引入改进迭代局部搜索的灰狼算法及应用

Improved iterative local search grey wolf algorithm and its application

针对标准灰狼算法(GWO)收敛速度慢,易陷入局部最优等缺点,提出一种引入改进迭代局部搜索的灰狼算法(IGWO)。首先,通过佳点集策略增强初始种群的均匀性与多样性;其次,采用双收敛因子,收敛因子基于种群位置非线性自适应更新,在种群寻优全期平衡全局勘探与局部开发能力;再次,在种群位置更新公式引入欧氏动态权重与莱维飞行策略,提升寻优精度,并帮助种群跳出局部最优值;最后,引入改进迭代局部搜索,使算法的搜索能力更加灵活,帮助算法加速收敛。通过10个基准测试函数的仿真分析及种群寻优平衡性对比,证明了IGWO具有更优的寻优精度、稳定性及收敛速度,随后将IGWO应用于工程优化问题中,相比GWO、GJO、WOA、HSSAHHO、SCHOA、NCPGWO、DSFGWO 7种算法,适应度分别优化了3.25%、27.2%、28.9%、3.15%、3.04%、0.23%、0.07%,证实了在工程应用中的可行性和有效性。

Aiming at the disadvantages of slow convergence of standard grey wolf optimizer(GWO)and easy to fall into local optimality,an improved grey wolf optimizer(IGWO)with improved iterative local search is proposed.First,the uniformity and diversity of the initial population were enhanced by the strategy of the best point set.Secondly,the dual convergence factor is used,which is nonlinear and adaptive updating based on population location,to balance the global exploration and local development ability in the whole period of population optimization.Thirdly,European dynamic weight and Levy flight strategy were introduced into the population position updating formula to improve the optimization accuracy and help the population jump out of the local optimal value.Finally,the improved iterative local search is introduced to make the search ability of the algorithm more flexible and help the algorithm accelerate convergence.Through the simulation analysis of 10 benchmark test functions and the comparison of population optimization balance,it is proved that IGWO has better optimization accuracy,stability and convergence speed.Then IGWO was applied to the engineering optimization problem.Compared with GWO,GJO,WOA,HSSAHHO,SCHOA,NCPGWO,DSFGWO 7 algorithms,the fitness was optimized by 3.25%,27.2%,28.9%,3.15%,3.04%,2.33%,0.07%,respectively.The feasibility and effectiveness in engineering application are proved.