计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (30): 29-31.

• 研究、探讨 • 上一篇    下一篇

采用种群划分的动态自适应免疫克隆选择算法

马春华1,钟 勇2,3   

  1. 1.绥化学院 计算机科学与技术系,黑龙江 绥化 152061
    2.中国科学院 成都计算机应用研究所,成都 610041
    3.中国科学院 研究生院,北京 100039
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-10-21 发布日期:2011-10-21

Dynamic adaptive immune clone selection algorithm by applying population division

MA Chunhua1,ZHONG Yong2,3   

  1. 1.Computer Science and Technology Department,Suihua College,Suihua,Heilongjiang 152061,China
    2.Chengdu Institute of Computer Application,Chinese Academy of Sciences,Chengdu 610041,China
    3.The Graduate School of the Chinese Academy of Sciences,Beijing 100039,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-10-21 Published:2011-10-21

摘要: 为了克服传统免疫克隆选择算法的种群缺乏多样性、抗体选择不具随机性的缺点,提出了一种新型动态自适应免疫克隆选择算法。在该算法求解过程中,根据抗体的亲和度将抗体种群动态地分为记忆单元和一般抗体单元,以球面杂交方式对种群进行调整并动态修正每个抗体的变异概率,从而保障了群体多样性,加快了算法的全局搜索速度。实例验证了所提算法具有较好的性能。

关键词: 免疫克隆选择算法, 抗体亲和度, 球面杂交, 变异概率

Abstract: In traditional immune clone selection algorithms,population lacks diversity and antibody cannot be randomly selected.In order to overcome above-mentioned shortcomings,a new dynamic adaptive immune clonal selection algorithm is proposed.In the proposed algorithm,population is firstly dynamically divided into the memory antibody units and the general antibody units according to antibody affinity.And then,population is adjusted by means of sphere crossover.Meanwhile,mutation probability of each antibody is dynamically amended so that population diversity is obtained and global search speed is accelerated.The better performance of the proposed algorithm is verified by examples.

Key words: clone selection algorithm, antibody affinity, sphere crossover, mutation probability