基于LonWorks技术的变风量空调多变量解耦控制的研究
|
摘要
变风量空调系统具有多变量、彼此耦合强烈、非线性等特点，其自动控制系统的多个控制回路同时工作时，各个控制回路之间具有强烈的耦合作用，相互影响，相互干扰。这种耦合作用直接影响到变风量空调系统的控制品质和稳定性，本文通过建立被控对象的数学模型和设计解耦控制系统的方法以减小多个控制回路之间的耦合，提高变风量空调系统的控制性能和稳定性。
本文首次建立了变风量空调系统的机理模型，并通过采用机理分析和实验数据分析相结合的方法，建立了所研究的五输入、五输出的变风量空调系统的传递函数矩阵，并对这五个控制回路的被控对象的非线性数学模型，采用两种方法加以处理：第一种方法是在系统的工作点附近进行泰勒级数展开，使输出的变化量与输入的变化量之间呈现近似的线性关系；第二种方法是将非线性的数学模型进行分段线性化，即用分段线性化来逼近非线性化的模型。本文研究表明，这两种线性化方法用来处理变风量空调系统的非线性问题都是有效的。
解耦控制是一个难度比较大的研究领域，很重要的原因是解耦补偿器的设计很困难，一般情况下，三输入、三输出系统解耦的难度就已经很大。本文采用前馈补偿法设计了变风量空调系统的解耦补偿器，并得出了五输入、五输出变风量空调系统前馈补偿解耦器的传递函数矩阵，该解耦补偿器可以使本文所研究的变风量空调控制系统的开环传递函数矩阵和闭环传递函数矩阵都变换为对角矩阵，从而解除五个控制回路之间的耦合，使变风量空调系统实现解耦。此前未见到国内外文献有这方面的报道。
本文采用逆推法与遗传算法相结合的PID控制器的设计方法可以有效地控制变风量空调系统。本文直接找出PID参数与闭环控制系统的特征方程的根之间的关系，只要使特征方程的根在z平面的单位圆内任意取值，然后再求出相应的PID参数，则闭环控制系统必然是稳定的。本文所采用的这种确定PID参数的方法，虽然可以使特征方程的根在z平面的单位圆内任意取值，但各种取值情况对变风量空调系统的控制品质的影响却不同，因此，要对各种特征方程的根的取值情况进行寻优，本文采用遗传算法对特征方程的根进行寻优，从而使控制器的参数得以优化。
西安建筑科技大学博士学位论文
一
本文在变风量空调系统解耦控制的研究过程中，将网络控制技术中的Lonworks技术与变
风量空调技术有机地结合起来，建立了基于Lonworks技术的变风量空调解耦控制实验系统，
目前国内尚未见到有关这方面的报道，该实验系统在国内具有先进水平。
本文提出了实现Lonworks和METASYS网络进行数据交换的新见解。在本文所研究的变风
量空调解耦控制系统的实验装置中，存在着Lonworks和METASYS两套系统，而这两套系统的
通信协议遵守不同的标准，使得这两套系统之间的数据交换出现困难。本文在处理这一问题时
采用了一个巧妙的方法。由于 LON网络、METASYS网络及Visual Basic都支持动态数据交换
（DDE）协议，因此，可以将 LON网络的数据通过该系统中的 LNS DDE Server送到 Visual
Basic上；同时将METASYS网络的数据通过 Metalink DDE送到同一个Visual Basic上，经过
Visual Basic进行处理后，分别送给LON网络和WITASYS网络，从而实现这两个网络的数据
交换。该方法简单实用，不仅节省了硬件投资，还节省了开发网关所需的时间。
VAV system has the feature of multi-variable, intense conpling and nonlinear. When the several control loops in the control system are working together.there is intense coupling among them. The loops influence as well as interfere each other.The coupling may directly affect the property and the stability of VAV system.This thesis focuses on reducing and eliminating the coupling among the control loops by means of setting up the mathematical model of VAV system and designing the decoupling control system. Thus the controlling property and the stability of VAV system may be well improved.
