基于自校正控制的大滞后控制过程的研究
|
摘要
在实际工业生产过程中,广泛存在着时滞。时滞的存在对工业控制系统的性能产生不利影响,特别是当时滞较大,惯性时间较长并且过程对象的动态特性随各种因素的影响而发生变化时,采用传统的控制方法,如PID控制方法,很难获得良好的控制效果,甚至会出现不稳定现象。因此,如何改善大滞后控制过程的控制品质是一个有着较大实际意义的课题。
本文对自校正控制理论的背景进行了介绍。研究了目前应用于实际工业大滞后系统中常用的PID调节器、Smith预估和自校正控制这三种控制算法,经仿真研究,分析比较了它们在大滞后控制系统应用中的优缺点,得出了自校正控制技术更具有优势的结论。
通过仿真研究了算法中参数的设置问题。本文对于模型的阶次,经过仿真发现,可将阶次设置为一个比实际模型的阶次稍大的值,这样既不影响系统的稳定性,又可以适应过程对象模型结构的慢时变性;使用损失函数法来确定模型的滞后时间常数;对于递推参数和协方差矩阵,在仿真开始时,利用自校正控制器先对其进行辨识,但控制器并不对控制输出进行计算,将自校正控制器投入运行时,这两个参数就已经是辨识好的并且适合于本系统的初值。这样设置参数能既保证控制系统的稳定性,又适应系统时变的动态特性。并且针对自校正调节器所存在的问题,即控制器的输出振荡的问题和系统崩溃问题,在控制算法中让遗忘因子等于0.9999,同时对控制器的输出、递推参数、增益向量和协方差矩阵引入加权滤波的控制思想,给出了一种修正的自校正控制算法。
最后将修正的自校正控制算法应用到换热元件综合试验室的加热器温度控制系统中进行仿真,这个系统是一个典型的具有大时滞、大惯性及慢时变特性的控制系统。仿真结果表明,其控制性能明显优于常规的PID控制器。
The time delay is widely existed in modern industrial process. It has negative effects on the control performance of the system, especially when delay time is large, inertia time is long and dynamic characteristics of the process object is influenced by the various factors. The conventional control method such as PID control can not meet requirements and even cause the stability of the system worse. Therefore, it is an important research with actual meaning to improve the control quality of the system with delay time.
In this paper, the theory of the self-tuning control is presented firstly. Then many control algorithms of current large time delay system are studied, for example PID regulator, Smith predictor controller and self-tuning controller. Through simulation, comparing their advantages and disadvantages in the large time delay control system, the result shows that self-tuning control is comparatively the best method.
The problem of how to set up the parameter is studied through simulation. In this paper, a lot of simulations show that the order of the model in the algorithm can be set a little larger than the actual object, after setting it like that, the controller do not affect the stability of the system, but also can adapt the time-varying character of the system; loss function is used to determine the time delay constant; and the recursive parameter and covariance matrix has been identified at the beginning of the simulation, the controller do not calculate the output, just identify the recursive parameter and covariance matrix, so when the self-tuning is used, these two parameters has been identified and very fit for the system. With these operations, the controller can not only guarantee stability of the control system, but also adapt the time-varying dynamic characteristics. Aiming at the problem of self-tuning regulator, that is oscillation of the controller output and breakdown of the system. Forgetting factor is determined to 0.9999, and the weighted filtering control thought is introduced into output of the controller, recursive parameter, gain vector, and covariance matrix. These two algorithms are combined together, and a kind of the modified self-tuning control is proposed.
At last, the modified self-tuning control arithmetic is applied to the temperature control system of the heater in hot-exchange integrated lab. The system typically possesses heating single direction, large time delay, big inertia and slow time varying characters. The result of the simulation reveals that its performance is much better than PID control.
