基于最优个体指导单纯形法改进的人工蜂群算法及应用

针对基本人工蜂群算法在求解复杂优化问题时,存在收敛精度低、收敛速度慢的缺点,提出一种基于最优个体指导单纯形法改进的人工蜂群算法。算法引入基于当前最优个体作为指导的单纯形法进行邻域搜索,以增强局部探索能力。同时采取保优策略,以加快收敛速度。通过6个标准测试优化问题的仿真实验表明,该算法较基本人工蜂群算法具有更高的求解精度和更快的收敛速度。将算法用于分数阶登革病毒传播模型的参数优化,所得的参数对应的模型输出与实际数据拟合情况较好。

IMPROVED ARTIFICIAL BEE COLONY ALGORITHM WITH SIMPLEX METHOD BASED ON OPTIMAL SOLUTION AND ITS APPLICATION

The basic artificial bee colony (ABC) algorithm has disadvantage of low convergence precision and slow convergence speed in solving complex optimization problems. To solve this problem, we proposed an improved artificial bee colony algorithm with simplex method based on optimal solution. Simplex method, which was based on the current optimal individual as guidance, was introduced to search the neighborhood to enhance the ability of local exploration. The optimization strategy was adopted to accelerate the convergence speed. The simulation results on 6 standard test optimization problems show that compared with the basic ABC algorithm, the improved algorithm has higher solution accuracy and faster convergence speed. The improved algorithm is applied to optimize the parameters of fractional dengue virus propagation model. The output of the model corresponding to the parameters is in better fit with the actual data.