| Computer Technology, Information Engineering |
|
|
|
|
| New hybrid decomposition many-objective evolutionary algorithm |
| GUO Xiao fang, WANG Yu ping, DAI Cai |
1. School of Computer, Xidian University, Xi'an 710071, China;
2. School of Science, Xi'an Technological University, Xi'an 710032, China |
|
|
|
Abstract A hybrid decomposition many-objective evolutionary algorithm based on a new dominance relation was proposed inspired by many-objective evolutionary algorithms based on decomposition in order to improve the diversity and convergence of the non-dominated solution set in manyobjective optimization problems. The sub-population evolutionary pattern was adopted, and a new efficiency order based dominance relation was designed to compare and update individuals inside each subpopulation, which helps to increase selective pressure and improve diversity. Powell search was used as the local search operator in order to improve the performance of local search. A hybrid evolution strategy combining traditional optimization method with evolutionary algorithm was adopted. Six standard benchmark problems with 5 to 20 objectives were tested to demonstrate the effectiveness of the algorithm. Experimental results showed that the algorithm performed better than other available algorithms in convergence and diversity．
|
|
Published: 23 July 2016
|
|
|
新的混合分解高维多目标进化算法
在基于分解技术求解高维多目标优化问题的思想启发下,为了提高多目标优化问题非支配解集合的分布性和收敛性,提出新的基于个体支配关系的混合分解高维多目标进化算法.该算法采用分子种群的进化模式,设计新的基于有效阶的个体支配关系用于个体的比较和更新操作,以便在增加个体选择压力的同时提高解集分布的多样性.为了改善该算法的局部搜索性能,将Powell搜索作为局部搜索算子,采用传统优化与进化算法相融合的混合进化策略.为了检验提出算法的性能,将提出算法用于求解5～20个目标的6类标准测试问题,与同类算法相比,该算法在收敛性和分布性方面均具有较大的改进和提高.
|
|
| ［1］ CHRISTIAN V L.A survey on multiobjective evolutionary algorithms for manyobjective problems ［J］. Computational Optimization and Applications, 2014, 58(3):707-756．
［2］ 公茂果,焦李成,杨咚咚,等. 进化多目标优化算法研究［J］. 软件学报,2009, 2(20): 271-289．
GONG Maoguo, JIAO Licheng, YANG Dongdong, et al. Research on evolutionary multiobjective optimization algorithms ［J］. Journal of Software, 2009, 2(20): 271-289．
［3］ 孔维健. 高维多目标进化算法研究综述［J］. 控制与决策, 2010,25（3）：321-326．
KONG Weijian. Survey on largedimensional multiobjective evolutionary algorithms ［J］. Control and Decision, 2010, 25(3): 321-326．
［4］ 巩敦卫. 基于集合的高维多目标优化问题的进化算法［J］. 电子学报, 2014, 42(1): 77-83．
GONG Dunwei. Solving manyobjective optimization problems using setbased evolutionary algorithms ［J］. Acta Electronica Sinica, 2014, 42(1): 77-83．
［5］ PIERRO D F, KHU S T, SAVIC D A.An investigation on preference order ranking scheme for multiobjective evolutionary optimization ［J］. IEEE Transactions on Evolutionary Computation, 2007, 11(1): 17-45．
［6］ ZOU Xiufen, CHEN Yu, LIU Minzhong. A new evolutionary algorithm for solving manyobjective optimization problems ［J］. IEEE Transactions on Systems, Man, and Cybernetics, part BCybernetics, 2008, 38(5): 1402-1412．
［7］ MOLINA J, SANTANA L, HERNANDEZDIAZ A, et al. Gdominance: reference point based dominance for multiobjective metaheuristics ［J］. European Journal of Operational Research, 2009, 197(2): 685-692．
［8］ DEB K, JAIN H. An evolutionary manyobjective optimization algorithm using referencepointbased nondominated sorting approach, part i: solving problems with box constraints ［J］. IEEE Transactions on Evolutionary Computation, 2014, 18(4): 577-601.
［9］ YUAN Yuan, HUA Xu, WANG Bo.A new dominance relation based evolutionary algorithm for manyobjective optimization ［J］. IEEE Transactions on Evolutionary Computation, 2016,20(1): 16-37．
［10］ ZITZLER E, SIMON K. Indicatorbased selection in multiobjective search ［C］∥ Conference on Parallel Problem Solving from Nature (PPSN VIII). UK: Springer, 2004: 832-842．
［11］ BEUME N, NAUJOKS B, EMMERICH M. SMSEMOA: multiobjective selection based on dominated hypervolume ［J］. European Journal of Operational Research, 2007, 181(3): 1653-1669．
［12］ WHILE L, BRADSTREET L, BARONE L.A fast way of calculating exact hypervolumes ［J］. IEEE Transactions on Evolutionary Computation, 2012, 16(1): 86-95．
［13］ ZHANG Q, LI H. MOEA/D: a multiobjective evolutionary algorithm based on decomposition ［J］. IEEE Transactions on Evolutionary Computation, 2007, 11(6): 712-731．
［14］ TAN Yanyan, JIAO Yongchang, LI Hong. MOEA/D +uniform design: a new version of MOEA/D for optimization problems with many objectives ［J］. Computers and Operations Research, 2013, 40(6): 1648-1660．
［15］ JIANG S, CAI Z, ZHANG J, et al. Multiobjective optimization by decomposition with Paretoadaptive weight vectors ［C］∥ Proceeding of 7th International Conference on Natural Computation. Shanghai: IEEE, 2011: 1260-1264．
［16］ LI H, ZHANG Q. Multiobjective optimization problems with complicated Pareto sets, MOEA/D and NSGAII ［J］. IEEE Transactions on Evolutionary Computation, 2009, 13(2): 284-302．
［17］ ZHOU A, ZHANG Q, ZHANG G. A multiobjective evolutionary algorithm based on decomposition and probability model ［C］∥ IEEE Congress on Evolutionary Computation. ［S.l.］: IEEE, 2012: 18.
［18］ QI Y. MOEA/D with adaptive weight adjustment ［J］. Evolutionary Computation, 2014, 22(2): 231-264．
［19］ LIU Hailin, GU Fangqing, ZHANG Qingfu. Decomposition of a multiobjective optimization problems into a number of simple multiobjective subproblems ［J］. IEEE Transactions on Evolutionary Computation, 2014, 18(3): 450-455．
［20］ Deb K, THIELE L, LAUMANNS M, et al. Scalable multiobjective optimization test problems ［C］∥IEEE Congress on Evolutionary Computation (CEC’02). ［S.l.］: IEEE, 2002: 825-830．
［21］ ZHANG Q, ZHOU A, ZHAO S, et al. Multiobjective optimization test instances for the CEC 2009 special session and competition ［R］. Clemson: University of Essex, 2008．
［22］ 代才. 基于分解的多目标进化算法研究［D］. 西安：西安电子科技大学,2014: 50-70．
DAI Cai. Research on evolutionary algorithms based on decomposition for manyobjective optimization problems ［D］. Xi'an: Xidian University, 2014: 50-70.
［23］ ZITZLER E, LAUMANNS M, THIELE L. Performance assessment of multiobjective optimizers: analysis and review ［J］. IEEE Transactions on Evolutionary Computation, 2003, 7(2): 117133. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