Here the theoretical model of VAV system is set up for the first time.Besides,the transfer function matrix of the 5 inputs and the 5 outputs is established by combining the theoretical analysis with the data analysis of experiment. Then the non-linear mathematical model of the five loops is delt with in two method: In the first method, Taylor progression is applied near the working point of the system in order to make the volume of change between the output and the input become similar to a linear relationship. In the second mothed, the non-linear mathematical model realize linear by means of being devided into sections.The study shows that both the motheds are effective in dealing with the non-linear issue in VAV system.
In this thesis, the decoupling unit of VAV system is designed in the way of feedforward conpensation. Then the transfer function matrix of the feedforward conpensation decoupling unit of the 5 inputs & 5 outputs VAV system is worked out. The decoupling conpensation unit can change the transfer function matrix of the open loop and closed loop control system into diagonal matrix,so that the coupling among the five control loops is eliminated,and VAV system is decoupled. Before this thesis, similar report is unseen in both domestic and foreign reference.
This thesis presents an effective way to design PID controller through a combination of adverse deduce and genetic algorithm. The relationship between PID parameter and the root of the characteristic equation of the closed loop control system has been found out directly. As long as the root of the characteristic equation is within the unit circle of z-plane and the corresponding PID parameter is got, the closed loop control system is sure to be stable. Although the root of the characteristic equation can vary within the unit circle of z-plane,different value may affect the control property differently. Therefore,it is better to optimize the root, here the optimizing method is carried out by genetic algorithm.
In this thesis, the VAV decoupling control experiment system based on LonWorks technology is set up by means of combination of Lon Works technology of networks control technology and VAV technology. So far, there is no such report in domestic reference. So the experiment system can be regarded as advanced.
This thesis offers a new way to exchange data between LonWorks and Metasys networks. Acturelly, there are two systems in the experiment unit of the VAV decoupling control system. They are LonWorks system and Metasys system. As the communication protocols follow their respective standard, the data exchanging between the two systems becomes difficult. Here an ingenious method is adopted. Since LON net, Metasys net and Visual Basic all support dynamic data exchange (DDE)protocol, the data of LON net can be conveyed to Visual Basic through LNS DDE Server of LON system. Meanwhile, the data of Metasys net can be conveyed to the same Visual Basic through Metalink DDE . Then the Visual Basic is expected to deal with the data, after that, the data are conveyed back to LON net and Metasys net. Thus the data exchange is realized. Simple and practical, this method helps to save the cost of the hardware as well as the time spent in exploiting the gateway.
本文首次建立了变风量空调系统的机理模型，并通过采用机理分析和实验数据分析相结合的方法，建立了所研究的五输入、五输出的变风量空调系统的传递函数矩阵，并对这五个控制回路的被控对象的非线性数学模型，采用两种方法加以处理：第一种方法是在系统的工作点附近进行泰勒级数展开，使输出的变化量与输入的变化量之间呈现近似的线性关系；第二种方法是将非线性的数学模型进行分段线性化，即用分段线性化来逼近非线性化的模型。本文研究表明，这两种线性化方法用来处理变风量空调系统的非线性问题都是有效的。
解耦控制是一个难度比较大的研究领域，很重要的原因是解耦补偿器的设计很困难，一般情况下，三输入、三输出系统解耦的难度就已经很大。本文采用前馈补偿法设计了变风量空调系统的解耦补偿器，并得出了五输入、五输出变风量空调系统前馈补偿解耦器的传递函数矩阵，该解耦补偿器可以使本文所研究的变风量空调控制系统的开环传递函数矩阵和闭环传递函数矩阵都变换为对角矩阵，从而解除五个控制回路之间的耦合，使变风量空调系统实现解耦。此前未见到国内外文献有这方面的报道。
本文采用逆推法与遗传算法相结合的PID控制器的设计方法可以有效地控制变风量空调系统。本文直接找出PID参数与闭环控制系统的特征方程的根之间的关系，只要使特征方程的根在z平面的单位圆内任意取值，然后再求出相应的PID参数，则闭环控制系统必然是稳定的。本文所采用的这种确定PID参数的方法，虽然可以使特征方程的根在z平面的单位圆内任意取值，但各种取值情况对变风量空调系统的控制品质的影响却不同，因此，要对各种特征方程的根的取值情况进行寻优，本文采用遗传算法对特征方程的根进行寻优，从而使控制器的参数得以优化。
西安建筑科技大学博士学位论文
一
本文在变风量空调系统解耦控制的研究过程中，将网络控制技术中的Lonworks技术与变
风量空调技术有机地结合起来，建立了基于Lonworks技术的变风量空调解耦控制实验系统，
目前国内尚未见到有关这方面的报道，该实验系统在国内具有先进水平。
本文提出了实现Lonworks和METASYS网络进行数据交换的新见解。在本文所研究的变风
量空调解耦控制系统的实验装置中，存在着Lonworks和METASYS两套系统，而这两套系统的
通信协议遵守不同的标准，使得这两套系统之间的数据交换出现困难。本文在处理这一问题时
