首页 | 本学科首页   官方微博 | 高级检索  
     

可求解多目标优化问题的演化博弈优化算法
引用本文:徐敏,张敏,王煦法.可求解多目标优化问题的演化博弈优化算法[J].小型微型计算机系统,2007,28(4):640-644.
作者姓名:徐敏  张敏  王煦法
作者单位:中国科学技术大学计算机科学与技术系 安徽合肥230027
基金项目:国家自然基金委员会海外青年学者合作研究基金(60428202)资助.
摘    要:借鉴演化博弈的思想和选择机制,提出了一种新的基于演化博弈的优化算法(EGOA)用于多目标问题的求解.算法框架具备对该类问题的通用性.为了对算法性能进行评估,采用了一组多目标优化问题(MOPs)的测试函数进行实验.实验结果表明,使用本算法搜索得到的演化稳定策略集合能够很好地逼近多目标优化问题的帕累托前沿,与一些经典的演化算法相比具有良好的问题求解能力.

关 键 词:多目标优化问题  演化博弈理论  复制者动态  演化算法
文章编号:1000-1220(2007)04-0640-05
修稿时间:2006-01-06

Evolutionary Game Based Optimization Algorithm for Multi-objective Optimization Problems
XU Min, ZHANG Min, WANG Xu-fa.Evolutionary Game Based Optimization Algorithm for Multi-objective Optimization Problems[J].Mini-micro Systems,2007,28(4):640-644.
Authors:XU Min  ZHANG Min  WANG Xu-fa
Abstract:Used the idea and the selection mechanism called replicator dynamics of evolutionary game theory for reference, and proposed an evolutionary game based optimization algorithm (EGOA) to solve multiobjective optimization problems (MOPs). The frame of algorithm has universal character for MOPs. To evaluate the newly designed approach, solved a group of test functions. The experiment results showed that the evolutionary stable strategy (ESS) set searched by EGOA is very close to the Pareto front of MOPs. Compared with some classic evolutionary algorithm, has a good ability of problem solving.
Keywords:multiobjective optimization problem  evolutionary game theory  replicator dynamics  evolutionary algorithm
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号