环境选择的双种群约束多目标狼群算法

doi:10.3778/j.issn.1002-8331.2502-0187

摘要/Abstract

摘要： 针对多目标狼群算法存在的搜索不充分、收敛性不足和多样性欠缺的问题，以及缺少对约束进行处理的问题，提出环境选择的双种群约束多目标狼群算法（multi-objective wolf pack algorithm for dual population constraints with environment selection，DCMOWPA-ES）。引入双种群约束处理方法给种群设置不同的搜索偏好，主种群运用可行性准则优先保留可行解，次种群通过[ε]约束探索不可行区域并将搜索结果传递给主种群，让算法能较好应对复杂的不可行区域，保障算法的可行性；提出维度选择的随机游走策略，使人工狼可自主选择游走方向，提高种群的全局搜索能力；设计精英学习的步长调整机制，人工狼通过向头狼学习的方式提升种群的局部搜索能力，确保算法的收敛性；采用环境选择的狼群更新策略，根据人工狼被支配的情况和所处位置的密度信息对其赋值，选择被支配数少且密度信息小的人工狼作为优秀个体，改善算法的多样性。为验证算法性能，将DCMOWPA-ES与六种新兴约束多目标优化算法在两组约束多目标测试集和汽车侧面碰撞设计问题上进行对比实验。实验结果表明，DCMOWPA-ES算法具备较好的可行性、收敛性和多样性。

关键词: 狼群算法, 双种群约束, 维度选择, 精英学习, 环境选择, 约束多目标优化

Abstract: To address the problems of insufficient search capability, poor convergence, and limited diversity in multi-objective wolf pack algorithms, along with inadequate constraint handling, a multi-objective wolf pack algorithm for dual population constraints with environment selection (DCMOWPA-ES) is proposed. A dual population constraint handling mechanism allocates distinct search preferences, the main population uses the constraint domination principle to prioritize the feasible solutions, and the secondary population explores the infeasible region through [ε] constraint handling approach and passes the search results to the main population, so that the algorithm can better deal with complex infeasible areas and ensure the feasibility of the algorithm. A random walk strategy of dimension selection is proposed, so that the artificial wolf can choose the walking direction independently and improve the global search ability of the population. The step size adjustment mechanism of elite learning is designed, and the artificial wolf improves the local search ability of the population by learning from the head wolf to ensure the convergence of the algorithm. The wolf pack update strategy of environment selection is adopted, and the artificial wolf is assigned a value according to the domination of the artificial wolf and the density information of the location, and the artificial wolf with a small number of domination and small density information is selected as an excellent individual to improve the diversity of the algorithm. To verify the performance of the algorithm, DCMOWPA-ES is compared with six emerging constrained multi-objective optimization algorithms on two constrained multi-objective test sets and the car-side impact design problem. The experimental results show that the DCMOWPA-ES algorithm has good feasibility, convergence and diversity.

Key words: wolf pack algorithm, dual population constraint, dimension selection, elite learning, environment selection, constrained multi-objective optimization

吕莉, 杨凌锋, 肖人彬, 孟振宇, 崔志华, 王晖. 环境选择的双种群约束多目标狼群算法[J]. 计算机工程与应用, 2025, 61(16): 116-131.

LYU Li, YANG Lingfeng, XIAO Renbin, MENG Zhenyu, CUI Zhihua, WANG Hui. Multi-Objective Wolf Pack Algorithm for Dual Population Constraints with Environment Selection[J]. Computer Engineering and Applications, 2025, 61(16): 116-131.