采用了一个巧妙的方法。由于 LON网络、METASYS网络及Visual Basic都支持动态数据交换
（DDE）协议，因此，可以将 LON网络的数据通过该系统中的 LNS DDE Server送到 Visual
Basic上；同时将METASYS网络的数据通过 Metalink DDE送到同一个Visual Basic上，经过
Visual Basic进行处理后，分别送给LON网络和WITASYS网络，从而实现这两个网络的数据
交换。该方法简单实用，不仅节省了硬件投资，还节省了开发网关所需的时间。
VAV system has the feature of multi-variable, intense conpling and nonlinear. When the several control loops in the control system are working together.there is intense coupling among them. The loops influence as well as interfere each other.The coupling may directly affect the property and the stability of VAV system.This thesis focuses on reducing and eliminating the coupling among the control loops by means of setting up the mathematical model of VAV system and designing the decoupling control system. Thus the controlling property and the stability of VAV system may be well improved.
Here the theoretical model of VAV system is set up for the first time.Besides,the transfer function matrix of the 5 inputs and the 5 outputs is established by combining the theoretical analysis with the data analysis of experiment. Then the non-linear mathematical model of the five loops is delt with in two method: In the first method, Taylor progression is applied near the working point of the system in order to make the volume of change between the output and the input become similar to a linear relationship. In the second mothed, the non-linear mathematical model realize linear by means of being devided into sections.The study shows that both the motheds are effective in dealing with the non-linear issue in VAV system.
In this thesis, the decoupling unit of VAV system is designed in the way of feedforward conpensation. Then the transfer function matrix of the feedforward conpensation decoupling unit of the 5 inputs & 5 outputs VAV system is worked out. The decoupling conpensation unit can change the transfer function matrix of the open loop and closed loop control system into diagonal matrix,so that the coupling among the five control loops is eliminated,and VAV system is decoupled. Before this thesis, similar report is unseen in both domestic and foreign reference.
This thesis presents an effective way to design PID controller through a combination of adverse deduce and genetic algorithm. The relationship between PID parameter and the root of the characteristic equation of the closed loop control system has been found out directly. As long as the root of the characteristic equation is within the unit circle of z-plane and the corresponding PID parameter is got, the closed loop control system is sure to be stable. Although the root of the characteristic equation can vary within the unit circle of z-plane,different value may affect the control property differently. Therefore,it is better to optimize the root, here the optimizing method is carried out by genetic algorithm.
In this thesis, the VAV decoupling control experiment system based on LonWorks technology is set up by means of combination of Lon Works technology of networks control technology and VAV technology. So far, there is no such report in domestic reference. So the experiment system can be regarded as advanced.
This thesis offers a new way to exchange data between LonWorks and Metasys networks. Acturelly, there are two systems in the experiment unit of the VAV decoupling control system. They are LonWorks system and Metasys system. As the communication protocols follow their respective standard, the data exchanging between the two systems becomes difficult. Here an ingenious method is adopted. Since LON net, Metasys net and Visual Basic all support dynamic data exchange (DDE)protocol, the data of LON net can be conveyed to Visual Basic through LNS DDE Server of LON system. Meanwhile, the data of Metasys net can be conveyed to the same Visual Basic through Metalink DDE . Then the Visual Basic is expected to deal with the data, after that, the data are conveyed back to LON net and Metasys net. Thus the data exchange is realized. Simple and practical, this method helps to save the cost of the hardware as well as the time spent in exploiting the gateway.
引文
[1] Alexander A. Tseytlin. Making VAV More Efficient and Flexible. Heating/Piping/Air Conditioning, 1993(2).
