An improved self-adaptive memetic differential evolution (DE) algorithm

被引：0

作者：

Zhao, Fuqing ^{[1
,2
]}

Huo, Mingming ^{[1
]}

机构：

[1] School of Computer and Communication Technology, Lanzhou University of Technology, Lanzhou 730050, China

[2] Key Laboratory of Contemporary Design, Integrated Manufacturing Technology, Ministry of Education, Northwestern Polytechnical University, 710072, China

来源：

Journal of Information and Computational Science | 2012年 / 9卷 / 15期

关键词：

Evolutionary algorithms - Local search (optimization) - Adaptive algorithms;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

This paper proposes an improved self-adaptive memetic differential evolution algorithm(IMDE). In the aspects of population initialization and local search, the normal distribution model is introduced to improve the classic differential evolution algorithm in order to guarantees its higher optimizing efficiency and accuracy. The self-adaptive operators of mutation and crossover are introduced which not only improve the global convergence, but also guarantee the convergence speed of the algorithm. The simulation results show that IMDE has good global convergence and can avoid premature convergence effectively. © 2012 by Binary Information Press.

引用

页码：4321 / 4328