基于多记忆抗体克隆选择的人工免疫网络算法

在优化克隆算法的研究中,针对传统的克隆选择算法存在收敛性差和局部最优问题,提出一种多记忆抗体克隆选择原理的人工免疫网络算法。在克隆选择算法的基础上通过引入替代阀值因子,利用随机生成的新抗体组成种群替代原种群中对抗原亲和力最小抗体,同时增设变异概率的概念,达到在一定程度上避免记忆抗体种群的退化现象,提高算法的全局优化能力,避免陷入局部最优。仿真结果表明,算法加快了种群亲和力成熟的进程,随着进化代数的增加检测率总体呈上升趋势,能更好的应用于大规模各种识别问题中。

Algorithm of Artificial Immune Network Based on Multiple Memory Antibody Clonal Selection

To overcome the shortcomings in traditional clonal selection algorithm,this paper proposed an artificial immune network algorithm based on multiple memory antibody clonal selection principle of antibody of.Based on clonal selection algorithm and by introducing replacement threshold factor,the algorithm used the new antibodies generated randomly to replace the original species composition of the population of the smallest antigen affinity antibodies,and meanwhile the concept of an additional mutation probability was introduced to avoid the degradation of memory antibody population and improve the ability of global optimization algorithms to avoid falling into local optimum.Simulation results show that the global optimization algorithm can be better applied to large-scale problems.