[2] 陈向阳．变风量空调系统的自动控制．暖通空调，Vol．27，No．3，34-39，1997．
[3] Darrell D Massie. Predicting control plant HVAC equipment performance using neural networks-laboratory system test results. ASHRAE Trans, 1998,104(1).
[4] Erik Jeannette. Experimental results of a predictive neural network HVAC controller. ASHRAE Trans, 1998,104(2).
[5] Avery,Gil. The Instability of VAV Systems. Heating/Piping/Air Conditioning, 1992(2).
[6] 霍小平．变风量系统的概念、分类及应用实例．暖通空调，Vol．27，No．5，22-26，1997．
[7] 彭士梅．空调风机的变风量调节．暖通空调，Vol．31，No．1，50-51，2001．
[8] 王维新，常本康．智能建筑中空调系统的综合节能．南京理工大学学报，1999，(10)．
[9] 刘作军，董砚．智能建筑VAV空调系统的节能控制．自动化与仪表，2000，(5)．
[10] 赵乱成：智能建筑设备自动化技术，西安：西安电子科技大学，2002．
[11] Tuncaly Tanyolu. A neural network approach with self-organized principal component analysis for identifcation of dehumidifying coils. ASHRAE Trans, 1999,105(1).
[12] 李太富．智能控制在空调控制中应用及探讨．四川工业学院学报，V19(1)．
[13] 吴爱国，高德云．VAV空调系统的智能控制．制冷学报，1999，(1)．
[14] Chen T,Athienitic AK. Ambient temperature and solar radiation prediction for predictive control of HVAC systems and a methodology for optimal building heating dynamic operation. ASHRAE Transactions, 1996.
[15] Ahmed O, Mitchell JW, Klein SA. Application of general regression neural network in HVAC process identification and control. ASHRAE Transactions, 1996.
[16] Brian A Rock. Performance of fixed, air-side economizer, and neural network demand-controlled ventilation in CAV system. ASHRAE Trans, 1998,104(2).
[17] 蒋红梅：模糊PID控制在变风量空调系统中的应用研究，硕士学位论文，西安建筑科技大学，13-27，2001．
[18] Larry R. VAV mixed air plenum pressure control. Heating/Piping/Air Conditioning,1995,(8).
[19] Osman Ahmed. Feedforward-feedback controller using general regression neural network(GRNN) for laboratory HVAC system: Part Ⅰ-pressure control. ASHRAE Trans, 1998,104 (2). Ⅱ Ⅲ
[20] 李克欣．变风量空调系统的VPT控制法及其应用．暖通空调，1999，(3)．
[21] 戴斌文，狄洪发，马先民．变风量空调系统风机静压控制方法研究．建筑热能通风空调，2000，(3)，6-10．
[22] 彭士梅．空调系统的VAV末端．暖通空调，Vol．29，No．3，42-45，1999．
[23] 金以慧：过程控制，北京：清华大学出版社，1993．
[24] Osman Ahmed. Feedforward-feedback controller using general regression neural network(GRNN) for laboratory HVAC system: Part Ⅱ- temperature control-cooling. ASHRAE Trans, 1998, 104(2).
[25] Osman Ahmed. Feedforward-feedback controller using general regression neural network(GRNN) for laboratory HVAC system: Part Ⅲ- temperature control-heating. ASHRAE Trans, 1998, 104(2).
[26] 钱以明，杨书明．VAV系统中新风量的控制．暖通空调，Vol．29，No．3，39-41，1999．
[27] 汪善国．新风量、新风和新回风系统及控制．暖通空调，Vol．27，No．1，23-28，1997．
[28] Stephen C Carpenter. Energy and IAQ impacts of CO_2-based demand-controlled ventilation. ASHRAE Trans, 1996,V102(2).
[29] David M Elovitz. Minimum outside air control methods for VAV systems. ASHRAE Trans, 1995,V101(2).
[30] Wang Shengwei, Jin Xinqiao. CO_2-based occupancy detection for on-line outdoor air control. Indoor and Built Environment 1998,7(3).
[31] 宋宏光，变风量系统最小新风量控制方法的讨论．暖通空调，1999，(3)．
[32] 吴茂杰，耿世彬．VAV系统最小新风量的控制．建筑热能通风空调，2000，(3)，38-39．
[33] Wang Shengwei, Jin Xinqiao. On-line outdoor air flow control by estimating occupancy. Air Conditioning in High Rise Building '97, Shanghai, 1997.
