Abstract: The Bacterial Foraging Optimization Algorithm(BFOA) has poor global search ability and is easily trapped into local opti- mum. In order to solve these problems, an adaptive hybrid BFOA fused with Artificial Bee Colony(ABC) algorithm is proposed. Firstly, Employed Bees Style Chemotaxis(EC) is proposed, which greatly enhances the algorithm’s capability of global searching. Then the original fixed step size chemotaxis is changed into an adaptive step size one, which improves the solution precision. On the basis of above, an evaluation method for diversity is put forward to switch two chemotaxis automatically. In order to overcome degradation of diversity caused by direct copy, a copy method based on t-distribution disturbance is proposed. A scout bees style migration based on opposition-based learning is put forward to avoid premature. Simulation experimental results show that the proposed algorithm has a better performance in terms of optimization ability, convergence speed and population diversity compared with ABC algorithm and BFOA.

Key words: Bacterial Foraging Optimization Algorithm(BFOA), Artificial Bee Colony(ABC) algorithm, adaptive step size, Employed Bees Style Chemotaxis(EC), t-distribution disturbance, opposition-based learning

摘要： 针对细菌觅食优化算法(BFOA)全局搜索能力差和易陷入局部最优的缺点，提出一种混合人工蜂群算法(ABC)的自适应细菌觅食优化算法。借鉴ABC的雇佣蜂行为，设计一种新的雇佣蜂式趋化方式，以提高算法的全局搜索能力。同时将原固定步长趋化改为自适应步长趋化，以提高算法的求解精度。引入种群多样性评价，依据评价结果完成2种趋化方式的自适应切换。为克服直接复制带来的多样性降低问题，提出基于t分布扰动的复制方式，同时设计基于对立学习的侦察蜂式迁移，以避免算法的早熟。仿真实验结果表明，与ABC和BFOA算法相比，该算法的寻优能力较强，在求解精度和收敛速度方面也具有较优的性能。

关键词: 细菌觅食优化算法, 人工蜂群算法, 自适应步长, 雇佣蜂式趋化, t分布扰动, 对立学习

CLC Number: 

• TP301