Abstract: An immune good-point set genetic algorithm is put forward based on the combination of immune system and good-point set theory.The method of creating good point is presented to modify crossover operator and initial population setting.A definition based on weighted Euclidean distance is proposed to calculate antibody similarity,concentration and fitness.Immune system is introduced to keep the diversities of population and make a good speed guiding direction that aims at the family whose ancestors have schemata with high fitness.The simulation results show that this algorithm has superiority in speed,accuracy and overcoming premature.

Key words: immune system, weighted Euclidean distance, good-point set, characteristic individual, diversity, uniform design

摘要： 结合免疫机制和数论中的佳点集理论,给出了一种免疫佳点集遗传算法。该算法把数论中佳点集理论运用于遗传算法交叉操作和种群初始化的改进,提出带权欧氏距离计算抗体的相似度、浓度和适应度,引入免疫机制使群体保持多样性和快速导向高适应度模式。实验结果验证了该算法可以有效地避免早熟,改善算法的全局收敛性,提高搜索效率。

关键词: 免疫机制, 带权欧氏距离, 佳点集, 特征个体, 多样性, 均匀设计