本文对自校正控制理论的背景进行了介绍。研究了目前应用于实际工业大滞后系统中常用的PID调节器、Smith预估和自校正控制这三种控制算法,经仿真研究,分析比较了它们在大滞后控制系统应用中的优缺点,得出了自校正控制技术更具有优势的结论。
通过仿真研究了算法中参数的设置问题。本文对于模型的阶次,经过仿真发现,可将阶次设置为一个比实际模型的阶次稍大的值,这样既不影响系统的稳定性,又可以适应过程对象模型结构的慢时变性;使用损失函数法来确定模型的滞后时间常数;对于递推参数和协方差矩阵,在仿真开始时,利用自校正控制器先对其进行辨识,但控制器并不对控制输出进行计算,将自校正控制器投入运行时,这两个参数就已经是辨识好的并且适合于本系统的初值。这样设置参数能既保证控制系统的稳定性,又适应系统时变的动态特性。并且针对自校正调节器所存在的问题,即控制器的输出振荡的问题和系统崩溃问题,在控制算法中让遗忘因子等于0.9999,同时对控制器的输出、递推参数、增益向量和协方差矩阵引入加权滤波的控制思想,给出了一种修正的自校正控制算法。
最后将修正的自校正控制算法应用到换热元件综合试验室的加热器温度控制系统中进行仿真,这个系统是一个典型的具有大时滞、大惯性及慢时变特性的控制系统。仿真结果表明,其控制性能明显优于常规的PID控制器。
The time delay is widely existed in modern industrial process. It has negative effects on the control performance of the system, especially when delay time is large, inertia time is long and dynamic characteristics of the process object is influenced by the various factors. The conventional control method such as PID control can not meet requirements and even cause the stability of the system worse. Therefore, it is an important research with actual meaning to improve the control quality of the system with delay time.
In this paper, the theory of the self-tuning control is presented firstly. Then many control algorithms of current large time delay system are studied, for example PID regulator, Smith predictor controller and self-tuning controller. Through simulation, comparing their advantages and disadvantages in the large time delay control system, the result shows that self-tuning control is comparatively the best method.
The problem of how to set up the parameter is studied through simulation. In this paper, a lot of simulations show that the order of the model in the algorithm can be set a little larger than the actual object, after setting it like that, the controller do not affect the stability of the system, but also can adapt the time-varying character of the system; loss function is used to determine the time delay constant; and the recursive parameter and covariance matrix has been identified at the beginning of the simulation, the controller do not calculate the output, just identify the recursive parameter and covariance matrix, so when the self-tuning is used, these two parameters has been identified and very fit for the system. With these operations, the controller can not only guarantee stability of the control system, but also adapt the time-varying dynamic characteristics. Aiming at the problem of self-tuning regulator, that is oscillation of the controller output and breakdown of the system. Forgetting factor is determined to 0.9999, and the weighted filtering control thought is introduced into output of the controller, recursive parameter, gain vector, and covariance matrix. These two algorithms are combined together, and a kind of the modified self-tuning control is proposed.
At last, the modified self-tuning control arithmetic is applied to the temperature control system of the heater in hot-exchange integrated lab. The system typically possesses heating single direction, large time delay, big inertia and slow time varying characters. The result of the simulation reveals that its performance is much better than PID control.
引文
[1]张培建,吴建国.大纯滞后系统的研究与应用.东南大学学报,2004,34(9):271-275.
[2]潘峰,韩如成.时变大时滞系统的控制方法综述.仪器仪表学报.2002,23(3):789-791.
[3]吴士昌,吴忠强.自适应控制.北京:机械工业出版社,2005.
[4]Astrom K J and Hagglund T.PID Controllers:Theory,design and tuning.Research Triangle Park:Instrument Society of America,1995.
[5]闫秀英.一种自校正PID控制器设计与仿真.系统仿真学报.2006,18(2):753-756.
[6]Tang K S,Man K F,Chen G et al.An optimal fuzzy PID controller.IEEE Trans.on Industrial Electronics,2001,48(4):757-765.
[7]Chen Y Q,Moore K L.Relay feedback tuning of robust PID controllers with iso-damping property.IEEE Transactions on Systems,Man,and Cybernetics,Part B.2005,35(1):23-31.
[8]Chee F,Fernando,T L,Savkin A V et al.Expert PID control system for blood glucose control in critically ill patients.IEEE Trans.on information technology in biomedicine.2003,7(4):419-425.
[9]Ho S J,Li S S,Shinn Y H.Optimizing fuzzy neural networks for tuning PID controllers using an orthogonal simulated annealing algorithm OSA.IEEE Trans.on Fuzzy Systems.2006,14(3):421-434.
[10]李正军.计算机控制系统.北京:机械工业出版社,2005.
[11]Crowe J,Ranmker B L.Dynamic matrix control-a computer control algorithm.Proc,Joint Automat,Control.Theory Appl..2002,149(1):26-31.
[12]Hang C C.A Performance Study of Control System with Dead Time.IEEE.IECI.1980,27(3):234-241.
[13]傅燕,周琼,杨建刚.改进型Smith预估控制系统的仿真.计算机仿真.1998,15(4):39-41.
[14]Shanmugathasan N,Johnston R D.Exploitation of Time Delays for Improved Process Control.Int.J.Control.1988,48(3):1137-1162.
