基于单神经元PID的并联机构稳定控制研究
|
摘要
并联机构由于非线性强、控制过程复杂,导致常规PID控制参数难以整定。为了实现对二自由度并联机构更精确的控制,结合神经网络,文中提出了一种基于单神经元的自适应PID控制策略,以实现PID控制参数的实时整定。通过dSPACE对被控对象进行半实物仿真,仿真结果表明,设计的单神经元PID控制器控制效果优于传统PID控制器。
Parallel mechanism is a closed loop system,it 's composed of multiple parallel chains.Compared with the traditional mechanism,it has a series of advantages,such as high stiffness,high precision,high bearing capacity and so on. However,because of its strong nonlinearity and complicated control process,the control parameters of conventional PID are difficult to set. To solve the above problem,on the basis of traditional PID control theory,an adaptive PID control strategy based on single neuron is proposed,in order to realize the real-time tuning of PID control parameters. The semi-physical simulation is designed to verify the correctness of algorithm. The simulation results show that the control effect of the single neuron PID controller is better than that of the traditional PID controller.
Parallel mechanism is a closed loop system,it 's composed of multiple parallel chains.Compared with the traditional mechanism,it has a series of advantages,such as high stiffness,high precision,high bearing capacity and so on. However,because of its strong nonlinearity and complicated control process,the control parameters of conventional PID are difficult to set. To solve the above problem,on the basis of traditional PID control theory,an adaptive PID control strategy based on single neuron is proposed,in order to realize the real-time tuning of PID control parameters. The semi-physical simulation is designed to verify the correctness of algorithm. The simulation results show that the control effect of the single neuron PID controller is better than that of the traditional PID controller.
引文
[1]Shu H L,Pi Y G. PID neural networks for time-de-laysystems[J].Computers and Chemical Engineering,2000,24:859-862.
[2]戴雅馨.浅析PID参数整定方法[J].纯碱工业,2009(6):15-17.
[3]李砚浓,李汀兰,姜艺,等.基于RBF神经网络自适应PID四旋翼飞行器控制[J].控制工程,2016,23(3):378-382.
[4]王述彦,师宇,冯忠绪.基于模糊PID控制器的控制方法研究[J].机械科学与技术,2011,30(1):166-172.
[5]Zhang C N,Zhao M,Wang M. Single Neuron Rational Model of Arithmetic and Logic Operations[J]. IEEE Transactions on Neural Networks,2011,3(11):739-747.
[6]朴海国,王志新,张华强.基于合作粒子群算法的PID神经网络非线性控制系统[J].控制理论与应用,2009,26(12):1318-1324.
[7]杨金芳,马平,张华磊.自适应单神经元网络在耦合系统中的应用[J].信息技术,2004,28(10):42-44.
[8]丁军,徐用懋.单神经元自适应PID控制器及其应用[J].控制工程,2004,11(1):27-30.
[9]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计[M].北京:清华大学出版社,2004:36-40.
[10]刘金琨.先进PID控制先进PID控制MATLAB仿真[M].北京:电子工业出版社,2006:301-324.
[11]潘峰,薛定宇,徐心和.基于d SPACE半实物仿真技术的伺服控制研究与应用[J].系统仿真学报,2004,16(5):936-939.
[12]朱冰,贾晓峰,王御,等.基于双dSPACE的汽车动力学集成控制快速原型试验[J].吉林大学学报:工学版,2016,46(1):8-14.
[2]戴雅馨.浅析PID参数整定方法[J].纯碱工业,2009(6):15-17.
[3]李砚浓,李汀兰,姜艺,等.基于RBF神经网络自适应PID四旋翼飞行器控制[J].控制工程,2016,23(3):378-382.
[4]王述彦,师宇,冯忠绪.基于模糊PID控制器的控制方法研究[J].机械科学与技术,2011,30(1):166-172.
[5]Zhang C N,Zhao M,Wang M. Single Neuron Rational Model of Arithmetic and Logic Operations[J]. IEEE Transactions on Neural Networks,2011,3(11):739-747.
[6]朴海国,王志新,张华强.基于合作粒子群算法的PID神经网络非线性控制系统[J].控制理论与应用,2009,26(12):1318-1324.
[7]杨金芳,马平,张华磊.自适应单神经元网络在耦合系统中的应用[J].信息技术,2004,28(10):42-44.
[8]丁军,徐用懋.单神经元自适应PID控制器及其应用[J].控制工程,2004,11(1):27-30.
[9]周开利,康耀红.神经网络模型及其MATLAB仿真程序设计[M].北京:清华大学出版社,2004:36-40.
[10]刘金琨.先进PID控制先进PID控制MATLAB仿真[M].北京:电子工业出版社,2006:301-324.
[11]潘峰,薛定宇,徐心和.基于d SPACE半实物仿真技术的伺服控制研究与应用[J].系统仿真学报,2004,16(5):936-939.
[12]朱冰,贾晓峰,王御,等.基于双dSPACE的汽车动力学集成控制快速原型试验[J].吉林大学学报:工学版,2016,46(1):8-14.