改进人工蜂群算法

针对人工蜂群算法存在的收敛速度慢、易陷入局部最优的缺点,利用自由搜索算法中的信息素、灵敏度模型代替传统的轮盘赌选择模型,并引入OBL策略产生新蜜源取代每次迭代的最差蜜源,提出了一种改进的人工蜂群算法,并结合NIT技术建立一种新的多峰优化方法.对9个标准测试函数仿真表明本文提出的改进算法不仅大大提高了最优解的精度而且缩短了运行时间,改进性能明显优于现有人工蜂群算法.实例测试表明该方法能够有效、精确地搜索各个峰值点.

A modified artificial bee colony algorithm and its application

Considering the shortcomings of the artificial bee colony algorithm due to slow convergence and ease of falling into the local optimum,the pheromone and the sensitivity model in a free search algorithm was used to replace the traditional roulette wheel selection model.Furthermore,OBL was introduced in this paper to present an improved artificial bee colony algorithm,replacing the worst bee colony with a new one generated in each iterative procedure.Then a new method for multi-modal function optimization based on this algorithm was designed.The simulation results of nine standard test functions show that in the case of less running time the new algorithm greatly improves the accuracy of the optimal solution,and it is much better than the existing artificial bee colony algorithms.The simulation of five test functions shows that this method can accurately search the various peaks.