一类改进的人工蜂群算法

A Kind of Improved Artificial Bee Colony Algorithm

对2005年Karaboga提出的模仿蜜蜂觅食行为的人工蜂群算法进行了研究,将粒子群算法中的惯性权重引入到人工蜂群算法中,提出了带惯性权重的改进的人工蜂群算法(Improved artificial bee colony algorithm with inertia weight,ABCIW)的方法.将ABCIW算法应用于求解基准函数的最小值问题,进而应用于优化BP神经网络的参数,对中国手足口病发病人数进行预测.与基本人工蜂群算法、快速人工蜂群算法和带记忆的人工蜂群算法相比较,ABCIW算法更适合解决函数的优化问题.对中国手足口病发病人数的预测说明了ABCIW算法具有较好的预测结果和较高的稳定性.

Artificial bee colony algorithm presented by Karaboga in 2005 is researched, which imitates the foraging behavior of honeybees. An inertia weight in particle swarm optimization is introduced into artificial bee colony algorithm, and an improved artificial bee colony algorithm with inertia weight（AB- CIW） is proposed. Apply ABCIW algorithm to solve the minimum problems of benchmark functions and further to optimize the parameters of BP neural network for predicting the onset number of hand-foot- mouth disease in China. By comparison with basic artificial bee colony algorithm, quick artificial bee col- ony algorithm and artificial bee colony algorithm with memory, ABCIW algorithm is more suitable for solving the optimization problems of functions. The prediction results of the onset number of hand-foot- mouth disease in China show that ABCIW algorithm has good prediction results and higher stability.