Abstract: In order to get the best solutions of the multi-model function, an immune clonal algorithm with self-adaptive population size is proposed. The self-adaptive population size changes with the evolutionary process are achieved to balance the impact of population size on the efficiency of the algorithm. In addition, in terms of multi-modal function optimization characteristics, Larmark learning is used as a local search strategy to enhance the search ability for the optimal solution. The experimental results show that the algorithm has better performances.

Key words: clone optimization, multi-model function, population size, local search

摘要： 为了尽可能求得多峰函数的最优解，提出了一种种群规模自适应调整的克隆算法。实现了种群规模根据进化过程自适应的变化，平衡了种群规模对算法效率的影响。此外，结合多峰函数优化的特点，为了增强算法搜索最优解的能力，采用Larmack学习策略作为局部搜索机制。实验结果表明，该算法求解效果较好。

关键词: 克隆优化, 多峰函数, 种群规模, 局部搜索