J4 ›› 2013, Vol. 40 ›› Issue (2): 48-53.doi: 10.3969/j.issn.1001-2400.2013.02.009

• 研究论文 • 上一篇    下一篇

一种实用的类电磁机制算法

姜建国1;王双记1,2;刘永青1;尚海豹1;朱溢能1   

  1. (1. 西安电子科技大学 计算机学院,陕西 西安  710071;
    2. 中国人民解放军91388部队,广东 湛江  524022)
  • 收稿日期:2011-09-27 出版日期:2013-04-20 发布日期:2013-05-22
  • 作者简介:姜建国(1956-),男,教授,E-mail: jgjiang@mail.xidian.edu.cn.
  • 基金资助:

    国家部委基础科研计划资助项目(D1120060967)

Practical electromagnetism-like mechanism algorithm

JIANG Jianguo1;WANG Shuangji1,2;LIU Yongqing1;SHANG Haibao1;ZHU Yineng1   

  1. (1. School of Computer Science and Technology, Xidian Univ., Xi'an  710071, China;
    2. Unit No. 91388 of PLA, Zhanjiang  524022, China)
  • Received:2011-09-27 Online:2013-04-20 Published:2013-05-22

摘要:

分析了类电磁机制算法的函数优化原理,针对原算法中存在的运算量大、收敛速度慢、合力计算公式病态性等问题,提出了一种新的实用的类电磁机制算法.采用归一化目标函数值方法,简化了电量计算公式;引入合力计算修正因子,消除了合力计算公式的病态性,削弱了距离因素对种群粒子受力的影响,加快了收敛速度.实验证明,改进后的算法具有更好的收敛效果和更高的执行效率.

关键词: 类电磁机制算法, 函数优化, 修正因子

Abstract:

The function optimization principle of the original Electromagnetism-like mechanism (EM) algorithm and some problems of it are analyzed, such as large computation, low convergence rate, and pathological feature in the total force formula. A new pratical EM algorithm is proposed according to the problems above. The power formula is simplified by adopting the method for normalizing the objective function value. The correction factor is introduced into the total force formula, so that the pathological feature in the total force formula is eliminated, the impact of the distance factor on the force between population particles weakened, and the convergence rate increased. Experiments show that the improved algorithm has a faster convergence speed and higher solution accuracy.

Key words: electromagnetism-like mechanism algorithm, function optimization, correction factor

中图分类号: 

  • TP301.6