论文部分内容阅读
针对传统免疫克隆选择算法收敛速度较慢的问题,结合克隆概率和免疫概率的自适应变换、群体灾变算法以及有无记忆库思想,提出了无记忆库的自适应免疫克隆选择算法与有记忆库的自适应免疫克隆选择算法,并将其应用于TSP问题.群体灾变算法的应用便于使算法尽快摆脱迟钝状态,并使算法能够保持抗体多样性.自适应方法的应用使得算法在进化初期有较强的全局搜索能力和较弱的局部搜索能力,随着进化的进行,全局搜索能力逐渐减弱,局部搜索能力逐渐增强,便于找到全局最优点.仿真实验结果表明,与传统的免疫克隆算法相比,该算法有效克服了早熟问题,保持了抗体的多样性,而且收敛速度较快.
Aiming at the problem of slow convergence speed of traditional immune clone selection algorithm, this paper proposes adaptive immune clone selection algorithm without memory bank and memory bank with memory transform based on the adaptive transformation of clone probability and immune probability, group catastrophe algorithm and the idea of memory with or without memory, Adaptive immune clonal selection algorithm and its application to TSP problems.The application of group catastrophe algorithm to facilitate the algorithm to get rid of insensitive state as soon as possible and enable the algorithm to maintain the antibody diversity.Application of adaptive method makes the algorithm in the early evolution of more Strong global search ability and weak local search ability, with the progress of evolution, the global search ability gradually weakened and the local search ability gradually increased, which is convenient to find the global optimum point.The simulation results show that compared with the traditional immune clone algorithm , The algorithm effectively overcomes the problem of precocious puberty, maintains the diversity of antibodies, and has a faster convergence rate.