改进BP神经网络PID控制的机械臂电液伺服系统  

Electro-hydraulic Servo System of Manipulator Controlled by Improved BP Neural Network PID

作　　者：张森 韦明 王豪 ZHANG Sen;WEI Ming;WANG Hao(Xi’an Aerospace Precision Mechatronics Institute,Xi’an 710000,China)

机构地区：[1]西安航天精密机电研究所,西安710000

出　　处：《自动化与仪表》2025年第4期23-28,共6页Automation & Instrumentation

摘　　要：针对机械臂电液伺服系统中的液压缸位置跟踪控制精度问题,提出一种基于改进BP神经网络的液压缸位移PID控制策略。首先,分析了电液伺服控制系统工作原理并建立数学模型;其次,引入自适应动量项、改进的激活函数及改进的拟牛顿法来优化BP神经网络,提高神经网络的映射能力以及响应速度,实现对PID控制参数的自适应整定;最后,在Matlab实验平台对液压缸位置跟踪和抗扰动能力进行仿真。仿真结果表明,与传统PID控制和BP-PID控制相比,改进的BP-PID控制抗干扰能力和鲁棒性更强,可以有效提高电液伺服系统位置跟踪精度和响应速度。Aiming at the accuracy of hydraulic cylinder position tracking control in electro-hydraulic servo system of manipulator,a PID control strategy of hydraulic cylinder displacement based on improved BP neural network is proposed.Firstly,the working principle of electro-hydraulic servo control system is analyzed and the mathematical model is established.Secondly,the adaptive momentum term,improved activation function and improved quasi-Newton method are introduced to optimize the BP neural network,improve the mapping ability and response speed of the neural network,and realize the adaptive tuning of PID control parameters.Finally,the position tracking and anti-disturbance ability of hydraulic cylinder are simulated on the Matlab experimental platform.The simulation results show that compared with traditional PID control and BP-PID control,the improved BP-PID control has stronger anti-disturbance ability and robustness,which can effectively improve the position tracking accuracy and response speed of electro-hydraulic servo system.

关 键 词：电液伺服系统 位置控制 改进的BP神经网络 PID控制器 

分 类 号：TP242.2[自动化与计算机技术—检测技术与自动化装置]