| 引用本文: | 侯中喜,陈小庆,郭良民.基于排挤机制改进的多目标进化算法.[J].国防科技大学学报,2006,28(4):18-21.[点击复制] |
| HOU Zhongxi,CHEN Xiaoqing,GUO Liangmin.An Improved Multi-objective Evolutionary Algorithm Based on Crowing Mechanism[J].Journal of National University of Defense Technology,2006,28(4):18-21[点击复制] |
|
| |
|
|
| 本文已被：浏览 6433次 下载 5927次 |
| 基于排挤机制改进的多目标进化算法 |
| 侯中喜, 陈小庆, 郭良民 |
|
(国防科技大学 航天与材料工程学院，湖南 长沙 410073)
|
| 摘要: |
| 进化算法是求解多目标优化问题(MOP)重要而有效的方法。为加快收敛速度，提高收敛精度，在已有算法(NSGA-Ⅱ)的基础上，引进小生境思想，提出了更为合理的排挤机制。通过典型应用函数的计算测试，结果表明：上述改进不仅具有较高的计算效率，而且能够得到分布更为合理的解，且能保持解的多样性分布。 |
| 关键词: 多目标优化 进化算法 排挤机制 |
| DOI： |
| 投稿日期：2006-01-11 |
| 基金项目:国家863基金资助项目(2005AA756050) |
|
| An Improved Multi-objective Evolutionary Algorithm Based on Crowing Mechanism |
| HOU Zhongxi, CHEN Xiaoqing, GUO Liangmin |
|
(College of Aerospace and Materials Engineering, National Univ. of Defense Technology, Changsha 410073, China)
|
| Abstract: |
| Evolutionary algorithms are the main and effective methods in solving multi-objective optimization problems (MOP). Based on the NSGA-Ⅱ algorithm studied and analyzed, we improved its crowding mechanism by introducing the Niche theory to expedite its convergence velocity and improve its convergence precision. The representative test functions show that the improvements have higher computational efficiency and can obtain a reasonable distributing solution; it can also maintain the solutions' diversity. |
| Keywords: multi-objective optimization evolutionary algorithm crowing mechanism |
|
|