参考文献

[1] HAN S, SUN X M. Optimizing product design using genetic algorithms and artificial intelligence techniques[J]. IEEE Access, 2024, 12: 151460-151475.
[2] 侯艳, 任丙飞, 滕少华, 等. 带罐约束的多目标短期炼油调度优化研究[J]. 江西师范大学学报 (自然科学版), 2023, 47(3): 307-316.
HOU Y, REN B F, TENG S H, et al. The multi-objective short-term scheduling optimization with charging-tank-switch-overlap constraint in refinery[J]. Journal of Jiangxi Normal University (Natural Science Edition), 2023, 47(3): 307-316.
[3] YU W, LI B Z, JIA H Y, et al. Application of multi-objective genetic algorithm to optimize energy efficiency and thermal comfort in building design[J]. Energy and Buildings, 2015, 88: 135-143.
[4] 曹嘉乐, 杨磊, 田井林, 等. 面向高维多目标优化的双阶段双种群进化算法[J]. 计算机工程与应用, 2024, 60(9): 159-171.
CAO J L, YANG L, TIAN J L, et al. Dual-stage dual-population evolutionary algorithm for many-objective optimization[J]. Computer Engineering and Applications, 2024, 60(9): 159-171.
[5] 梁正平, 王侃, 周倩, 等. 基于进化多任务的稀疏大规模多目标优化[J]. 计算机学报, 2025, 48(2): 358-380.
LIANG Z P, WANG K, ZHOU Q, et al. Evolutionary multi-task for sparse large-scale multi-objective optimization[J]. Chinese Journal of Computers, 2025, 48(2): 358-380.
[6] 赵嘉, 钟劲文, 肖人彬, 等. 随机游走和特殊拥挤距离更新的多模态多目标狼群算法[J/OL]. 计算机工程与应用 [2024-11-07]. https://kns.cnki.net/kcms/detail/11.2127.TP. 20241106.
1618.024.html.
ZHAO J, ZHONG J W, XIAO R B, et al. Multi-modal multi-objective wolf pack algorithm with random walk and special crowding distance update[J/OL]. Computer Engineering and Applications [2024-11-07]. https://kns.cnki.net/kcms/detail/11.2127.TP.20241106.1618.024.html.
[7] 于琨, 张正本, 海本斋. 基于多目标全局约束的任务分配和调度算法[J]. 计算机工程与应用, 2018, 54(8): 55-60.
YU K, ZHANG Z B, HAI B Z. Global constraints for multi-objective based task allocation and scheduling algorithm[J]. Computer Engineering and Applications, 2018, 54(8): 55-60.
[8] QIAO K J, LIANG J, YU K J, et al. Constraints separation based evolutionary multitasking for constrained multi-objective optimization problems[J]. IEEE/CAA Journal of Automatica Sinica, 2024, 11(8): 1819-1835.
[9] WANG B C, LI H X, FENG Y, et al. An adaptive fuzzy penalty method for constrained evolutionary optimization[J]. Information Sciences, 2021, 571: 358-374.
[10] ZOU J, SUN R Q, YANG S X, et al. A dual-population algorithm based on alternative evolution and degeneration for solving constrained multi-objective optimization problems[J]. Information Sciences, 2021, 579: 89-102.
[11] FAN Z, LI W J, CAI X Y, et al. Push and pull search for solving constrained multi-objective optimization problems[J]. Swarm and Evolutionary Computation, 2019, 44: 665-679.
[12] COELLO C A C, PULIDO G T, LECHUGA M S. Handling multiple objectives with particle swarm optimization[J]. IEEE Transactions on Evolutionary Computation, 2004, 8(3): 256-279.
[13] 肖人彬, 陈峙臻. 从群智能优化到群智能进化[J]. 南昌工程学院学报, 2023, 42(1): 1-10.
XIAO R B, CHEN Z Z. From swarm intelligence optimization to swarm intelligence evolution[J]. Journal of Nanchang Institute of Technology, 2023, 42(1): 1-10.
[14] XIAO R B. Four development stages of collective intelligence[J]. Frontiers of Information Technology & Electronic Engineering, 2024, 25(7): 903-916.
[15] MING F, GONG W Y, LI D C, et al. A competitive and cooperative swarm optimizer for constrained multiobjective optimization problems[J]. IEEE Transactions on Evolutionary Computation, 2023, 27(5): 1313-1326.
[16] 吴虎胜, 张凤鸣, 吴庐山. 一种新的群体智能算法: 狼群算法[J]. 系统工程与电子技术, 2013, 35(11): 2430-2438.
WU H S, ZHANG F M, WU L S. New swarm intelligence algorithm: wolf pack algorithm[J]. Systems Engineering and Electronics, 2013, 35(11): 2430-2438.
[17] JIANG H, YU Q Z, HAN D, et al. A path planning method for unmanned aerial vehicle based on improved wolf pack algorithm[J]. Concurrency and Computation: Practice and Experience, 2024, 36(14): e8095.
