直线二级倒立摆的一种稳定控制算法研究  

Research on a Stabilityle Control Algorithm of Pendulum Control Algorithm of theLinear Double Inverted Pendulum

作　　者：蒋凌云 徐炳吉[1] 张峰华 张钦琛 JIANG Ling-yun;XU Bing-ji;ZHANG Feng-hua;ZHANG Qin-chen(School of Information Engineering,China University of Geosciences(Beijing),Beijing 100083,China)

机构地区：[1]中国地质大学(北京)信息工程学院,北京100083

出　　处：《计算机仿真》2023年第2期314-320,共7页Computer Simulation

摘　　要：针对线性二次型调节器(LQR)的权重矩阵Q和R的选取没有一个固定解析类方法,LQR的控制性能好坏基本取决于人工经验对Q和R的选取,若系统状态变量变多,调参就变得过于繁琐,以及基本粒子群算法(PSO)存在收敛速度慢,容易陷入局部最优等缺点,从惯性权重非线性动态调整,学习因子非线性动态调整以及基于自然选择机理三个角度来改进基本粒子群算法。并将改进粒子群算法应用于LQR参数优化,对直线二级倒立摆进行稳定控制。仿真结果表明,改进粒子群算法优化后的LQR相比一些传统的方法优化的LQR有着更好的控制效果。There is no fixed analytical method for the selection of the weight matrix Q and R of linear quadratic regulator(LQR).The control performance of LQR basically depends on the selection of Q and R by artificial experience.If the system state variables become more,the parameter adjustment will become too cumbersome,and the basic particle swarm optimization(PSO)has the disadvantages of slow convergence speed and easy to fall into local optimum.In the paper,the basic particle swarm optimization algorithm was improved from three aspects:nonlinear dynamic adjustment of inertia weight,nonlinear dynamic adjustment of learning factor and natural selection mechanism.For linear quadratic regulator(LQR)selection of weighting matrix Q and R does not have a fixed analytical method,LQR control performance is good or bad depends on artificial basic experience,if there are more system state variables,adjustment becomes too complicated,as well as the basic particle swarm optimization algorithm(PSO)is slow convergence speed,easily falling into the Local optimal solution,In this paper,the basic particle swarm optimization algorithm is improved from three aspects:inertia weight nonlinear dynamic adjustment,learning factor nonlinear dynamic adjustment and natural selection mechanism.The improved particle swarm optimization algorithm was applied to the optimization of LQR parameters to control the stability stabilize pendulum ofthe the linear double inverted pendulum.The simulation results show that the optimized LQR by the improved particle swarm optimization algorithm has better control effect than the optimized LQR by some traditional methods.

关 键 词：改进粒子群算法 线性二次型调节器 倒立摆 遗传算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]