采用神经网络PID控制器改进单杆液压执行器的研究
|
摘要
针对单杆液压执行器运动轨迹误差较大问题,采用改进神经网络PID控制器进行改进。建立单杆液压执行器简图模型,给出末端执行器位移方程式和液压缸流量方程式。分析PID控制器控制流程,引用BP神经网络结构,采用粒子群算法优化BP神经网络PID控制器参数,通过MATLAB软件对单杆液压执行器跟踪误差进行仿真。结果显示:单杆液压执行器采用PID控制器,X和Y方向产生的最大误差分别为8.10×10-5 m和8.90×10-5 m,跟踪误差较大;单杆液压执行器采用改进BP神经网络PID控制器,X和Y方向产生的最大误差分别为1.80×10-5m和2.10×10-5 m,跟踪误差较小。采用改进BP神经网络PID控制器,单杆液压执行器跟踪精度较高,能够实现高精度定位要求。
An improved neural network PID controller is adopted to solve the problem of large trajectory error of single rod hydraulic actuator. A simple model of single hydraulic actuator is established, and displacement equation of end effector and flow equation of hydraulic cylinder are given. The control flow of PID controller is analyzed, BP neural network structure is quoted and particle swarm optimization(PSO) is used to optimize the parameters of BP neural network PID controller. The tracking error of single rod hydraulic actuator is simulated by MATLAB software. The results show that the single bar hydraulic actuator uses PID controller, the maximum error in X and Y direction is 8.10×10-5 m and 8.90×10-5 m respectively. The single bar hydraulic actuator uses improved BP neural network PID controller, the maximum error of X and Y direction is 1.80×10-5 m and 2.10×10-5 m, and the tracking error is small. Using the improved BP neural network PID controller, the single rod hydraulic actuator has higher tracking accuracy and can achieve high precision positioning requirements.
An improved neural network PID controller is adopted to solve the problem of large trajectory error of single rod hydraulic actuator. A simple model of single hydraulic actuator is established, and displacement equation of end effector and flow equation of hydraulic cylinder are given. The control flow of PID controller is analyzed, BP neural network structure is quoted and particle swarm optimization(PSO) is used to optimize the parameters of BP neural network PID controller. The tracking error of single rod hydraulic actuator is simulated by MATLAB software. The results show that the single bar hydraulic actuator uses PID controller, the maximum error in X and Y direction is 8.10×10-5 m and 8.90×10-5 m respectively. The single bar hydraulic actuator uses improved BP neural network PID controller, the maximum error of X and Y direction is 1.80×10-5 m and 2.10×10-5 m, and the tracking error is small. Using the improved BP neural network PID controller, the single rod hydraulic actuator has higher tracking accuracy and can achieve high precision positioning requirements.
引文
[1]张翼鹏.液压变压器控制直线执行机构工作特性研究[D].哈尔滨:哈尔滨工业大学,2013.
[2]吴杰.新型单轨机车液压系统的研发[D].青岛:山东科技大学,2014.
[3]KARVONEN M,HEIKKIL?M,HUOVA M,et al.Analysis by Simulation of Different Control Algorithms of A Digital Hydraulic Two-Actuator System[J].International Journal of Fluid Power,2014,15(1):33-44.
[4]陈健,高慧斌,郭劲,等.数字化单杆控制系统设计[J].光学精密工程,2013,21(11):2844-2849.
[5]BENDER F A,G?LTZ S,BR?UNL T,et al.Model-ing and Offset-Free Model Predictive Control of a Hydraulic Mini Excavator[J].IEEE Transactions on Automation Science and Engineering,2017,14(4):1683-1692.
[6]王鹏飞,刁修慧.液压式挖掘机单杆操作系统的设计[J].机床与液压,2014,42(20):117-119.
[7]庞文杰,王云超,叶存达,等.多轴车辆转向系统神经网络控制算法研究[J].计算机仿真,2015,32(6):172-176.
[8]ANUPAM Kumar,VIJAY Kumar.Evolving an interval type-2 fuzzy PID controller for the redundant robotic manipul-ator[J].Expert Systems with Applications,2017,73(8):162-176.
[9]随聪慧.粒子群算法的改进方法研究[D].成都:西南交通大学,2010.
[10]BOUKENS M,BOUKABOU A.Design of an intelligent optimal neural network-based tracking controller for nonholonomic mobile robot systems[J].Neurocomputing,2016,226.
[2]吴杰.新型单轨机车液压系统的研发[D].青岛:山东科技大学,2014.
[3]KARVONEN M,HEIKKIL?M,HUOVA M,et al.Analysis by Simulation of Different Control Algorithms of A Digital Hydraulic Two-Actuator System[J].International Journal of Fluid Power,2014,15(1):33-44.
[4]陈健,高慧斌,郭劲,等.数字化单杆控制系统设计[J].光学精密工程,2013,21(11):2844-2849.
[5]BENDER F A,G?LTZ S,BR?UNL T,et al.Model-ing and Offset-Free Model Predictive Control of a Hydraulic Mini Excavator[J].IEEE Transactions on Automation Science and Engineering,2017,14(4):1683-1692.
[6]王鹏飞,刁修慧.液压式挖掘机单杆操作系统的设计[J].机床与液压,2014,42(20):117-119.
[7]庞文杰,王云超,叶存达,等.多轴车辆转向系统神经网络控制算法研究[J].计算机仿真,2015,32(6):172-176.
[8]ANUPAM Kumar,VIJAY Kumar.Evolving an interval type-2 fuzzy PID controller for the redundant robotic manipul-ator[J].Expert Systems with Applications,2017,73(8):162-176.
[9]随聪慧.粒子群算法的改进方法研究[D].成都:西南交通大学,2010.
[10]BOUKENS M,BOUKABOU A.Design of an intelligent optimal neural network-based tracking controller for nonholonomic mobile robot systems[J].Neurocomputing,2016,226.