[34] J P Kettler. controlling minimum ventilation volume in VAV systems. ASHRAE J, 1998, (5).
[35] D A Stanke. Ventilation where it is needed. ASHRAE J, October 1998.
[36] 高维平，栾国颜，闫丽萍．多变量控制系统控制回路的识别．计算机与应用化学，Vol．15，No．1，53-55，1998．
[37] Cao,Y.Y.,Lam,J.,& Sun, Y.X. Static output feedback stabilization: An ILMI approach. Automatica, 1998.
[38] 孙宁，等．变风量系统耦合特性研究．1998年全国暖通空调学术会议论文集，387-395．
[39] 巩小军．变风量空调系统控制的有关问题与工程实践．智能建筑，2000，(1)，67-71．
[40] Ding Z. Robust adaptive control of nonlinear output-feedback systems underbounded disturbances. IEE Proc. Control Theory Appl.,1998.
[41] Ding Z. Global adaptive output feedback stabilization of nonlinear systems of any relative degree with unknown high frequency gain. IEEE Trans. Autom. Control, 1998.
[42] Chen, H., & Allgower. F. A quasi-infinite horizon nonlinear model predictive control scheme with guaranteed stability. Automatica, 1998.
[43] De Nicolao, G.,Magni, L., & Scattolini, R. Stabilizing receding horizon control of nonlinear time-varying systems. IEEE Trans. Autom. Control, 1998.
[44] 刘豹，王正欣：系统辨识，北京：机械工业出版社，135-170，1993．
[45] 方崇智，萧德云：过程辨识，北京：清华大学出版社，1988．
[46] 秦勇，贾利民，张锡第．多变量模糊系统建模与控制理论．计算机仿真，Vol．16，No．3，15-18，1999．
[47] 赵铁山．基于模糊控制理论的无模型自适应控制技术．计算机自动测量与控制，2001，(3)，15-17．
[48] 马维明．多变量复杂系统的模糊控制．自动化与仪表，Vol．13，No．4，32-34，1998．
[49] 李玉云，王永骥．人工神经网络在暖通空调领域的应用研究发展．暖通空调，2001，(1)．
[50] 靳其兵，顾树生．一种参数快速收敛的多变量系统神经网络PID控制．控制与决策，Vol．13(s)，448-452，1998．
[51] 毛义梅，罗海福，张晶．一种PID参数模糊自整定控制器的设计与仿真．自动化与仪表，Vol．16，No．3，36-38，2001．
[52] 佟绍成，柴天佑．一类多变量非线性动态系统的鲁棒自适应模糊控制．自动化学报，Vol．26，No．5，630-636，2000．
[53] 晋欣桥，夏清．多区域VAV空调系统及其局部DDC控制器的动态模拟．制冷学报，1999，(1)，17-23．
[54] 杨自厚：自动控制原理，北京：冶金工业出版社，20-23，1990．
[55] 周谟仁：流体力学泵与风机，北京：中国建筑工业出版社，1994．
[56] 黄俊，王兆安：电力电子变流技术，北京：机械工业出版社，1994．
[57] 黄甘洲，王宜贵，刘立军：计算机网络，西安：西安电子科技大学，2000．
[58] 王桂增，王诗宓：高等过程控制，北京：清华大学出版社，2002．
[59] 陈芝久：制冷装置自动化，北京：机械工业出版社，8-29，2002．
[60] Wang Shengwei. Dynamic simulation of building VAV air-conditioning system and evaluation of EMCS on-line control strategies. Building and Environment, 1999,34,681-705.
[61] Wang Shengwei, Jin xinqiao. Model-based optimal control of VAV air-conditioning system using genetic algorithm. Building and Environment,2000,35,471-487.
[62] 王晓哲，李界家，吴成东，顾树生．多变量系统解耦方法综述，沈阳建筑工程学院学报，Vol．16，No．2，143-145，2000，(2)．
[63] 邵裕森：过程控制及仪表，上海：上海交通大学出版社，1999．
[64] Ding Z. Almost disturbance decoupling of uncertain output feedback systems, IEE Proc.Control Theory Appl., 1999.