[15]Huang S A,Tank K.Adaptive Control of Machanical Systems Using Neural Networks.IEEE Trans.on Systems,Man,and Cybernetics,Part C:Applications and Reviews.2007,37(5):897-903.
[16]陆会明,孙敬松,邓慧.电站典型大滞后过程自适应预估控制器设计及应用.南京航空航天大学学报.2006,38(1):151-153.
[17]侯远龙,王引生.自抗扰控制在电液伺服系统中的应用研究.机床与液压.2008,36(8):220-221.
[18]Huang S,Ta K K,Lee T H.An improvement on stable adaptive control for a class of nonlinear systems.IEEE Trans.on Automatic Control.2004,49(8):1398-1403.
[19]林金星,沈炯,李益国.基于多模型切换的过热汽温自适应预测控制.东南大学学报.2008,38(1):69-74.
[20]Bernard J A.Use of Rule-based System for Process Control.IEEE Contr.Syst.Mag.1988,8(5):3-13.
[21]Lightbody G,George W I.Nonlinear Control Structures Based on Embedded Neural System Models.IEEE Trans,on Neural Networks.1997,8(3):553-567.
[22]王耀南,童调生,蔡自兴.基于神经元网络的智能PID控制及应用.信息与控制.1994,23(3):185-189.
[23]Kickert W J.and Lemke H R.Aplication of a Fuzzy Controller in Warm Water Plant.Automatics.1976,12(2):301-308.
[24]田存生,马世伟,王玉曾.基于神经网络的具有Smith预估器的PID控制.兰州大学学报(自然科学版).1995,31(4):105-108.
[25]Chen S B.Self-learning Fuzzy Neural Networks for Control of Uncertain Systems with Time Delays.IEEE Trans,on Systems,Man,and Cybernetics-Part B:Cybernetics.1997,27(1):142-148.
[26]李清泉.系统辨识与自适应控制.北京:科学出版社,1990.
[27]王鹏,彭光正,伍清河.自校正调节器在具有阻力负载的气动位置伺服系统中的应用.控制理论与应用,2004,21(4):561-564.
[28]徐政.减摇鳍的自适应控制.上海:上海交通大学,2002.
[29]邓自立,郭一新.动态系统分析及其应用.沈阳:辽宁科学技术出版社,1985.
[30]刘金琨.先进PID控制及其MATLAB仿真.北京:电子工业出版社,2003.
[31]Ge M,Chiu M S,Wang Q G.Robust PID controller design via LMI approach.Journal of process control,2002,(12):3-13.
[32]Lee K C,Lee S.Performance evaluation of switched Ethernet for networked control systems.IECON Proceedings,28th Industrial Electronics Conference(IECON 02),Sevilla,Spain,2002,(4):3170-3175.
[33]陈夕松,汪木兰.过程控制系统.北京:科学出版社,2004.
[34]谢新民,丁峰.自适应控制系统.北京:清华大学出版社,2002.
[35]李清泉.自校正调节器的收敛性和控制幅度.清华大学学报(自然科学).1986.29-37.
[36]Zhihong Xiu,Guang Ren.Optimization Design of TS-PID Fuzzy Controllers Based on Genetic Algorithms.WCICA 2004,45(3):2476-2480.
[37]贺生明.平整机延伸率自校正控制系统的研究.沈阳:东北大学,2007.
[38]李晨明,谌玉红,张百海.基于模糊自整定PID的单片机温度控制系统设计.科学技术与工程,2008,8(15).
[39]朱云鹏.换热元件试验系统改造.大连:大连理工大学,2005.
[2]潘峰,韩如成.时变大时滞系统的控制方法综述.仪器仪表学报.2002,23(3):789-791.
[3]吴士昌,吴忠强.自适应控制.北京:机械工业出版社,2005.
[4]Astrom K J and Hagglund T.PID Controllers:Theory,design and tuning.Research Triangle Park:Instrument Society of America,1995.
[5]闫秀英.一种自校正PID控制器设计与仿真.系统仿真学报.2006,18(2):753-756.
[6]Tang K S,Man K F,Chen G et al.An optimal fuzzy PID controller.IEEE Trans.on Industrial Electronics,2001,48(4):757-765.
[7]Chen Y Q,Moore K L.Relay feedback tuning of robust PID controllers with iso-damping property.IEEE Transactions on Systems,Man,and Cybernetics,Part B.2005,35(1):23-31.
[8]Chee F,Fernando,T L,Savkin A V et al.Expert PID control system for blood glucose control in critically ill patients.IEEE Trans.on information technology in biomedicine.2003,7(4):419-425.
