| 基于个体密集距离的多目标进化算法 |
| |
| 引用本文： | 雷德明,吴智铭.基于个体密集距离的多目标进化算法[J].计算机学报,2005,28(8):1320-1326. |
| |
| 作者姓名： | 雷德明 吴智铭 |
| |
| 作者单位： | 上海交通大学自动化研究所,上海,200030 |
| |
| 基金项目： | 本课题得到国家自然科学基金（60074011,70071017）资助. |
| |
| 摘 要： | 外部种群维护和适应度赋值是多目标进化算法（MOEA）的两个重要部分，该文首先对这两个问题目前已有的处理方法进行了分析，然后提出了基于个体密集距离的外部种群维护方法，并在将所有个体根据Pareto支配关系分成四个层次的基础上，给出了一种由个体密集距离定义的适应度函数，最后将基于个体密集距离的多目标进化算法CMOEA应用于几个常用的测试函数，并和SPEA，SPEA-2进行了比较，计算结果表明CMOEA具有良好的搜索性能．
|
| 关 键 词： | 密集距离 维护 适应度赋值 多目标进化算法 |
| 收稿时间： | 2004-06-04 |
| 修稿时间： | 2004-06-04 |
Crowding-Measure Based Multi-Objective Evolutionary Algorithm |
| |
| LEI De-ming,WU Zhi-Ming.Crowding-Measure Based Multi-Objective Evolutionary Algorithm[J].Chinese Journal of Computers,2005,28(8):1320-1326. |
| |
| Authors: | LEI De-ming WU Zhi-Ming |
| |
| Abstract: | External population maintenance and fitness assignment are the two main parts of multi-objective evolutionary algorithm. In this paper, the authors first analyze the related works on the above problems. Secondly, an external population maintenance method in terms of the crowding measure of individual is proposed, and then all individuals are categorized into four kinds according to Pareto dominance relationship and the crowding-measure based fitness function is defined. Finally, the proposed algorithm CMOEA are applied to seven test functions and the comparison between CMOEA, SPEA and SPEA-2 demonstrates that CMOEA is provided with good performance. |
| |
| Keywords: | crowding measure maintenance fitness assignment multi-objective evolutionaryalgorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
