基于优势肽和免疫记忆的混合蚁群算法

Hybrid Ant Colony Algorithm Based on the Superior Peptide and Immune Memory

为了克服基本蚁群算法求解速度慢、易于出现早熟和停滞现象的缺陷,借鉴免疫算法中的免疫记忆和优势肽选择继承的思想,提出了基于优势肽和免疫记忆的混合蚁群算法(SPIM-ACA)。该算法在原有蚁群模型基础上增加内部记忆库,将记忆库中的解对应免疫抗体,将问题对应为抗原,运用免疫算子和优势肽选择算法进行新解的构造和记忆库的更新。将该算法从解的质量和多样性方面与传统蚁群算法、免疫算法及已有的改进算法进行了比较,结果表明:本文提出的算法不但明显提高了两个传统算法的性能,而且为解决其他组合优化问题提供了一个新的思路。

In order to overcome the shortcomings of the basic ant colony algorithm, such as slow convergence speed, precocity and stagnation, this paper proposes a hybrid ant colony algorithm based on the superior peptide and immune memory （SPIM-ACA） by using the mechanism of immune memory and the superior peptide selection and implantation. SPIM-ACA adds the interior and exterior memory library to the ant colony mode, takes the solutions in the memory library as antibodies and the problem as antigen, and undergoes the construction of solution and the updating of pheromone concentration by using the above mechanism. The results of experiment for solving TSP （traveling salesman problem） indicate that the proposed algorithm is superior to some standard algorithms, such as the hasic ant colony algorithm, the immune algorithm and so on, in the respects of the quality and the diversity of the performance.