基于单纯形的改进精英人工蜂群算法
An Improved Multi-elitist Artificial Bee Colony Algorithm Based on Nelder-Mead Simplex Method

针对人工蜂群算法收敛速度慢, 求解精度不高, 易陷入局部最优等问题, 基于受粒子群启发的多精英人工蜂群优化算法, 引入了蜂群中的精英个体和全局最优个体增强开发全局最优解的能力。文章中，在雇佣蜂阶段借助精英个体引导蜜源搜索，并利用蜂群中蜜源的质量排序重新构造蜜源的选择概率公式；在跟随蜂阶段，选择种群最优蜜源引领蜂群，加强算法对全局最好解的局部开采能力，同时将随机选择邻居蜜源变为最优定向选择。最后利用单纯形算法对精英解集进行再次更新， 进一步平衡蜂群的全局搜索和局部寻优能力。数值实验表明改进的新算法的寻优精度和收敛速度均有明显提高。
There were some problems in the Artificial Bee Colony (ABC) algorithm, such as the slow convergence speed, low solution precision and easy to fall in local optimum. Inspired by particle swarm optimization algorithm, multi-elitist artificial bee colony algorithm for real-parameter optimization use of global best solution and an elitist randomly selected from the elitist set were adopted to enhance the exploitation of the global best solution. In this paper, we the elitist to guide the nectar search during the employed bee process was introduced. And the selection probability formula of food source was reconstructed by using the quality of food source. In the onlooker bee stage, the best food source was selected to guide the swarm to enhance the exploitation of the global best solution, and the and the neighbor food source was selected to be the optimally directional choice. As the same time, a simplex method was used on elitist solution set to balance the exploration and exploitation ability of the algorithm. The numerical experiment results showed that the proposed algorithm had higher searching precision and faster convergence speed.