| 一种求解约束多目标问题的协作进化算法 |
| |
| 引用本文： | 王蕊,顾清华. 一种求解约束多目标问题的协作进化算法[J]. 控制与决策, 2021, 36(11): 2656-2664 |
| |
| 作者姓名： | 王蕊 顾清华 |
| |
| 作者单位： | 西安建筑科技大学管理学院,西安710055;西安建筑科技大学管理学院,西安710055;西安建筑科技大学资源工程学院,西安710055 |
| |
| 基金项目： | 国家自然科学基金项目(51774228,51864046);陕西省自然科学基金杰青项目(2020JC-44);中国博士后科学基金项目(2019M662505). |
| |
| 摘 要： | 针对约束多目标进化算法求解约束多目标问题时难以平衡收敛性、多样性和可行性的问题,提出一种协作进化算法(ConMOEA).将自适应形状估计进化算法(AGE-MOEA)和非支配排序遗传算法(NSGA-Ⅱ)优势融合,采用Deb约束支配原则非支配排序组合种群实现个体优选,在临界层中根据最大拥挤距离或生存值选择所需个体,最终形成...
|
| 关 键 词： | 约束多目标优化 Deb约束支配 AGE-MOEA NSGA-Ⅱ 收敛性 多样性 |
A collaborative evolutionary algorithm for solving constrained multi-objective problems |
| |
| WANG Rui,GU Qing-hua. A collaborative evolutionary algorithm for solving constrained multi-objective problems[J]. Control and Decision, 2021, 36(11): 2656-2664 |
| |
| Authors: | WANG Rui GU Qing-hua |
| |
| Affiliation: | School of Management,Xián University of Architecture and Technology,Xián 710055,China; School of Management,Xián University of Architecture and Technology,Xián 710055,China;School of Resources Engineering, Xián University of Architecture and Technology,Xián 710055,China |
| |
| Abstract: | The balance of convergence, diversity and feasibility is a difficulty for the constrained multi-objective evolutionary algorithms. Thus, a collaborative constrained multi-objective algorithm (ConMOEA) is proposed, which integrates the advantages of the adaptive geometry estimation based MOEA (AGE-MOEA) and the non-dominated sorting genetic algorithm(NSGA II). Firstly, the Deb constraint dominance is applied to sort the combined population. Then the individuals in critical layer are selected according to the maximum crowding distance or individual survival score. Finally, a new population is formed that can fast approach the Pareto front and has good distribution. The effectiveness of the proposed algorithm is validated by comparing with NSGA-II-CDP, C-TAEA, PPS, ToP, A-NSGA-III, AGE-MOEA on the DOC test suit. And the performance of algorithms is evaluated by the inverted generational distance (IGD) and hypervolume (HV). The simulation results show that the ConMOEA has better convergence and diversity. |
| |
| Keywords: | |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《控制与决策》浏览原始摘要信息 |
|
点击此处可从《控制与决策》下载全文 |
|
