Abstract: A comprehensive test and study of artificial bee colony algorithm’s performance is done. A series of experiments including effect of dimension and colony size, effect of scout bees and effect of initial region scaling are taken and analyzed. Meanwhile, inspired by genetic algorithm, ABC algorithm is applied with typical selection mechanisms and the performance with different selection mechanisms is compared. The experimental results show that ABC algorithm can obtain the global best optimum result in the condition of setting colony size be 40, dimension be 10 or 30, initial region scaling be symmetric distributed and the selection mechanism be deterministic sampling or remainder stochastic sampling with replacement instead of roulette wheel selection used in ABC algorithm.

Key words: artificial bee colony algorithm, function optimization, selection mechanism, parameter optimization

摘要： 对蜂群算法的性能进行全面的测试和研究，实验分析了维数和粒子数对算法的影响，侦察蜂的活动对算法的影响以及初始解的位置对算法的影响。同时受遗传算法的启发，将典型的选择机制应用到蜂群算法并对其进行改进，并比较不同选择机制下蜂群算法的性能。实验结果表明，在粒子数为40，维数为10或者30，均匀分布初始解的位置，采用确定式选择法和无放回余数选择法代替蜂群算法中轮盘赌的选择方法的条件下，蜂群算法得到整体最好的优化结果。

关键词: 蜂群算法, 函数优化, 选择机制, 参数优化