复烤机智能控制系统研究
|
摘要
烟叶复烤是在烤烟生产(初烤)的基础上再次烘烤形成的。烟叶复烤是烟叶从农产品转变到工业生产原料的过程，是烟草整体加工中非常重要的环节。通过打叶复烤可以减少烟叶的造碎，可以向卷烟厂提供高质量和规格化的原料，有利于烟片的长期贮存和醇化，节约贮存、醇化和运输费用。同时，在国际交易中，产品的形式主要是烟片和烟梗。因而烟片复烤的质量决定着一个复烤厂的生存和命运，特别是，随着中国加入WTO，国际化进一步加深，其重要性不言而语。据统计，每担烟每波动一个百分点，损失上百元人民币。仅宝丰复烤厂(年生产能力30万担)，每年损失就达数几百万元人民币。因此提高复烤厂的自动化程度，意义重大。
论文在熟悉复烤机工艺础上，针对烤段机理较明确的特点，建立了机理模型；针对回潮段机理不太明确的特点，建立了系统的辨识模型；并分析了影响系统的温湿度的因素。
复烤机复烤过程对象模型具有大滞后、多变量、非线性和扰动复杂等特点，属难控系统。经典控制理论和现代控制理论都要求被控对象有精确的数学模型，而复烤机的复烤过程的变化因素很多。因此，仅靠传统的控制方法难以使系统长期稳定在好的控制效果上。模糊控制、专家控制、学习控制、神经元网络控制等智能控制的研究，为解决模型不确定的系统提供了有力的工具。但同时我们也注意到。仅仅靠智能控制，忽略系统提供的定性模型，也是对信息资源的浪费，也不能达到预期效果。根据我们长期研究和实验的经验，应该把传统控制和智能控制结合起来，才能更好地控制对象。本文采用了基于模糊神经网络的预测控制。
预测控制是基于模型的控制，是20世纪70年代后期发展起来的一类新型的
郑州大学工学硕卜论文
计算机控制算法。这种算法的本质特征是它的三要素:预测模型、滚动优化和反
馈校正。预测模型，它用来描述对象的动态行为，根据系统当前的输入、输出信
息、预测未来的输出值。在线性系统情况下，可利用叠加原理，并利用系统的阶
跃响应、脉冲响应，或者状态空间模型等，预报系统未来的输出。在非线性系统
情况下，如果其非线性结构未知，且不能充分描述，则对其输出就难以作出精确
的预报，从而导致控制失败。由于模糊神经网络可做到较精确描述非线性动态过
程，因此，可用于这类基于模型的控制中。由于神经网络本身具有插值能力，因
此可以克服单纯模糊控制中“分挡”的限制，提高了控制精度;通过网络前向运
算实现模糊推理算法，大大简化了模糊推理的计算，提高了规则查询、推理的实
时性。在此基础上，论文提出了一种改进的四层框架的网络结构。该模型具有计
算简单;网络的每层推导均有明显的物理意义;具有全局逼近的性质。
利用C++编写了算法程序，给出了仿真结果，证明了系统具有很强的鲁棒性、
自学习和适应能力，从而达到了平稳控制的效果。利用组态软件编写了监控界面，
根据profibos协议，设计了系统的通信结构。根据烟草的不同品种的特性，进行
了三波段六光束测定和在线测量的扫描速率的测定，选择了水分测量的波段为
1.94um。并选择了其他硬件。
简一言之:以现场总线为物理实现形式，采取分段控制策略，融合传统控制和
智能控制的优点，实现系统的平稳控制、可靠控制。
To redry the tobacco leaf is a process in which the tobacco leaves become into the material of cigarette from farm produce .It plays a very important role in the process of making cigarette. When the baked tobacco leaves is baked again, then the redried tobacco leaves can be obtained .It has many advantages: the ratio of the broken leaves will be reduced greatly; The cigarette factories can get the high quality and standard material ;It also helps to store the leaves for a long time and does good to alcoholizing, so a lot of money will be saved. At present time, the main products are the leaves and stems in the international trade. Therefore, the profit of the cigarette is in the hand of the quality of the redried tobacco. Especially, after we have joined the WTO, more and more merchandises will be traded. If there is one percent above or below the standard about the containing water for one dan tobacco leaves, one hundred RMB will be lost .For Baofeng tobacco redrying factory, which has a product capacity of 30 dan, several million RMB is lost every year .So it is necessary to improve the automatic control system.
