计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (33): 78-81.

• 学术探讨 • 上一篇    下一篇

基于人类繁殖现象的遗传算法研究

严太山1,2,陶永芹1,崔杜武1   

  1. 1.西安理工大学 计算机科学与工程学院,西安 710048
    2.湖南理工学院 计算机系,湖南 岳阳 414000
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-11-21 发布日期:2007-11-21
  • 通讯作者: 严太山

Research on genetic algorithm based on human reproduction phenomenon

YAN Tai-shan1,2,TAO Yong-qin1,CUI Du-wu1   

  1. 1.School of Computer Science and Engineering,Xi’an University of Technology,Xi’an 710048,China
    2.Department of Computer Science,Hunan Institute of Technology,Yueyang,Hunan 414000,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-21 Published:2007-11-21
  • Contact: YAN Tai-shan

摘要: 标准遗传算法(SGA)只是对自然界遗传进化过程的比较简单的模拟,较少考虑人类特有的繁殖方式。提出一种基于人类繁殖现象的遗传算法(HRGA),该算法的遗传算子包括选择算子、助长算子、交叉算子和变异算子,遗传个体具有雄性和雌性两种不同的性别,融合了个体的年龄和个体间的亲缘关系两种特征,在允许的年龄范围内,异性个体进行严格的远缘繁殖,从而克服了标准遗传算法容易出现的早熟收敛现象,提高了算法的收敛速度。通过对函数最优化问题的求解试验,证明了该算法具有很强的跳出局部收敛的能力,其全局收敛速度和最优解的质量明显高于标准遗传算法,同时也证明了该算法的有效性。

关键词: 遗传算法, 人类繁殖现象, 函数最优化, 全局最优解

Abstract: Standard Genetic Algorithm(SGA) only simulates natural evolution process simply and considers human own reproduction mode less.A genetic algorithm based on Human Reproduction Phenomenon(HRGA) has been proposed in this paper.The genetic operators of this algorithm include selection operator,help operator,crossover operator and mutation operator.The genetic individuals are separated into male individual and female individual,the age feature and consanguinity feature are fused into individuals.Two individuals with opposite sex can reproduce the next generation if they are distant consanguinity individuals and their age is allowable.So,the premature convergence of standard genetic algorithm is overcome and the convergence speed of algorithm is enhanced.Experiments have been taken on function optimization.The experimental results show that this algorithm is very strong to avoid dropping into local convergence,the global convergence speed and optimal solutions are all better than that of standard genetic algorithm.Meanwhile,the validity of this algorithm has been proved.

Key words: genetic algorithm, human reproduction phenomenon, function optimization, global optimal solution