融合对数交叉概率因子和随机迁移的差分进化算法  

Differential Evolution Algorithm Merging Logarithm Crossing Probability and Random Migration

作　　者：马小华[1] 李济民[1] 

机构地区：[1]北方民族大学信息与系统科学研究所,宁夏银川750021

出　　处：《太原理工大学学报》2011年第3期228-231,共4页Journal of Taiyuan University of Technology

基　　金：国家自然科学基金项目资助(60962006);宁夏高等学校科研项目(2009JY008);国家民委科研项目资助(10BF07)

摘　　要：针对差分进化算法在后期收敛缓慢和易陷入局部极值缺点,提出了一种带有对数递增交叉概率因子和随机迁移算子的差分进化算法。这个算法增强了收敛速度和精度,同时也提高了全局寻优能力。数值实验结果表明,所提出的算法LMDE比基本DE和带混沌差分进化算法CDE在收敛性和稳健性以及全局寻优能力方面更好。For the shortcomings of being slow in convergence in the late and easy to fall into local minima,a differential evolution algorithm merging logarithm crossing probability and random migration was presented.This algorithm enhanced the convergence speed and precision,and also improved the ability of global optimization.Numerical results show that the proposed algorithm LMDE was better than the basic differential evolution algorithm DE and the chaotic differential evolution algorithm CDE in terms of convergence and robustness as well as the ability of global optimization.

关 键 词：差分进化 自适应交叉概率 随机迁移 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]