[65] 董玲，白焰．多变量系统的神经元解耦控制．现代电力，Vol．16，No．1，11-15，1999．
[66] 李秀改，侯媛彬．模糊神经网络在温度解耦控制中的研究与应用．现代电子技术，2000，(6)，4-6．
[67] 黄道平，朱学峰，胥布工．自适应神经元非模型多变量优化补偿控制．控制理论与应用，Vol．15，No．3，346-350，1998．
[68] 舒怀林．PID神经元网络对强耦合带时延多变量系统的解耦控制．控制理论与应用，1998，6．
[69] 李学锋，杨先发，周凤歧．多变量模糊解耦控制系统研究与应用，西北工业大学学报，Vol．13，No．1，56-59，1995．
[70] 席爱民：计算机控制系统，西安：陕西科学技术出版社，91-111，1994．
[71] 吴沧浦：最优控制的理论与方法，北京：国防工业出版社，2000．
[72] 屈文忠，邱阳．基于遗传算法的多变量模糊控制器的设计方法．西安交通大学学报，Vol．33，No．6，101-103，1999．
[73] 徐丽娜：神经网络控制，哈尔滨：哈尔滨工业大学出版社，67-86，1999．
[74] 王锦标，方崇智：过程计算机控制，北京：清华大学出版社，67-86，1997．
[75] 李江伟，汪仁煌．信息网络技术对工业控制技术发展的促进作用．电气自动化，2001，(1)．
[76] 黄一夫：微型计算机控制技术，北京：机械工业出版社，2000．
[77] 阳宪惠：现场总线技术及其应用，北京：清华大学出版社，4-8，1999．
[78] 杨育红：LON网络控制技术及应用，西安：西安电子科技大学，1999．
[79] 杨育红，涂敏，李滨：LON网络程序设计，西安：西安电子科技大学出版社，2001．
[80] 林生：计算机通信与网络教程，北京：清华大学出版社，1999．
[81] 徐银泉：交流调速系统及其应用，北京：中国纺织出版社，1996．
[82] 张树兵，戴红，陈哲：Visual Basic 6.0程序设计，北京：清华大学出版社，297-312，2001．
[2] 陈向阳．变风量空调系统的自动控制．暖通空调，Vol．27，No．3，34-39，1997．
[3] Darrell D Massie. Predicting control plant HVAC equipment performance using neural networks-laboratory system test results. ASHRAE Trans, 1998,104(1).
[4] Erik Jeannette. Experimental results of a predictive neural network HVAC controller. ASHRAE Trans, 1998,104(2).
[5] Avery,Gil. The Instability of VAV Systems. Heating/Piping/Air Conditioning, 1992(2).
[6] 霍小平．变风量系统的概念、分类及应用实例．暖通空调，Vol．27，No．5，22-26，1997．
[7] 彭士梅．空调风机的变风量调节．暖通空调，Vol．31，No．1，50-51，2001．
[8] 王维新，常本康．智能建筑中空调系统的综合节能．南京理工大学学报，1999，(10)．
[9] 刘作军，董砚．智能建筑VAV空调系统的节能控制．自动化与仪表，2000，(5)．
[10] 赵乱成：智能建筑设备自动化技术，西安：西安电子科技大学，2002．
[11] Tuncaly Tanyolu. A neural network approach with self-organized principal component analysis for identifcation of dehumidifying coils. ASHRAE Trans, 1999,105(1).
[12] 李太富．智能控制在空调控制中应用及探讨．四川工业学院学报，V19(1)．
[13] 吴爱国，高德云．VAV空调系统的智能控制．制冷学报，1999，(1)．
[14] Chen T,Athienitic AK. Ambient temperature and solar radiation prediction for predictive control of HVAC systems and a methodology for optimal building heating dynamic operation. ASHRAE Transactions, 1996.
[15] Ahmed O, Mitchell JW, Klein SA. Application of general regression neural network in HVAC process identification and control. ASHRAE Transactions, 1996.
[16] Brian A Rock. Performance of fixed, air-side economizer, and neural network demand-controlled ventilation in CAV system. ASHRAE Trans, 1998,104(2).
[17] 蒋红梅：模糊PID控制在变风量空调系统中的应用研究，硕士学位论文，西安建筑科技大学，13-27，2001．
[18] Larry R. VAV mixed air plenum pressure control. Heating/Piping/Air Conditioning,1995,(8).
