计算机与现代化 ›› 2011, Vol. 1 ›› Issue (1): 42-3.doi: 10.3969/j.issn.1006-2475.2011.01.012

• 算法分析与设计 • 上一篇    下一篇

基于蛛网结构的改进遗传算法研究

钟郁楠   

  1. 中国矿业大学电子信息科学与技术系,江苏 徐州 221116
  • 收稿日期:2010-09-17 修回日期:1900-01-01 出版日期:2011-01-20 发布日期:2011-01-20

An Improved Genetic Algorithm Based on Structure of Cobweb

ZHONG Yu-nan   

  1. Department of Electronic Information Science, China University of Mining and Technology, Xuzhou 221116, China
  • Received:2010-09-17 Revised:1900-01-01 Online:2011-01-20 Published:2011-01-20

摘要:

针对遗传算法对搜索空间变化的适应能力差,提出一种改进的遗传算法。根据蜘蛛在蜘蛛网上快速抓捕猎物的行为进行深入研究,并将其与GA模型相结合。通过改进遗传算法的初始化方式和适应度的评估方式,增强遗传算法对搜索空间变化的适应能力。对复杂函数的寻优实验验证改进后的遗传算法的有效性,并通过与传统SGA的实际计算结果相对比,更进一步说明算法的改进效果。

关键词: 遗传算法, 蛛网结构, De Jong函数, 适应性, 基本遗传算法

Abstract:

An improved Genetic Algorithm model is described based on the structure of cobweb. To abate the disadvantage related to the fitness of genetic algorithm, the improved genetic algorithm model is combined with several optimal methods brought from the study concerning the structure of cobweb. The improved genetic algorithm model improves the initialization and fitness assessment successfully. By the comparison between two results conducted by the improved genetic algorithm model and the standard model SGA respectively, the accessibility of the new genetic algorithm model is verified. 

Key words: Genetic Algorithm(GA), structure of Cobweb, function De Jong, fitness, SGA