改进灰狼优化算法分数阶PID控制器参数整定  

Improved Gray Wolf Optimization Algorithm for Fractional Order PID Controller Parameter Tuning

作　　者：贺斌 李岚卿 程江勇超 周希正 He Bin;Li Lanqing;Cheng Jiangyongchao;Zhou Xizheng(College of Civil Engineering,Jiangxi Science and Technology Normal University,Nanchang 330038,China)

机构地区：[1]江西科技师范大学土木工程学院,南昌330038

出　　处：《科技通报》2024年第5期39-45,共7页Bulletin of Science and Technology

基　　金：江西省教育厅科学技术研究项目(GJJ2201317)。

摘　　要：针对分数阶PID(proportion integration differentiation)控制器参数整定难的问题,本文提出改进的灰狼优化算法(improved grey wolf optimizer,IGWO)对分数阶PID控制器进行参数整定的方法。IGWO算法采用logistic映射来初始化种群位置,提高种群的多样性,采用非线性收敛因子,增强全局搜索能力和局部开发能力,采用动态权重策略,根据适应度值调整α、β、δ狼的权重。为验证IGWO算法的有效性选取4个基准测试函数进行寻优和对2个经典被控系统进行控制器设计,并与传统灰狼算法、粒子群算法、遗传算法和粒子群结合灰狼算法进行对比分析,结果显示,在寻优测试中IGWO算法在收敛速度和解的精度上更有优势,采用IGWO算法设计得到的控制器的控制性能更好。Aiming at the difficulty of parameter tuning of fractional order PID controller,this paper presents an improved Grey Wolf optimization algorithm(IGWO)for parameter tuning of fractional order PID controller.The IGWO algorithm uses logistic mapping to initialize the population position and improve the diversity of the population.It uses a nonlinear convergence factor to enhance the global search ability and local exploitation ability.It adopts a dynamic weighting strategy and adjusts the weights ofα,βandδWolf according to the fitness value.In order to verify the effectiveness of IGWO algorithm,four benchmark test functions are selected for optimization and controller design of two classical controlled systems,and compared with traditional Grey Wolf algorithm,particle swarm algorithm,genetic algorithm and particle swarm combined Grey Wolf algorithm,the results show that,In the optimization test,IGWO algorithm has more advantages in convergence speed and precision,and the controller designed by IGWO algorithm has better control performance.

关 键 词：分数阶PID 优化算法 参数整定 鲁棒性 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]