治之的策略将一个高维问题分解成若干个子问题,然后分别优化每个子问题.然而,现有的分解方法通常需要花费大量的计算成本来获得精确的变量分组.通过采用递归交互检测中的历史信息简化分组过程,能够避免检测某些集合的相互关系,本文提出了一种新型三层递归差分分组策略(NTRDG).与其他4种现有的分组方法相比,NTRDG在不影响分组精度的情况下计算成本消耗较低.仿真结果表明,NTRDG在求解大规模全局优化问题时具有很强的竞争力.The cooperative coevolution algorithm performs well in solving large-scale global optimization problems.The core idea of cooperative coevolution is to utilize a divide-and-conquer strategy for decomposing high-dimensional problems into multiple subproblems,which are then processed individually and separately.However,existing decomposition methods typically require significant computational cost to obtain accurate variable grouping.To address this issue,a novel three-level recursive differential grouping strategy(NTRDG)is proposed in this paper,which simplifies the grouping process by utilizing historical information in recursive interaction detection and avoids the detection of relationships among certain sets,leading to a lower computational cost without sacrificing grouping accuracy.Simulation results demonstrate that compared to four existing methods,NTRDG exhibits a stronger competitiveness in solving large-scale global optimization problems.

关 键 词：全局优化 协同进化 分解方法 三层递归差分分组 递归搜索 

分 类 号：O224[理学—运筹学与控制论]