| 基于递阶和小生境的离散分段遗传算法 |
| |
| 引用本文： | 高婷,鲁汉榕,乐艳丽,史原.基于递阶和小生境的离散分段遗传算法[J].计算机工程与设计,2007,28(11):2646-2648. |
| |
| 作者姓名： | 高婷 鲁汉榕 乐艳丽 史原 |
| |
| 作者单位： | 1. 空军雷达学院,研究生管理大队,湖北,武汉,430019
2. 空军雷达学院,信息与指挥自动化系,湖北,武汉,430019 |
| |
| 摘 要： | 为了解决简单遗传算法过早收敛的问题,并进一步改善简单遗传算法的寻优质量,在分析递阶遗传算法和小生境遗传算法的基础上,提出了离散分段遗传算法.该方法在微观上,采用了递阶遗传算法的递阶编码方式和小生境的选择思想.宏观上,通过分层多级寻优操作来适当加快遗传算法的寻优速度.该算法非常适合解决多峰值优化问题,同时也能够有效地修复早熟现象的影响,加快收敛速度.实验表明该方法在性能方面明显优于简单遗传算法.
|
| 关 键 词： | 遗传算法 离散化 小生境遗传算法 递阶遗传算法 早熟现象 小生境遗传算法 离散 分段遗传算法 based genetic algorithm 性能 实验 收敛速度 影响 早熟现象 修复 优化问题 多峰值 操作 寻优 思想 的选择 编码方式 微观 方法 |
| 文章编号： | 1000-7024（2007）11-2646-03 |
| 修稿时间： | 2006-06-22 |
Discrete segmented genetic algorithm based on HGA and NGA |
| |
| GAO Ting,LU Han-rong,LE Yan-li,SHI Yuan.Discrete segmented genetic algorithm based on HGA and NGA[J].Computer Engineering and Design,2007,28(11):2646-2648. |
| |
| Authors: | GAO Ting LU Han-rong LE Yan-li SHI Yuan |
| |
| Affiliation: | 1. Group of Graduate Management, Air Force Radar Academy, Wuhan 430019, China; Department of Information and Command Automation, Air Force Radar Academy, Wuhan 430019, China |
| |
| Abstract: | In order to improve the ability of searching the optimum, based on an insight into the premature phenomenon of genetic algorithms, a genetic algorithm is introduced combining HGA and NGA. From the low level point of view, the algorithm adopts the hierarchical coding methods in HGA and the selection methods in NGA. From the high level point of view, multilevel action is used to improve the speed of SGA. The algorithm suits to solve the problem of multi-peak values. And it can amend the premature convergence problems. Tests show that it has better performance compared with the simple genetic algorithm. |
| |
| Keywords: | GA discretizafion NGA HGA premature phenomenon |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
