首页|基于RBF神经网络整定PID的电液比例系统位置控制研究

基于RBF神经网络整定PID的电液比例系统位置控制研究

扫码查看
针对凿岩机械臂的电液比例系统位置控制精度问题,提出了一种基于径向基函数(RBF)神经网络整定PID的电液比例系统位置控制方法.首先,在AMESim中搭建了阀控非对称液压缸的电液比例系统简化模型,设置了各个模块的参数;然后,利用MATLAB/Simulink搭建了系统闭环控制模型,通过不断更新RBF网络模型并修正PID参数,实现了基于RBF神经网络整定PID的电液比例系统位置控制目的;结合AMESim搭建的电液比例系统模型和Simulink下搭建的控制器进行了联合仿真;最后,基于凿岩台车机械臂实验平台,进行了电液比例系统位置控制实验.仿真结果表明:在受到外部干扰的情况下,RBF神经网络整定PID控制系统能够在0.3s内控制活塞杆重新运行至目标位置,平均响应时间为 1.5 s,位置精度误差不超过 5 mm.实验结果表明:与常规PID控制方法相比,RBF神经网络整定PID控制活塞杆位置精度误差降低了75%,位置精度误差在工程实际要求的 10 mm范围以内,因此,RBF神经网络整定PID算法可以有效提高电液比例系统的位置控制精度,满足凿岩机械臂实际工作中对电液比例系统位置精度的控制要求.
Position control of electro-hydraulic proportional system based on RBF neural network tuning PID
Aiming at the position control accuracy of electro-hydraulic proportional system of rock drilling manipulator,a method of position control of the electro-hydraulic proportion system based on radical basis function(RBF)neural network tuning PID was proposed.Firstly,a simplified model of the electro-hydraulic proportion system for a valve-controlled non-symmetric hydraulic cylinder was built in AMESim,and parameters for each module were set.Then,a closed-loop control model of the system was constructed using MATLAB/Simulink.The RBF network model was continuously updated and the PID parameters were corrected,position control of the electro-hydraulic proportion system based on RBF neural network tuning PID was achieved.The built model of the electro-hydraulic proportion system in AMESim and the controller constructed in Simulink were combined for joint simulation.Finally,based on the mechanical arm experimental platform of a rock drilling rig,an electro-hydraulic proportional system position control experiment was conducted.The simulation results indicate that,under external disturbances,the RBF neural network tuning PID control system can control the piston rod to return to the target position within 0.3 s,with an average response time of 1.5 s,and the position accuracy error does not exceed 5 mm.The experimental results show that compared with conventional PID control method,the RBF neural network tuning PID control piston rod position accuracy error is reduced by 75%,and the position accuracy error is within the 10 mm range required by engineering practice.Therefore,the RBF neural network tuning PID algorithm can effectively improve the position control accuracy of the electro-hydraulic proportional system,and meet the control requirements for the position accuracy of the electro-hydraulic proportional system in the actual work of rock drilling robotic arms.

rock drilling robotic armradical basis function(RBF)neural network tuning PIDposition control accuracy of electro-hydraulic proportional systemjoint simulationMATLAB/SimulinkAMESim

陈翰文、徐巧玉、徐恺、张正

展开 >

河南科技大学 机电工程学院,河南 洛阳 471000

洛阳银杏科技有限公司,河南 洛阳 471000

凿岩机械臂 径向基函数神经网络整定PID 电液比例系统位置控制精度 联合仿真 MATLAB/Simulink AMESim

国家自然科学基金

51205108

2024

机电工程
浙江大学 浙江省机电集团有限公司

机电工程

CSTPCD北大核心
影响因子:0.785
ISSN:1001-4551
年,卷(期):2024.41(3)
  • 22