摘要
提出了一个并行自适应动态克隆选择算法。在算法的每次迭代过程中,首先动态地计算出每个抗体的变异概率,然后根据抗体的亲和度将抗体种群动态分为记忆单元和一般抗体单元,紧接着以球面杂交方式对种群进行调整,从而加快了算法的全局搜索速度。同时,针对算法计算量大、耗时长的缺点,结合已有的并行计算技术,构造出了算法的并行计算方法。实例验证了所提并行算法的有效性、可行性。
A Parallel Adaptive Dynamic Clone Selection Algorithm（PADCSA） is proposed.Firstly, mutation probability of each antibody is dynamically calculated.And then, according to antibody affinity, antibody populations are dynamically divided into memory antibody units and general antibody units.Subsequently, antibody populations are adjusted by sphere crossover so that global search speed of the proposed algorithm is accelerated.Meanwhile, according to larger calculation and longer consumed time,parallel computation technology is introduced into the provided algorithm too.The effectiveness and the feasibility of the proposed parallel algorithm are verified by examples.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第25期37-39,63,共4页
Computer Engineering and Applications
基金
人工智能四川省(高校)重点实验室科研基金项目(No.2008RK010)
关键词
克隆选择
变异概率
抗体亲和度
并行计算技术
clone selection
mutation probability
antibody affinity
parallel computation technology
