基于变异策略的自适应七星瓢虫优化算法

Adaptive seven-spot ladybird optimization based on mutation strategy

针对七星瓢虫优化算法易陷入局部最优、求解精度不高的缺陷,提出基于变异策略的自适应七星瓢虫优化算法。为提高算法求解质量,在每次迭代搜索时利用柯西变异策略增加解的多样性,引入竞争淘汰机制淘汰适应度值较差的个体;同时,为了提高算法的收敛性能,在算法搜索后期利用混沌变异策略对种群中最优和较优个体进行混沌变异操作,并对学习因子进行自适应更新调整。利用标准测试函数进行实验仿真,结果表明改进算法不仅提高了求解精度,同时有效避免了局部收敛问题。

Considering the problems of falling into the local optimum and low solving precisions of seven-spot ladybird optimization（SLO）, this paper proposed an adaptive seven-spot ladybird optimization algorithm based on mutation strategy. In order to improve the quality of solutions, it applied Cauthy mutation to increase the diversity of population and competition mechanism to eliminate bad individual according to its fitness in every iteration. To improve convergence performance of the SLO, this paper varied the best and the second-best based on chaos mutation strategy in the late iterations. Meanwhile, it adjusted the acceleration coefficient adaptively. Simulation experimental results based on a set of widely used benchmark functions show that the improved algorithm can yield solutions with higher precision, while effectively avoiding local convergence problems.