[19] Osman Ahmed. Feedforward-feedback controller using general regression neural network(GRNN) for laboratory HVAC system: Part Ⅰ-pressure control. ASHRAE Trans, 1998,104 (2). Ⅱ Ⅲ
[20] 李克欣．变风量空调系统的VPT控制法及其应用．暖通空调，1999，(3)．
[21] 戴斌文，狄洪发，马先民．变风量空调系统风机静压控制方法研究．建筑热能通风空调，2000，(3)，6-10．
[22] 彭士梅．空调系统的VAV末端．暖通空调，Vol．29，No．3，42-45，1999．
[23] 金以慧：过程控制，北京：清华大学出版社，1993．
[24] Osman Ahmed. Feedforward-feedback controller using general regression neural network(GRNN) for laboratory HVAC system: Part Ⅱ- temperature control-cooling. ASHRAE Trans, 1998, 104(2).
[25] Osman Ahmed. Feedforward-feedback controller using general regression neural network(GRNN) for laboratory HVAC system: Part Ⅲ- temperature control-heating. ASHRAE Trans, 1998, 104(2).
[26] 钱以明，杨书明．VAV系统中新风量的控制．暖通空调，Vol．29，No．3，39-41，1999．
[27] 汪善国．新风量、新风和新回风系统及控制．暖通空调，Vol．27，No．1，23-28，1997．
[28] Stephen C Carpenter. Energy and IAQ impacts of CO_2-based demand-controlled ventilation. ASHRAE Trans, 1996,V102(2).
[29] David M Elovitz. Minimum outside air control methods for VAV systems. ASHRAE Trans, 1995,V101(2).
[30] Wang Shengwei, Jin Xinqiao. CO_2-based occupancy detection for on-line outdoor air control. Indoor and Built Environment 1998,7(3).
[31] 宋宏光，变风量系统最小新风量控制方法的讨论．暖通空调，1999，(3)．
[32] 吴茂杰，耿世彬．VAV系统最小新风量的控制．建筑热能通风空调，2000，(3)，38-39．
[33] Wang Shengwei, Jin Xinqiao. On-line outdoor air flow control by estimating occupancy. Air Conditioning in High Rise Building '97, Shanghai, 1997.
[34] J P Kettler. controlling minimum ventilation volume in VAV systems. ASHRAE J, 1998, (5).
[35] D A Stanke. Ventilation where it is needed. ASHRAE J, October 1998.
[36] 高维平，栾国颜，闫丽萍．多变量控制系统控制回路的识别．计算机与应用化学，Vol．15，No．1，53-55，1998．
[37] Cao,Y.Y.,Lam,J.,& Sun, Y.X. Static output feedback stabilization: An ILMI approach. Automatica, 1998.
[38] 孙宁，等．变风量系统耦合特性研究．1998年全国暖通空调学术会议论文集，387-395．
[39] 巩小军．变风量空调系统控制的有关问题与工程实践．智能建筑，2000，(1)，67-71．
[40] Ding Z. Robust adaptive control of nonlinear output-feedback systems underbounded disturbances. IEE Proc. Control Theory Appl.,1998.
[41] Ding Z. Global adaptive output feedback stabilization of nonlinear systems of any relative degree with unknown high frequency gain. IEEE Trans. Autom. Control, 1998.
[42] Chen, H., & Allgower. F. A quasi-infinite horizon nonlinear model predictive control scheme with guaranteed stability. Automatica, 1998.
[43] De Nicolao, G.,Magni, L., & Scattolini, R. Stabilizing receding horizon control of nonlinear time-varying systems. IEEE Trans. Autom. Control, 1998.
