基于单神经元PSD算法控制串级倒立摆系统  

Control of cascade inverted pendulum system based on single neuron PSD algorithm

作　　者：陈强 姚风麒 王国庆 金映波 李羽辰 Chen Qiang;Yao Fengqi;Wang Guoqing;Jin Yingbo;Li Yuchen(Anhui University of Technology,Ma’anshan Anhui,243000)

机构地区：[1]安徽工业大学,安徽马鞍山243000

出　　处：《电子测试》2022年第5期74-76,73,共4页Electronic Test

摘　　要：倒立摆是具有非线性,多变量,强耦合性的综合系统模型,传统的控制方法在解决其复杂多变的控制问题上的能力是有限的。本文在基于串级PID倒立摆的基础上,利用单神经元的自学习和自适应能力设计其外环控制器。在单神经元PID控制中,增益K的选取与实际系统的控制对象特性息息相关,且K值的给定对系统有很大影响。针对单神经元中固定增益K值可能无法调节出较为满意的控制效果,提出基于PSD控制算法可自动调整增益K值的单神经元自适应PID控制策略。在此基础上进行仿真实验,结果表明较之于串级PID控制倒立摆有更好的控制效果和稳定性。Inverted pendulum is a comprehensive system model with nonlinear,multivariable and strong coupling.The ability of traditional control methods to solve its complex and changeable control problems is limited.Based on the cascade PID inverted pendulum,the outer loop controller is designed by using the self-learning and adaptive ability of single neuron.In single neuron PID control,the selection of gain K is closely related to the characteristics of the control object of the actual system,and the given value of K has a great influence on the system.In view of the fact that the fixed gain K value in a single neuron may not be able to adjust a satisfactory control effect,a single neuron adaptive PID control strategy based on PSD control algorithm is proposed.The simulation results show that the inverted pendulum has better control effect and stability than cascade PID control.

关 键 词：串级倒立摆 单神经元 PSD控制算法 

分 类 号：TPH73[自动化与计算机技术]