CAJ | 학술논문

为了充分利用种群的全局统计信息，将分布估计算法引入二进制人工蜂群算法，提出一种基于分布估计的二进制人工蜂群算法。该算法利用分布估计算法获得的全局统计信息引导候选解的产生，提高了全局探索能力；采用直接针对离散域的多维邻域更新策略，加快了收敛速度，降低了计算复杂度。仿真结果表明，与传统二进制人工蜂群算法相比，改进算法在优化精度、收敛速度和鲁棒性方面均有明显改善。
위료충분이용충군적전국통계신식，장분포고계산법인입이진제인공봉군산법，제출일충기우분포고계적이진제인공봉군산법。해산법이용분포고계산법획득적전국통계신식인도후선해적산생，제고료전국탐색능력；채용직접침대리산역적다유린역경신책략，가쾌료수렴속도，강저료계산복잡도。방진결과표명，여전통이진제인공봉군산법상비，개진산법재우화정도、수렴속도화로봉성방면균유명현개선。
To make full use of the population’s global statistics information,this paper introduced estimation of distribution algorithm into binary artificial bee colony algorithm,and proposed a binary artificial bee colony algorithm based on estimation of distribution algorithm.It guided the candidate solution by the global statistics information from the estimation of distribution algorithm,which could enhance the global exploration ability.It designed the multidimensional update strategy in the discrete domain,which quickens convergence speed and reduces the computational complexity.The simulation results show that,com-pared with the traditional binary artificial bee colony algorithm,the performances of the improved algorithm are better in the optimization accuracy,convergence speed and robustness.