[9]Ho S J,Li S S,Shinn Y H.Optimizing fuzzy neural networks for tuning PID controllers using an orthogonal simulated annealing algorithm OSA.IEEE Trans.on Fuzzy Systems.2006,14(3):421-434.
[10]李正军.计算机控制系统.北京:机械工业出版社,2005.
[11]Crowe J,Ranmker B L.Dynamic matrix control-a computer control algorithm.Proc,Joint Automat,Control.Theory Appl..2002,149(1):26-31.
[12]Hang C C.A Performance Study of Control System with Dead Time.IEEE.IECI.1980,27(3):234-241.
[13]傅燕,周琼,杨建刚.改进型Smith预估控制系统的仿真.计算机仿真.1998,15(4):39-41.
[14]Shanmugathasan N,Johnston R D.Exploitation of Time Delays for Improved Process Control.Int.J.Control.1988,48(3):1137-1162.
[15]Huang S A,Tank K.Adaptive Control of Machanical Systems Using Neural Networks.IEEE Trans.on Systems,Man,and Cybernetics,Part C:Applications and Reviews.2007,37(5):897-903.
[16]陆会明,孙敬松,邓慧.电站典型大滞后过程自适应预估控制器设计及应用.南京航空航天大学学报.2006,38(1):151-153.
[17]侯远龙,王引生.自抗扰控制在电液伺服系统中的应用研究.机床与液压.2008,36(8):220-221.
[18]Huang S,Ta K K,Lee T H.An improvement on stable adaptive control for a class of nonlinear systems.IEEE Trans.on Automatic Control.2004,49(8):1398-1403.
[19]林金星,沈炯,李益国.基于多模型切换的过热汽温自适应预测控制.东南大学学报.2008,38(1):69-74.
[20]Bernard J A.Use of Rule-based System for Process Control.IEEE Contr.Syst.Mag.1988,8(5):3-13.
[21]Lightbody G,George W I.Nonlinear Control Structures Based on Embedded Neural System Models.IEEE Trans,on Neural Networks.1997,8(3):553-567.
[22]王耀南,童调生,蔡自兴.基于神经元网络的智能PID控制及应用.信息与控制.1994,23(3):185-189.
[23]Kickert W J.and Lemke H R.Aplication of a Fuzzy Controller in Warm Water Plant.Automatics.1976,12(2):301-308.
[24]田存生,马世伟,王玉曾.基于神经网络的具有Smith预估器的PID控制.兰州大学学报(自然科学版).1995,31(4):105-108.
[25]Chen S B.Self-learning Fuzzy Neural Networks for Control of Uncertain Systems with Time Delays.IEEE Trans,on Systems,Man,and Cybernetics-Part B:Cybernetics.1997,27(1):142-148.
[26]李清泉.系统辨识与自适应控制.北京:科学出版社,1990.
[27]王鹏,彭光正,伍清河.自校正调节器在具有阻力负载的气动位置伺服系统中的应用.控制理论与应用,2004,21(4):561-564.
[28]徐政.减摇鳍的自适应控制.上海:上海交通大学,2002.
[29]邓自立,郭一新.动态系统分析及其应用.沈阳:辽宁科学技术出版社,1985.
[30]刘金琨.先进PID控制及其MATLAB仿真.北京:电子工业出版社,2003.
[31]Ge M,Chiu M S,Wang Q G.Robust PID controller design via LMI approach.Journal of process control,2002,(12):3-13.
[32]Lee K C,Lee S.Performance evaluation of switched Ethernet for networked control systems.IECON Proceedings,28th Industrial Electronics Conference(IECON 02),Sevilla,Spain,2002,(4):3170-3175.
[33]陈夕松,汪木兰.过程控制系统.北京:科学出版社,2004.
[34]谢新民,丁峰.自适应控制系统.北京:清华大学出版社,2002.
[35]李清泉.自校正调节器的收敛性和控制幅度.清华大学学报(自然科学).1986.29-37.
[36]Zhihong Xiu,Guang Ren.Optimization Design of TS-PID Fuzzy Controllers Based on Genetic Algorithms.WCICA 2004,45(3):2476-2480.
[37]贺生明.平整机延伸率自校正控制系统的研究.沈阳:东北大学,2007.
[38]李晨明,谌玉红,张百海.基于模糊自整定PID的单片机温度控制系统设计.科学技术与工程,2008,8(15).
[39]朱云鹏.换热元件试验系统改造.大连:大连理工大学,2005.