首页 | 本学科首页   官方微博 | 高级检索  
     检索      

用群体启发进化规划求解高维优化问题
引用本文:窦全胜,周春光,徐中宇,潘冠宇.用群体启发进化规划求解高维优化问题[J].吉林大学学报(理学版),2005,43(5):622-626.
作者姓名:窦全胜  周春光  徐中宇  潘冠宇
作者单位:(1.山东工商学院信息与电子工程学院,山东省烟台264005;2.吉林大学计算机科学与技术学院,长春130012;3.长春工业大学计算机科学与工程学院,长春130012)
基金项目:国家自然科学基金(批准号:60433020),教育部“符号计算与知识工程”重点实验室基金
摘    要:提出一种新的进化规划方法,群体启发进化规划(PHEP),在进化过程中,通过控制群体的4个参数,把握群体中个体分布情况,并通过这些信息有效地调整个体的变异步长,克服了传统EP方法变异步长修正的盲目性.将PHEP方法应用于高维优化问题,实验结果表明,PHEP方法在高维条件下的性能明显优于其他EP方法.

关 键 词:进化规划  高维优化  群体  
文章编号:1671-5489(2005)05-0622-05
收稿时间:2005-01-10
修稿时间:2005年1月10日

Population Heuristic Evolutionary Programming for High-dimension Optimization
DOU Quan-sheng,ZHOU Chun-guang,XU Zhong-yu,PAN Guan-yu.Population Heuristic Evolutionary Programming for High-dimension Optimization[J].Journal of Jilin University: Sci Ed,2005,43(5):622-626.
Authors:DOU Quan-sheng  ZHOU Chun-guang  XU Zhong-yu  PAN Guan-yu
Institution:(1.School of Information and Electronics Engineering,Shandong Institute of Business and Technology,Yantai264005,Shandong Province,China;2.College of Computer Science and Technology,Jilin University,Changchun130012,China;3.School of Computer Science and Engineering,Changchun University of Technology,Changchun130012,China)
Abstract:A new evolutionary programming method known as population heuristic evolutionary programming(PHEP) was proposed in this paper.The information of distribution-status of population can be known by(controlling) four parameters of population in the evolution process and the mutation size of individuals can be adjusted according to such information,so as to overcome the deficiency of traditional EP,which updates the(mutation) size blindly.PHEP was tested by using benchmark functions under high-dimension condition,the(experi)mental results show that the performance of PHEP is better than that of other EP method obviously(under) high-dimension condition.
Keywords:evolutionary programming  high-dimension optimization  population
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《吉林大学学报(理学版)》浏览原始摘要信息
点击此处可从《吉林大学学报(理学版)》下载免费的PDF全文
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