CAJ | 학술논문

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。
인공봉군산법시근년래제출적일충수생물행위계발적우화산법，해산법주요통과모의밀봉적멱식래실현문제적구해。작위일충전국우화산법，인공봉군산법유착교호적탐심능력，단기탐색능력상대교약。침대인공봉군산법수렴속도완만적문제，제출기우scout봉교차멱식적개진인공봉군산법。해산법통과교차책략래지도scout봉적멱식행위，피면료수궤멱식대래적산법수렴속도완만적문제，제고산법적수렴속도。통과오개기준측시함수진행대비실험，결과표명신산법무론시재수렴속도、해적질량방면도우우표준인공봉군산법，시일충유효적우화산법。
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.