We set up the physical model for the redrying segment and statically model for the ordering model according to their characters, after we learned the technical skill of the redrying machine. Then we get the factors, which affect the temperature and humid of the tobacco.
The tobacco redryer is a plant, which has a large lag and multiple parameters, non-linear feature. So it is very difficult to control .the traditional control cannot qualify because it requires the exact model to control. Intelligence control, such as the fuzzy control, expert system and learned control, become a powerful tool to solute the plant that has an unclear model. However, if we give up the method of the traditional control, it means that much important information has been wasted.
According to our experience, we should combine the traditional control and 1C. In this paper, we introduce the MFC based on ANN.
MFC is a control method based on the model, which was put forward in 1770s. It has three parts: model predicting, continual optical and backward. Model predicting is to setup the model according the input and output data. For the linear system, it is very easy and difficult for the nonlinear system. In the paper, we use the FNN to predict the
model, which combine the advantages of fuzzy and ANN. We present a four layers FNN. which has a clear meanings and simple calculation.
According to its simulink programmed by C++, we analyze its result and it proofs that this arithmetic can satisfy our requirement. Then we give the supervision interface using the configure software. In the whole design of physics structure to realize the control idea, we introduce the advanced field bus of SIEMENS PLC in order to update the control system in the future. According to the spectrum experiment, we select 1. 94um as the measure wavelength. Other hardwires are also selected.
Briefly, what we have done is to design a control system which can guarantees the redryer work calmly, combining the traditional control method and intelligence control, making the advantage of the field bus as the physical form, controlling each subsystem separately by dividing the whole system into 12 subsystems.
论文在熟悉复烤机工艺础上，针对烤段机理较明确的特点，建立了机理模型；针对回潮段机理不太明确的特点，建立了系统的辨识模型；并分析了影响系统的温湿度的因素。
复烤机复烤过程对象模型具有大滞后、多变量、非线性和扰动复杂等特点，属难控系统。经典控制理论和现代控制理论都要求被控对象有精确的数学模型，而复烤机的复烤过程的变化因素很多。因此，仅靠传统的控制方法难以使系统长期稳定在好的控制效果上。模糊控制、专家控制、学习控制、神经元网络控制等智能控制的研究，为解决模型不确定的系统提供了有力的工具。但同时我们也注意到。仅仅靠智能控制，忽略系统提供的定性模型，也是对信息资源的浪费，也不能达到预期效果。根据我们长期研究和实验的经验，应该把传统控制和智能控制结合起来，才能更好地控制对象。本文采用了基于模糊神经网络的预测控制。
预测控制是基于模型的控制，是20世纪70年代后期发展起来的一类新型的
郑州大学工学硕卜论文
计算机控制算法。这种算法的本质特征是它的三要素:预测模型、滚动优化和反
馈校正。预测模型，它用来描述对象的动态行为，根据系统当前的输入、输出信
息、预测未来的输出值。在线性系统情况下，可利用叠加原理，并利用系统的阶
跃响应、脉冲响应，或者状态空间模型等，预报系统未来的输出。在非线性系统
情况下，如果其非线性结构未知，且不能充分描述，则对其输出就难以作出精确
的预报，从而导致控制失败。由于模糊神经网络可做到较精确描述非线性动态过
程，因此，可用于这类基于模型的控制中。由于神经网络本身具有插值能力，因
此可以克服单纯模糊控制中“分挡”的限制，提高了控制精度;通过网络前向运
算实现模糊推理算法，大大简化了模糊推理的计算，提高了规则查询、推理的实
时性。在此基础上，论文提出了一种改进的四层框架的网络结构。该模型具有计
算简单;网络的每层推导均有明显的物理意义;具有全局逼近的性质。
利用C++编写了算法程序，给出了仿真结果，证明了系统具有很强的鲁棒性、
自学习和适应能力，从而达到了平稳控制的效果。利用组态软件编写了监控界面，
根据profibos协议，设计了系统的通信结构。根据烟草的不同品种的特性，进行
了三波段六光束测定和在线测量的扫描速率的测定，选择了水分测量的波段为
1.94um。并选择了其他硬件。
简一言之:以现场总线为物理实现形式，采取分段控制策略，融合传统控制和
智能控制的优点，实现系统的平稳控制、可靠控制。
To redry the tobacco leaf is a process in which the tobacco leaves become into the material of cigarette from farm produce .It plays a very important role in the process of making cigarette. When the baked tobacco leaves is baked again, then the redried tobacco leaves can be obtained .It has many advantages: the ratio of the broken leaves will be reduced greatly; The cigarette factories can get the high quality and standard material ;It also helps to store the leaves for a long time and does good to alcoholizing, so a lot of money will be saved. At present time, the main products are the leaves and stems in the international trade. Therefore, the profit of the cigarette is in the hand of the quality of the redried tobacco. Especially, after we have joined the WTO, more and more merchandises will be traded. If there is one percent above or below the standard about the containing water for one dan tobacco leaves, one hundred RMB will be lost .For Baofeng tobacco redrying factory, which has a product capacity of 30 dan, several million RMB is lost every year .So it is necessary to improve the automatic control system.
