| 新型复杂进化全局优化算法的研究 |
| |
| 引用本文： | 郝海青,马航,庄健.新型复杂进化全局优化算法的研究[J].电子学报,2013,41(4):704-709. |
| |
| 作者姓名： | 郝海青 马航 庄健 |
| |
| 作者单位： | 1. 华南理工大学机械与汽车工程学院,广东广州 510006;2. 广州大学土木工程学院,广东广州 510006;3. 南通大学附属第二医院呼吸内科,江苏南通226001;4. 西安交通大学机械学院,陕西西安 710049 |
| |
| 基金项目： | 广东省战略性新兴产业核心技术攻关项目，陕西省科技厅工业公关项目，中央高校基本科研业务费专项资金 |
| |
| 摘 要： | 文章在复杂系统思想激励下设计了一种新型的基于复杂系统改进的进化算法,该算法改进了进化算法的交叉、选择、变异和进化策略,体现了进化过程中能量分布、空间搜索、信息利用的复杂性,并保持了进化算法的简单结构框架;进而通过计算实例分析了新型复杂系统进化算法两个主要参数对算法性能影响;最后,新算法测试了CEC'2012大规模全局优化竞赛中的函数集合,并与其他优秀算法的测试结果进行了对比,结果表明本文所提出的基于复杂系统改进的进化算法综合性能强于所有的对比算法.
|
| 关 键 词： | 复杂系统 进化算法 大规模 全局优化 |
| 收稿时间： | 2012-04-05 |
The Study of a New Complex System Evolutionary Algorithm for Global Optimal Problems |
| |
| HAO Hai-qing
,
MA Hang
,
ZHUANG Jian.The Study of a New Complex System Evolutionary Algorithm for Global Optimal Problems[J].Acta Electronica Sinica,2013,41(4):704-709. |
| |
| Authors: | HAO Hai-qing MA Hang ZHUANG Jian |
| |
| Affiliation: | 1. School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou, Guangdong 510006, China;2. School of Civil Engineering, Guangzhou University, Guangzhou, Guangdong 510006, China;3. The Respiratory Department of the Second Affiliated Hospital of Nantong University, Nantong, Jiangsu 226001, China;4. School of Mechanical Engineering, Xi'an Jiaotong University, Xi'an, Shaanxi 710049, China |
| |
| Abstract: | Inspired by some complex system concepts,a new evolutionary algorithm is designed in paper.In the new algorithm,crossover,selecting operator,mutating operator and evolutionary strategy are improved by some characteristic properties of complex system.These improvements represent the complexity of energy distribution,space searching and information use during the evolution.And the simple procedure structure of evolutionary algorithm is reserved.Then,the influences of two parameters are discussed in paper.Finally,all functions of CEC'2012 competition on large scale problems are tested by the new evolutionary algorithm.And testing results of the new evolutionary algorithm are compared with those of other excellent algorithms.The results show that comprehensive capabilities of the new evolutionary algorithm,presented in the paper,are better than all comparing algorithms. |
| |
| Keywords: | complex system evolution algorithm large scale global optimization |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《电子学报》浏览原始摘要信息 |
|
点击此处可从《电子学报》下载全文 |
|
