改进的人工蜂群算法性能

Performance of an improved artificial bee colony algorithm

为克服人工蜂群算法容易陷入局部最优解的缺点,提出一种新的改进型人工蜂群算法。首先,利用数学中的外推技巧定义了新的位置更新公式,由此构造出一种具有引导趋势的蜂群算法;其次,为了克服算法在进化后期位置相似度高、更新速度慢的缺陷,将微调机制引入算法中,讨论摄动因子范围,由此提高算法在可行区域内的局部搜索能力。最后通过3个基准函数仿真测试,结果表明:与常规算法相较,改进后在搜索性能和精度方面均有明显提高。

An improved algorithm based on Artificial Bee Colony（ABC） algorithm was proposed to solve the problem that traditional ABC algorithm is inclined to fall into local optima.In the first stage,the improved ABC algorithm was derived from the skills of extrapolation in mathematics to update the new location of ABC.In the second stage,in order to overcome the deficiency of high position similarity in later stage of evolution and slow renewal rate and enhance the ability of local search in feasible region,a fine-tuning mechanism was introduced to ABC.Simultaneously,the effect of convergence subjected to different perturbation factors was discussed.Finally,the simulation results in three benchmark functions show that the proposed algorithm has better performance than traditional algorithm in search ability and accuracy.