蜂群算法研究综述

Overview of research on bee colony algorithms

蜂群算法是一种模仿蜜蜂繁殖、采蜜等行为的新兴的群智能优化技术,近几年备受研究者关注。初步探讨了蜂群算法的理论基础,详细论述了基于蜜蜂繁殖行为和采蜜行为的两类蜂群算法的生物学机理及其最常见算法的应用研究情况,并分析比较了遗传算法、蚁群算法、粒子群算法和蜂群算法的优缺点、适用范围及性能。最后,总结了现有蜂群算法存在的问题,并指出其未来的研究方向。

Bee colony algorithms are new swarm intelligence techniques inspired by the intelligent behaviors of real honey bees such as the reproductive behavior and the foraging behavior.More recently,researchers have become very interested in it and its related research.Therefore,this paper preliminary studied the theoretical basis of bee colony algorithms.According to the different bee behaviors,bee colony algorithms were mainly classified into two types,namely the reproductive behavior and the foraging behavior.Then discussed and illustrated the biological mechanism and the most popular algorithm of each type in detail,respectively.Moreover,analyzed and compared genetic algorithm,ant colony optimization,particle swarm optimization and bee colony algorithms in terms of advantages and disadvantages,application fields and performances.Finally,summarized the existing problems in current research on the bee colony algorithms and suggested some future research directions to address the problems.