| 基于临界自组织模型的进化算法 |
| |
| 引用本文： | 关志华,寇纪淞,李敏强.基于临界自组织模型的进化算法[J].系统工程学报,2004,19(3):321-325. |
| |
| 作者姓名： | 关志华 寇纪淞 李敏强 |
| |
| 作者单位： | 1. 天津大学系统工程研究所,天津,300072;天津工业大学管理学院,天津,300160
2. 天津大学系统工程研究所,天津,300072 |
| |
| 基金项目： | 国家自然科学基金资助项目(69974026，70171002). |
| |
| 摘 要： | 将临界自组织(self-organlzed criticality，SOC)概念引入到进化算法领域，提出了基于临界自组织的进化算法(SOCEA)，引入灾变机制来模拟临界自组织状态中的剧烈变化过程，采用临界自组织的模型构造进化算法的算子、通过实验测试可以得出SOCEA算法要优于同类的对照算法，并且计算量开销也比较小、
|
| 关 键 词： | 临界自组织模型 进化算法 遗传算子 |
| 文章编号： | 1000-5781(2004)03-0321-05 |
Evolutionary algorithm based on self-organized criticality model |
| |
| GUAN Zhi-hua.Evolutionary algorithm based on self-organized criticality model[J].Journal of Systems Engineering,2004,19(3):321-325. |
| |
| Authors: | GUAN Zhi-hua |
| |
| Institution: | GUAN Zhi-hua~ |
| |
| Abstract: | This paper applies the concept of self-organized criticality (SOC) to evolutionary algorithms (EAs) and their operator, and presents a new SOCEA (evolutionary algorithm based on self-organized criticality model ) algorithm. The paper introduces the mass extinction mechanism to simulate the process of the violent changes of SOC, and applies the model of SOC to build the operators of EAs. The results show that the SOCEA clearly outperforms the standard EAs. The great performance of this algorithm is based on a very simple modification of the standard EAs, and so the new algorithm adds almost no extra computational costs. |
| |
| Keywords: | self-organized criticality model evolutionary algorithm genetic operators |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
