计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230700155-11.doi: 10.11896/jsjkx.230700155

• 人工智能 • 上一篇    下一篇

基于双重方向向量的大规模多目标进化算法

韩立君, 王鹏, 李瑞旭, 刘仲尧   

  1. 烟台大学计算机与控制工程学院 山东 烟台 264005
  • 发布日期:2024-06-06
  • 通讯作者: 王鹏(to_wangpeng@163.com)
  • 作者简介:(hanlijun20210317@163.com)
  • 基金资助:
    国家自然科学基金(62072392,61972360,62103350);山东省重大科技创新工程项目(2019522Y020131);山东省自然科学基金(ZR2020QF113,ZR2020QF046,ZR2021QF086)

Dual Direction Vectors-based Large-scale Multi-objective Evolutionary Algorithm

HAN Lijun, WANG Peng, LI Ruixu, LIU Zhongyao   

  1. College of Computer and Control Engineering,Yantai University,Yantai,Shandong 264005,China
  • Published:2024-06-06
  • About author:HAN Lijun,born in 1997,postgraduate.Her main research interests include large-scale multi-objective evolutionary algorithm and so on.
    WANG Peng,born in 1987,Ph.D,lecturer,postgraduate supervisor.is a member of CCF(No.F9977M).His main research interests include evolutionary computation,service computing,and swarm intelligence algorithm.
  • Supported by:
    National Natural Science Foundation of China(62072392,61972360,62103350) ,Major Innovation Project of Science and Technology of Shandong Province(2019522Y020131) and Natural Science Foundation of Shandong Province,China(ZR2020QF113,ZR2020QF046,ZR2021QF086).

PDF (PC)

280

摘要： 大规模多目标优化问题的决策空间维度高达数百维,在巨大的搜索空间中实现快速收敛同时高效保持种群多样性极具挑战。针对上述问题,文中提出了一种基于双重方向向量的大规模多目标进化算法(DDLE),该算法的主要思想是利用两类不同的方向向量引导种群进化,提高算法的搜索效率。首先,设计了一种收敛性方向向量生成策略提升算法的收敛速度;其次,推出了一种多样性方向向量生成策略增强种群的多样性;最后,提出了一种基于自适应的环境选择算子动态平衡种群进化过程中的收敛性与多样性。为验证DDLE的性能,将其与5种先进的算法在72个大规模基准测试问题上进行了对比实验。实验结果表明,DDLE在求解大规模多目标优化问题上相较于其它对比算法具有显著优势。

关键词: 进化算法, 大规模多目标优化, 双重方向向量, 收敛性方向向量, 多样性方向向量

Abstract: The decision space dimension of large-scale multi-objective optimization problemsis up to hundreds of dimensions.It is extremely challenging to achieve fast convergence in the huge search space while efficiently maintaining the diversity of the population.To address the above problems,a dual direction vectors-based large-scale multi-objective evolutionary algorithm(DDLE) is proposed in the paper.The main idea of the algorithm is to utilize two different types of direction vectors to guide the population evolution and improve the search efficiency of the algorithm.First,a convergent direction vector generation strategy is designed to improve the convergence speed of the algorithm.Second,a diversity direction vector generation strategy is introduced to enhance the diversity of the population.Finally,an adaptive environment-based selection operator is proposed to dynamically balance the convergence and diversity in the process of population evolution.To verify the performance of DDLE,it is compared with five state-of-the-art algorithms in experiments on 72 large-scale benchmark test problems.Experimental results show that DDLE has a significant advantage over other compared algorithms in solving large-scale multi-objective optimization problems.

Key words: Evolutionary algorithms, Large-scale multi-objective optimization, Dual direction vectors, Convergence direction vector, diversity direction vector

中图分类号: 

