摘要
针对传统的灰狼算法(GWO)易陷入局部最优、后期收敛速度慢等问题,提出一种非线性控制参数组合调整策略。对3种不同的非线性参数控制策略的调节因子a进行仿真,并分析影响搜索参数A的因素;对5组不同的调节参数值进行基准函数的测试仿真,选择权重系数的非线性控制参数组合策略的最佳参数值。仿真结果表明,所提出的非线性控制参数组合调整策略优于GWO 2、GWO 3,以及文献[10]提出的改进灰狼优化算法,为组合策略在智能算法中的运用提供了验证。
In order to improve the traditional gray wolf optimization(GWO)algorithm,which is easy to fall into local optimum and slow at convergence speed,in this paper,a nonlinear control parameter combination adjustment strategy is proposed.The adjustment factor a of three different nonlinear parameter control strategies was simulated and the factors affecting the search parameter A were analyzed.Five groups were adjusted differently,the parameter value was used for the test simulation of the reference function,and the optimal parameter value of the nonlinear control parameter combination strategy of the weight coefficient was selected.The simulation results show that the proposed nonlinear control parameter combination adjustment strategy is better than the GWO 2,GWO 3 and the improved gray wolf optimization algorithm proposed in reference[10],which provides verification for the application of combination strategy in intelligent algorithms.
作者
张孟健
龙道银
杨小柳
王霄
杨靖
Zhang Mengjian;Long Daoyin;Yang Xiaoliu;Wang Xiao;Yang Jing(The Electrical Engineering College,Guizhou University,Guiyang 550025,Guizhou,China;Power China Guizhou Engineering Co.,Ltd.,Guiyang 550001,Guizhou,China)
出处
《计算机应用与软件》
北大核心
2021年第5期250-255,322,共7页
Computer Applications and Software
基金
国家自然科学基金项目(61861007,61640014)
贵州省工业攻关项目(黔科合支撑[2019]2152,黔科合支撑[2016]2302)
物联网技术案例库(KCALK201708)
自动化专业卓越工程师计划项目(ZYS2015004)
贵州省联合基金项目(黔科合LH字[2017]7228)
贵州省研究生创新基金项目(YJSCXJH[2019]005)。
关键词
灰狼优化算法
非线性控制参数策略
权值系数
搜索参数
控制参数
Grey wolf optimizer
Nonlinear control parameter strategy
Weight coefficient
Search parameter
Control parameter
