基于人工免疫的多峰函数优化算法研究

摘要

研究函数优化问题,提高寻优效率.针对寻找函数最优解,当函数是具有多个峰值的函数时,传统的基于蚁群的函数优化算法易陷人局部极小,不能找到真正的全局最优解的问题.提出一种基于人工免疫的改进蚁群函数优化方法,在传统的蚁群算法基础上引入人工免疫的思想,加入抗体浓度作为蚂蚁选择下-条路径时的条件,由空间信息素、抗体浓度以及抗体适应度值等全局信息来决定抗体被选择的概率,避免了只依赖单一的信息素寻优而陷入局部极小的问题.实验证明,这种方法不仅能解决多峰函数寻优易陷入局部极小的问题,而且具有很高的迭代寻优效率,取得了满意的结果.%Research function optimization problems and improve the optimal efficiency. This paper presented an improved ant colony optimization method based on artificial immune function. The method joined antibody concentra-tion as the conditions under which the ants chose next path. Then using the global information such as space informa-tion, antibody concentration and antibody fitness value, the selected probability of antibody was determined, which a-voided to fall into the local minimum caused by only rely on a single pheromones optimization. Experiments show that the method can not only solve the problem that multi-modal function optimization easily get into local minimum, but also have high iterative optimization efficiency, and obtaine satisfactory results.