  • TP301
[1]CAI X Y,MA Z Y,ZHANG F,et al.A daptive decomposition-based multi-task collabo-rative expensive multi-objective optimization algorithm[J].Journal of Computer Science,2021,44(9):1934-1948.
[2]TIAN Y,CHENG R,ZHANG X Y,et al.A strengthened dominance relation considering convergence and diversity for evolutionary many-objective optimization[J].IEEE Transactions on Evolutionary Computation,2018,23(2):331-345.
[3]LIU X F,ZHAN Z H,GAO Y,et al.Coevolutionary particleswarm optimization with bottleneck objective learning strategy for many-objective optimization[J].IEEE Transactions on Evolutionary Computation,2018,23(4):587-602.
[4]OLOWU T,JAFARI H,MOGHADDAMI M,et al.Multiphysics and multiobjective design optimization of high-frequency transformers for solid state transformer applications[J].IEEE Transactions on Industry Applications,2020,57(1):1014-1023.
[5]LI Y Z,NI Z X,ZHAO T Y,et al.Coordinated scheduling for improving uncertain windpower adsorption in electricvehicles－windintegrated power systems by multiobjective optimization approach[J].IEEETransactions on Industry Applications,2020,56(3):2238-2250.
[6]GU Z M,WANG G G.Improving NSGA-IIIalgorithms with in-formation feedback models for large-scalemany-objective optimization[J].Future Generation Computer Systems,2020,107:49-69.
[7]HONG W J,YANG P,TANG K.Evolutionary computationfor large-scale multi-objective optimization:A decade of progresses[J].International Journal of Automation and Computing,2021,18(2):155-169.
[8]MA L B,HUANG M,YANG S X,et al.Anadaptive localizeddecision variable analysis approach to large-scale multiobjective and many-objective optimization[J].IEEE Transactions on Cybernetics,2021,52(7):6684-6696.
[9]WAN H D,JIAO L C,SHANG R H,et al.A memetic optimiza-tion strategy based on dimension reduction in decision space[J].Evolutionary Computation,2015,23(1):69-100.
[10]PARSONS L,HAQUE E,LIU H.Subspace clustering for high dimensional data:a review[J].Acmsigkdd Explorations News Letter,2004,6(1):90-105.
[11]LIU J C,LI F,WANG H H,et al.A review of research on evolutionary high-dimensional multiobjective optimization algorithms[J].Control and Decision,2018,33(5):879-887.
[12]ANTONIO L M,COELLO C A C.Use of cooperative coevolution for solving larges-cale multi objective optimization problems[C]//IEEE Congress on Evolutionary Computation.2013:2758-2765.
[13]HE C,CHENG R,LI L H,et al.Large-scale multi objective optimization viare for mulated decision variable analysis[J].IEEE Transactions on Evolutionary Computation,2022,26:1-1.
[14]FENG Y L,FENG L,KWONG S,et al.A multivariation multifactorial evolutionary algorithm for large-scale multiobjective optimization[J].IEEE Transactions on Evolutionary Computation,2021,26(2):248-262.
[15]HE C,CHENG R,YAZDANI D.Adaptive offspring generation for evolutionary large-scale multiobjective optimization[J].IEEE Transactions on Systems,Man,and Cybernetics:Systems,2020,52(2):786-798.
[16]SUN G,WU J J,CHEN H,et al.Hidden preference-basedmulti-objective evolutionary algorithm based on chebyshev distance[J].Computer Science,2020,41(12):2543-2549.
[17]DEB K,PRATAP A,AGARWAL S,et al.A fast and elitist multiobjective genetic algorithm:NSGAII[J].IEEE Transactions on Evolutionary Computation,2002,6(2):182-197.
[18]ZITZLER E,LAUMANNS M,THIELE L.S-PEA2:Improving the strength Pareto evolutionary algorithm[J].Technical Report,2001,103:95-100.
[19]ZHANG Q F,LI H.MOEA/D:A multiobjective evolutionaryalgorithm based on decomposition[J].IEEE Transactions on Evolutionary Computation,2007,11(6):712-731.
[20]LIU H L,GU F Q,ZHANG Q F.Decomposition of a multiobjective optimization problem into a number of simple multiobjective subproblems[J].IEEE Transactions on Evolutionary Computation,2014,18(3):450-455.
[21]ZITZLER E,KUNZLI S.Indicator-basedsel-ection in multiobjective search[C]//Parallel Problem Solving from Nature.Springer Berlin Heidelberg,2004:832-842.
[22]TIAN Y,CHENG R,ZHANG X Y,et al.An indicator-based multiobjective evolutionary algorithm withreference point adaptation for better versatility[J].IEEE Transactionson Evolutionary Computation,2017,22(4):609-622.
[23]LIU Z Z,WANG Y.Handling constrained multiobjective optimization problems with constraints in both the decision and object-ive spaces[J].IEEE Transactions on Evolutionary Computation,2019,23(5):870-884.
[24]MA X L,LIU F,QI Y T,et al.A multiobjec-tive evolutionary algorithm based on decision variable analyses for multiobjective optimization problems with large-scale variables[J].IEEE Transactions on Evolutionary Computation,2015,20(2):275-298.
[25]HEINER Z,HISAO I,SANAZ M,et al.A framework for large-scale multiobjective optimization based on problem transformation[J].IEEE Transactions on Evolutionary Computation,2017,22(2):260-275.
[26]HE C,LI L H,TIAN Y,et al.Accelerating large-scale multiobjective optimization via problem reformulation[J].IEEE Transactions on Evolutionary Computation,2019,23(6):949-961.
[27]LIANG Z P,LIU C,WANG Z Q,et al.Alarge-scale multi-objective optimization algorithm based on archiving and weight expan-sion[J].Journal of Computer Science,2022,45(5):951-972.
[28]TIAN Y,ZHENG X T,ZHANG X Y,et al.Efficient large-scale multiobjective optimization based on a competitive swarm optimizer[J].IEEE Transactions on Cybernetics,2019,50(8):3696-3708.
[29]CHENG R,JIN Y C,OLHOFER M,et al.A reference vectorguided evolutionary algorithm for many-objective optimization[J].IEEE Transactions on Evolutionary Computation,2016,20(5):773-791.
[30]CHENG R,JIN Y C,OLHOFER M.Test problems for large-scale multiobjective and many-objective optimization[J].IEEE Transactions on Cybernetics,2016,47(12):4108-4121.
[31]LI L H,HE C,CHENG R,et al.Afast sam-pling based evolutionary algorithm for million-dimensional multiobjective optimization[J].Swarm and Evolutionary Computation,2022,75:101181.
[32]ZHANG K,SHEN C N,YEN G G.Multipo-pulation-based differential evolution for large-scale many-objective optimization[J].IEEE Transactions on Cybernetics,2022,10:1-13.
[33]CHEN H K,CHENG R,WEN J M,et al.Solving large-scale many-objective optimization problems by covariance matrix adap-tation evolution strategy with scalable small subpopulations[J].Information Sciences,2020,509:457-469.
[34]TIAN Y,CHENG R,ZHANG X Y,et al.Pla-tEMO:A MATLAB platform for evolutionary multi-objective optimization[educational forum][J].IEEE Computational Intelligence Ma-gazine,2017,12(4):73-87.
[35]ZHOU A M,JIN Y C,ZHANG Q F,et al.Combining model-based and genetics-based offspring generation for multi-objective optimization using a convergence criterion[C]//2006 IEEE International Conference on Evolutionary Computation.2006:892-899.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发