[18] KULIEV E V, ZAPOROZHETS D Y, KUREICHIK V V, et al. Wolf pack algorithm for solving VLSI design tasks[J]. Journal of Physics: Conference Series, 2019, 1333(2): 022009.
[19] 赵嘉, 吕丰, 肖人彬, 等. 自适应分组和拥挤距离更新的多目标狼群算法[J]. 控制与决策, 2024, 39(11): 3772-3780.
ZHAO J, LV F, XIAO R B, et al. Multi-objective wolf pack algorithm based on adaptive grouping strategy and crowding distance[J]. Control and Decision, 2024, 39(11): 3772-3780.
[20] WANG L H, NG A H C, DEB K. Multi-objective evolutionary optimisation for product design and manufacturing[M]. London: Springer, 2011.
[21] DEB K, PRATAP A, AGARWAL S, et al. A fast and elitist multiobjective genetic algorithm: NSGA-II[J]. IEEE Transactions on Evolutionary Computation, 2002, 6(2): 182-197.
[22] TAKAHAMA T, SAKAI S. Constrained optimization by the ε constrained differential evolution with gradient-based mutation and feasible elites[C]//Proceedings of the 2006 IEEE International Conference on Evolutionary Computation. Piscataway: IEEE, 2006: 1-8.
[23] RUNARSSON T P, YAO X. Stochastic ranking for constrained evolutionary optimization[J]. IEEE Transactions on Evolutionary Computation, 2000, 4(3): 284-294.
[24] PANICHELLA A. An improved Pareto front modeling algorithm for large-scale many-objective optimization[C]//Proceedings of the 2022 Genetic and Evolutionary Computation Conference. New York: ACM, 2022: 565-573.
[25] SUN R Q, ZOU J, LIU Y, et al. A multistage algorithm for solving multiobjective optimization problems with multiconstraints[J]. IEEE Transactions on Evolutionary Computation, 2023, 27(5): 1207-1219.
[26] LIU Z Z, WANG B C, TANG K. Handling constrained multiobjective optimization problems via bidirectional coevolution[J]. IEEE Transactions on Cybernetics, 2022, 52(10): 10163-10176.
[27] QIAO K J, LIANG J, LIU Z Y, et al. Evolutionary multitasking with global and local auxiliary tasks for constrained multi-objective optimization[J]. IEEE/CAA Journal of Automatica Sinica, 2023, 10(10): 1951-1964.
[28] MING F, GONG W Y, WANG L, et al. Constrained multi-objective optimization via multitasking and knowledge transfer[J]. IEEE Transactions on Evolutionary Computation, 2024, 28(1): 77-89.
[29] TIAN Y, WANG R Q, ZHANG Y J, et al. Adaptive population sizing for multi-population based constrained multi-objective optimization[J]. Neurocomputing, 2025, 621: 129296.
[30] TIAN Y, ZHU W J, ZHANG X Y, et al. A practical tutorial on solving optimization problems via PlatEMO[J]. Neurocomputing, 2023, 518: 190-205.
[31] COELLO C A C, CORTéS N C. Solving multiobjective optimization problems using an artificial immune system[J]. Genetic Programming and Evolvable Machines, 2005, 6(2): 163-190.
[32] ZITZLER E, THIELE L, LAUMANNS M, et al. Performance assessment of multiobjective optimizers: an analysis and review[J]. IEEE Transactions on Evolutionary Computation, 2003, 7(2): 117-132.
[33] MA Z W, WANG Y. Evolutionary constrained multiobjective optimization: test suite construction and performance comparisons[J]. IEEE Transactions on Evolutionary Computation, 2019, 23(6): 972-986.
[34] FAN Z, LI W J, CAI X Y, et al. An improved epsilon constraint-handling method in MOEA/D for CMOPs with large infeasible regions[J]. Soft Computing, 2019, 23: 12491-12510.
[35] DEB K, JAIN H. An evolutionary many-objective optimization algorithm using reference-point-based nondominated sorting approach, part II: handling constraints and extending to an adaptive approach[J]. IEEE Transactions on Evolutionary Computation, 2014, 18(4): 602-622.
[36] KUMAR A, WU G H, ALI M Z, et al. A benchmark-suite of real-world constrained multi-objective optimization problems and some baseline results[J]. Swarm and Evolutionary Computation, 2021, 67: 100961.