摘要: 提出一种新的动态自适应克隆选择并行算法。在每次迭代过程中,动态计算每个抗体的变异概率,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,以球面杂交方式对种群进行调整,加快算法的全局搜索速度。同时针对算法计算量大的缺点,设计对应的并行计算方法。实例结果表明,该算法耗时较少,收敛精度较高。
关键词:
克隆选择,
变异概率,
抗体亲和度,
球面杂交,
并行计算
Abstract: A new dynamic adaptive clone selection algorithm is proposed. Mutation probability of each antibody is dynamically calculated. 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. The effectiveness and the feasibility of the proposed algorithm are verified by examples. Example shows that the proposed algorithm has less time-consuming and higher convergence precision.
Key words:
clone selection,
mutation probability,
antibody affinity,
sphere crossover,
parallel computation
中图分类号:
李红婵, 朱颢东. 一种新的动态自适应克隆选择并行算法[J]. 计算机工程, 2011, 37(15): 161-163.
LI Gong-Chan, SHU Hao-Dong. Dynamic Adaptive Clone Selection Parallel Algorithm[J]. Computer Engineering, 2011, 37(15): 161-163.