您的位置:首页  > 论文页面

一种求解约束优化问题的差分进化算法

发表时间:2010-04-30  浏览量:1450  下载量:717
全部作者: 刘若辰,李勇,焦李成
作者单位: 智能感知与图像理解教育部重点实验室,西安电子科技大学智能信息处理研究所
摘 要: 基于差分进化(differential evolution,DE)算法的相关理论以及随机排序约束处理方法,提出一种解决约束优化问题(constrained optimization problem,COP)的算法。该算法对种群进化采用随机保留次优解的策略,有效地提高了种群多样性。对13个标准测试问题的测试结果表明:与动态惩罚函数的进化算法、人工免疫响应约束进化策略、可行性规则的差分进化算法以及采用随机排序的进化策略相比,新算法在收敛速度和求解精度上均具有一定的优势。
关 键 词: 人工智能;差分进化算法;约束优化;随机排序
Title: A differential evolution algorithm for constrained optimization problem
Author: LIU Ruochen, LI Yong, JIAO Licheng
Organization: Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education, Institute of Intelligent Information Processing, Xidian University
Abstract: Based on differential evolution and stochastic ranking strategy, a differential evolution algorithm for constrained optimization problem is proposed in this paper. The proposed algorithm reserves sub-optimal solutions in the process of population evolution, which effectively enhances the diversity of the population. The experiment results on 13 well-known benchmark problems show that the proposed algorithm is capable of improving the search performance significantly in convergent speed and precision with respect to four techniques representative of the state-of-the-art in constrained optimization such as evolutionary algorithm based on homomorphous maps, artificial immune response constrained evolutionary strategy, constraint handling differential evolution, and evolutionary strategies based on stochastic ranking.
Key words: artificial intelligent; differential evolution algorithm; constrained optimization; stochastic ranking
发表期数: 2010年4月第8期
引用格式: 刘若辰,李勇,焦李成. 一种求解约束优化问题的差分进化算法[J]. 中国科技论文在线精品论文,2010,3(8):830-837.
 
0 评论数 0
暂无评论
友情链接