We set up the physical model for the redrying segment and statically model for the ordering model according to their characters, after we learned the technical skill of the redrying machine. Then we get the factors, which affect the temperature and humid of the tobacco.
The tobacco redryer is a plant, which has a large lag and multiple parameters, non-linear feature. So it is very difficult to control .the traditional control cannot qualify because it requires the exact model to control. Intelligence control, such as the fuzzy control, expert system and learned control, become a powerful tool to solute the plant that has an unclear model. However, if we give up the method of the traditional control, it means that much important information has been wasted.
According to our experience, we should combine the traditional control and 1C. In this paper, we introduce the MFC based on ANN.
MFC is a control method based on the model, which was put forward in 1770s. It has three parts: model predicting, continual optical and backward. Model predicting is to setup the model according the input and output data. For the linear system, it is very easy and difficult for the nonlinear system. In the paper, we use the FNN to predict the
model, which combine the advantages of fuzzy and ANN. We present a four layers FNN. which has a clear meanings and simple calculation.
According to its simulink programmed by C++, we analyze its result and it proofs that this arithmetic can satisfy our requirement. Then we give the supervision interface using the configure software. In the whole design of physics structure to realize the control idea, we introduce the advanced field bus of SIEMENS PLC in order to update the control system in the future. According to the spectrum experiment, we select 1. 94um as the measure wavelength. Other hardwires are also selected.
Briefly, what we have done is to design a control system which can guarantees the redryer work calmly, combining the traditional control method and intelligence control, making the advantage of the field bus as the physical form, controlling each subsystem separately by dividing the whole system into 12 subsystems.
引文
[1] 《烟草科技》，1987年，第4期，2～5，4页
[2] 打叶复烤理论与技术，江西科学技术出版社1995
[3] 打叶技术与主要设备北京长征高科技1992
[4] 金闻博 1995 打叶复烤理论与技术江西科学技术出版社
[5] 云南省烟草公司基建处 1996 云南省打叶复烤发展建设体会
[6] Cardwell昆明船舶公司秦皇岛烟机公司打叶复烤设备技术资料
[7] 《卷烟工艺规范》中国烟草总公司1994版
[8] 《三益烟草有限公司12000kg／h打叶复烤线设备交验报告》
[9] 陈铁军．链系统方法及其应用[M]．郑州：河南科学技术出版社，1993，5．
[10] 孙增圻．智能控制理论与技术．清华大学出版社，1997．4
[11] 李士勇．模糊控制·神经控制和智能控制论．哈尔滨工业大学出版社，1998．9
[12] 易继锴，候媛彬．智能控制技术．北京工业大学出版社，1999．9
[13] 冯冬青，谢宋和等．模糊智能控制[M]．北京：化学工业出版社．1998，9．
[14] 林小峰，廖志伟，方辉．隶属函数对模糊控制性能的作用与影响[J]．电机与控制学报，1998，2(4)：197～200．
[15] 周鸣争．基于遗传算法的模糊隶属函数的优化及应用[J]．安徽机电学院学报，1998，13(4)：13～17．
[16] 邓兵，梁文林．模糊控制系统中量化因子与比例因子的研究[J]．电光与控制，1999，(1)：6～12．
[17] 晏勇，杜继宏，冯元琨．模糊控制[J]．计算机自动测量与控制，1999，7(1)：55～57．
[18] 孙增圻．智能控制理论与技术[M]．清华大学出版社，广西科学技术出版社．1997．4．
[19] 刘向杰，柴天佑，刘红波．动力锅炉燃烧系统的模糊控制策略[J]．自动化学报，1998，24(4)：535～538．
[20] 张凤芝，任长明，宗润宽等．氢氮比预测模糊控制[J]．工业控制计算机，2000，13(3)：47～48．
[21] 王玉，张永明，吴瑞生等．模糊控制在核子秤系统中的应用研究[J]．核电子学与探测技术，2001，21(1)：37～39．
[22] 于微波，刘俊萍，何伟明．烟叶发酵温度的模糊控制[J]．吉林工学院学报，
1998，19(1)：34～38．
[23] 刘增良主编．模糊技术与应用选编[M]．北京：北京航空航天大学出版社，1997，5．
[24] Hunt K J, et al. Neural Networks for Control Systems-A Survey. Automatica, 1992,28(6).
