基于翻筋斗觅食策略的灰狼优化算法

灰狼优化算法(grey wolf optimization,GWO)存在收敛的不合理性等缺陷,目前对GWO算法的收敛性改进方式较少,除此之外,当GWO迭代至后期,所有灰狼个体都逼近α狼、β狼、δ狼,导致算法陷入局部最优。针对以上问题,提出了一种增强型的灰狼优化算法(disturbance and somersault foraging-grey wolf optimization,DSF-GWO)。首先引入一种扰动因子,平衡了算法的开采和勘探能力;其次引入翻筋斗觅食策略,在后期使其不陷入局部最优的同时也使得前期的群体多样性略有提升。对DSF-GWO算法的寻优性能进行验证,选取14个单/多峰目标函数进行实验,在相同的参数设置下,结果表明DSF-GWO算法在寻优性能上较GWO算法有明显优势。

Grey wolf optimization algorithm based on somersault foraging strategy

GWO algorithm has drawbacks such as the irrationality of convergence.At present,there are few ways to improve the convergence of GWO algorithm.In addition,in the GWO iteration to the later stage,all gray individual wolves approachαwolf,βwolf andδwolf,causing the algorithm to fall into a local optimum.In order to address the above problems,this paper proposed an enhanced GWO algorithm——DSF-GWO.This algorithm introduced a disturbance factor to balance the mining and exploration capabilities.Secondly,it introduced a somersault foraging strategy to prevent the group from falling into a local optimum,while also making the group’s diversity slightly improve at the early stage.This paper verified the optimization performance of DSF-GWO by 14 unimodal and multimodal objective functions.Under the same parameter settings,the experimental results show that DSF-GWO algorithm has obvious advantages in optimization performance over GWO algorithm.