融入等温过程的改进灰狼优化算法

Improved gray wolf optimization algorithm integrating isothermal process

灰狼优化算法是一种新颖的群智能优化算法,针对该算法存在的平衡全局探索和局部开发效率低、易陷入局部极值的问题,提出融入等温过程的改进灰狼优化算法IGWOSA。为了平衡算法开发与探索的能力,IGWOSA在灰狼位置更新操作后,融入等温过程。根据metropolis准则对更新的新位置进行取舍,从而增添了算法跳出局部极值的能力。同时,对α、β、δ灰狼赋予高斯扰动变异操作,进一步提升搜索效率。实验结果表明,对于13个基准函数,改进策略能有效提升算法性能;高斯扰动对算法性能有显著提升效果;IGWOSA与最先进的同类算法EOGWO、EGWO、CGWO相比,在搜索效率和性能方面优势明显。其中,IGWOSA尤其擅长处理单峰函数,更是以数量级的优势优于对比算法,但是,在处理多峰函数时,EGWO以微弱的优势优于IGWOSA。

Grey wolf optimization(GWO)algorithm is a new type of swarm intelligence optimization algorithm.In view of the fact that the algorithm is low efficiency and easy falling into local extremum in the process of balancing global search and local development,a hybrid gray wolf optimization algorithm named IGWOSA which integrates isothermal process is proposed.In order to balance the ability of algorithm'development and search,IGWOSA is integrated with the isothermal process after the location update operation.According to the criterion of metropolis,the updated location is screened to increase the algorithm's ability of jumping out of local extremum.In order to improve the search efficiency,Gaussian perturbation mutation operation is applied to each generation of α,β and δ individuals by GWO algorithm.The experiments on 13 benchmark functions show that the improved strategy can effectively improve the performance of the algorithm,and the Gaussian perturbation can significantly improve the performance of the algorithm.In addition,IGWOSA has significant advantages in search efficiency and performance in comparison with the similar algorithms of EOGWO,EGWO and CGWO.Among them,IGWOSA is especially good at dealing with unimodal function,and it is superior to the contrast algorithms by an order of magnitude.However,EGWO is slightly superior to IGWOSA in dealing with multimodal function.