一种改进的人工蜂群算法

A Modified Artificial Bee Colony Algorithm

人工蜂群算法是近年来提出的一种受生物行为启发的优化算法,该算法主要通过模拟蜜蜂的觅食来实现问题的求解。作为一种全局优化算法,人工蜂群算法有着较好的探寻能力,但其探索能力相对较弱。针对人工蜂群算法收敛速度缓慢的问题,提出基于scout蜂交叉觅食的改进人工蜂群算法。该算法通过交叉策略来指导scout蜂的觅食行为,避免了随机觅食带来的算法收敛速度缓慢的问题,提高算法的收敛速度。通过五个基准测试函数进行对比实验,结果表明新算法无论是在收敛速度、解的质量方面都优于标准人工蜂群算法,是一种有效的优化算法。

Artificial bee colony （ABC） algorithm invented recently is a biological-inspired optimization algorithm, which simulates the foraging behaviors of honey bee swarm. As one of the global optimization algorithms, ABC is good at exploration but poor at exploitation. A modified artificial bee colony （MABC） algorithm based on crossover strategy of scout is proposed for slow convergence of basic ABC. MABC avoids the problem of slow convergence came with ran-dom foraging and increases the convergence speed by means of crossover strategy which guides the scout foraging be-havior. The proposed algorithm is tested on five different scale problems and compared with basic ABC. The compari-son results show that MABC is an effective algorithm, and is better than basic ABC in not only the convergence speed but also the solution quality.