基于改进灰狼算法的LQR优化控制方法研究  

Research on LQR Optimal Control Method Based on Improved Gray Wolf Algorithm

作　　者：宋涛涛 李艳萍[1] 李洪港 SONG Tao-tao;LI Yan-ping;LI Hong-gang(School of Information and Electrical Engineering,Shandong Jianzhu University,Jinan Shandong 250101,China)

机构地区：[1]山东建筑大学信息与电气工程学院,山东济南250101

出　　处：《计算机仿真》2024年第2期339-343,372,共6页Computer Simulation

基　　金：国家自然科学基金项目(62133008)。

摘　　要：针对二级倒立摆在使用LQR(线性二次调节器)进行优化控制过程中,由经验选取的加权矩阵Q和R参数存在着较大的随机性和不稳定性问题,提出了改进灰狼算法优化控制器加权矩阵Q和R的方法。为灰狼算法设计了基于二次余弦规律的自适应收敛因子a和增强α狼适应度值fα的比例权重方法。增强了算法迭代前期的全局搜索能力和后期的收敛速度,通过MATLAB/Simulink仿真,并与传统灰狼算法相比较,得出改进算法能够有效降低倒立摆回归平衡状态时的超调量,更快达到稳定状态,使控制效果更加理想。Aiming at the problem of significant randomness and instability in the weighting matrix Q and R parameters selected by experience during the optimization control process using LQR(linear quadratic regulator)for a two-stage inverted pendulum,an improved grey wolf algorithm is proposed to optimize the controller weighting matrix Q and R.An adaptive convergence factor a based on quadratic cosine and a proportionalαweight method to enhance wolf fitness value fαwere designed for Gray Wolf algorithm.The global search ability in the early stage of the algorithm iteration and the convergence speed in the later stage were enhanced,and the optimal weighting matrix Q and R parameters were obtained.By MATLAB/Simulink simulation,the improved algorithm can effectively reduce the overshoot when the inverted pendulum returns to the balance state,reach the stable state faster and achieve better control effect compared with the traditional gray wolf algorithm.

关 键 词：灰狼算法 线性二次调节器 二级倒立摆 收敛因子 适应度值 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]