[44] 刘豹，王正欣：系统辨识，北京：机械工业出版社，135-170，1993．
[45] 方崇智，萧德云：过程辨识，北京：清华大学出版社，1988．
[46] 秦勇，贾利民，张锡第．多变量模糊系统建模与控制理论．计算机仿真，Vol．16，No．3，15-18，1999．
[47] 赵铁山．基于模糊控制理论的无模型自适应控制技术．计算机自动测量与控制，2001，(3)，15-17．
[48] 马维明．多变量复杂系统的模糊控制．自动化与仪表，Vol．13，No．4，32-34，1998．
[49] 李玉云，王永骥．人工神经网络在暖通空调领域的应用研究发展．暖通空调，2001，(1)．
[50] 靳其兵，顾树生．一种参数快速收敛的多变量系统神经网络PID控制．控制与决策，Vol．13(s)，448-452，1998．
[51] 毛义梅，罗海福，张晶．一种PID参数模糊自整定控制器的设计与仿真．自动化与仪表，Vol．16，No．3，36-38，2001．
[52] 佟绍成，柴天佑．一类多变量非线性动态系统的鲁棒自适应模糊控制．自动化学报，Vol．26，No．5，630-636，2000．
[53] 晋欣桥，夏清．多区域VAV空调系统及其局部DDC控制器的动态模拟．制冷学报，1999，(1)，17-23．
[54] 杨自厚：自动控制原理，北京：冶金工业出版社，20-23，1990．
[55] 周谟仁：流体力学泵与风机，北京：中国建筑工业出版社，1994．
[56] 黄俊，王兆安：电力电子变流技术，北京：机械工业出版社，1994．
[57] 黄甘洲，王宜贵，刘立军：计算机网络，西安：西安电子科技大学，2000．
[58] 王桂增，王诗宓：高等过程控制，北京：清华大学出版社，2002．
[59] 陈芝久：制冷装置自动化，北京：机械工业出版社，8-29，2002．
[60] Wang Shengwei. Dynamic simulation of building VAV air-conditioning system and evaluation of EMCS on-line control strategies. Building and Environment, 1999,34,681-705.
[61] Wang Shengwei, Jin xinqiao. Model-based optimal control of VAV air-conditioning system using genetic algorithm. Building and Environment,2000,35,471-487.
[62] 王晓哲，李界家，吴成东，顾树生．多变量系统解耦方法综述，沈阳建筑工程学院学报，Vol．16，No．2，143-145，2000，(2)．
[63] 邵裕森：过程控制及仪表，上海：上海交通大学出版社，1999．
[64] Ding Z. Almost disturbance decoupling of uncertain output feedback systems, IEE Proc.Control Theory Appl., 1999.
[65] 董玲，白焰．多变量系统的神经元解耦控制．现代电力，Vol．16，No．1，11-15，1999．
[66] 李秀改，侯媛彬．模糊神经网络在温度解耦控制中的研究与应用．现代电子技术，2000，(6)，4-6．
[67] 黄道平，朱学峰，胥布工．自适应神经元非模型多变量优化补偿控制．控制理论与应用，Vol．15，No．3，346-350，1998．
[68] 舒怀林．PID神经元网络对强耦合带时延多变量系统的解耦控制．控制理论与应用，1998，6．
[69] 李学锋，杨先发，周凤歧．多变量模糊解耦控制系统研究与应用，西北工业大学学报，Vol．13，No．1，56-59，1995．
[70] 席爱民：计算机控制系统，西安：陕西科学技术出版社，91-111，1994．
[71] 吴沧浦：最优控制的理论与方法，北京：国防工业出版社，2000．
[72] 屈文忠，邱阳．基于遗传算法的多变量模糊控制器的设计方法．西安交通大学学报，Vol．33，No．6，101-103，1999．
[73] 徐丽娜：神经网络控制，哈尔滨：哈尔滨工业大学出版社，67-86，1999．
[74] 王锦标，方崇智：过程计算机控制，北京：清华大学出版社，67-86，1997．
[75] 李江伟，汪仁煌．信息网络技术对工业控制技术发展的促进作用．电气自动化，2001，(1)．
[76] 黄一夫：微型计算机控制技术，北京：机械工业出版社，2000．
[77] 阳宪惠：现场总线技术及其应用，北京：清华大学出版社，4-8，1999．
[78] 杨育红：LON网络控制技术及应用，西安：西安电子科技大学，1999．
[79] 杨育红，涂敏，李滨：LON网络程序设计，西安：西安电子科技大学出版社，2001．
[80] 林生：计算机通信与网络教程，北京：清华大学出版社，1999．
[81] 徐银泉：交流调速系统及其应用，北京：中国纺织出版社，1996．
[82] 张树兵，戴红，陈哲：Visual Basic 6.0程序设计，北京：清华大学出版社，297-312，2001．