[25] Hunt K J, Sharbaro D. Neural networks for nonlinear internal model control. IEE Proceedings-D, 1991,138(5):431～438.
[26] 张乃尧，栾天．基于模糊神经网络的模型参考自适应控制[J]．自动化学报，1996，22(4)：476～480．
[27] Berenji H R, Khedkar P. Learning and tuning fuzzy logic controllers through reinforcements. IEEE Trans. On Neural Networks, 1992,3(5):724～740.
[28] Astrm K. J., Anton J. J., Arzen K. E.. Expert control. Automatica, 1986,22:227～286.
[29] Zhou Q J., Bai J K.. An intelligent control of novel design. Proc of Multinational Instrumentational Coference, 1983,137～150.
[30] 李淑兰主编．自然科学学科发展战略报告：自动化科学与技术．北京：科学出版社，1995．
[31] 张再兴，孙增圻．关于专家控制[J]．信息与控制，1994，24(3)：167～172．
[32] 黄山松，杜继宏，冯元琨．人工神经网络及其在控制中的应用[J]．计算机自动测量与控制，1999，7(2)：57～59．
[33] Special Issue on Application of Neural Networks. Proc. IEEE, 1996,84(10).
[34] Widrow B et al. Neural Networks, Application in Industry, Business and Science. Communication of the ACM, 37,1994:93～105.
[35] 冯慈璋，电磁场，人民教育出版社，1979，95
[36] 方志成，无损检测自动化与信息处理，国防工业出版社，1985，35～40
[37] 夏新民，烟叶复烤温湿度自动控制系统的分析《洛阳工学院学报》，1994年，第4期，72-76页
[38] 张建明，王宁等，PID自适应调整增益的神经元非模型控制，机电工程，1999(5)
[39] 万百五，大系统智能控制的进展，控制理论与应用，1995(1)
[40] 钟庆昌，谢剑英等，变参数PID控制器，信息与控制，1998(8)
[41]陶永华，尹怡欣等，《新型PID控制及其应用》，机械工业出版社，1999
[42]张峥 张嵘 Profibus现场总线在高效方坯连铸机控制系统中的应用，《世界仪表与自动化》，2002．1
[43]冯晓升，中国现场总线标准体系的形成与实施，《世界仪表与自动化》，2001．1
[44]郝赤强，温湿度微机自动测试系统的研制，《黑龙江烟草》，1995年，第8期，24-26页
[45]杨翠容，模糊解耦温湿度控制仪表及其应用，仪表技术与传感器，1998年第7期20-23
[46]李东辉，Fuzzy控制规则自调整和Fuzzy控制系统寻优及其仿真研究，模糊数学，1986，3：53-61．
[47]王金章等，模糊调节参数的在线自整定，控制与决策，1989，4：40-45．
[48]胡家耀等，模糊控制在退火炉燃烧过程控制中的应用，自动化学报，1989，6(15)：501-507．
[49]H Ying. Sufficient conditions on uniform approximation of multivariate functions by General Takagi-Sugeno fuzzy systems with linear rule consequent. IEEE Transactions on Systems, Man,& Cybernetics, 1998,28A:515-520.
[50]张乃尧，典型模糊控制器的结构分析[J]，模糊系统与数学，1997，11(2)：10～21．
[51]Tanaka K, Sugeno M, Stability analysis and design of fuzzy control systems. Fuzzy Sets and Systems, 1992,45(2):135-156.
[52]Holden, A.D.C. and Suddarth, S.C. Combined Netural-Net/knowledge-Based Adaptive Systems for large-Scale Dynamical Control. Internatimal Journal of Pattern Recognition and Artificial Intelligence. 1991(5)
[53]Widrow B er.al. Neural Networks, Application in Industry, Business and Science. Communication of the ACM, 37,1994:93～105
[54]Special Issue on Application of Neural Networks. Proc. IEEE, 1996,84(10)
[55]Kung S Y et al, Neural Networks for Intelligent Multimedia Processing ,Proc. IEEE, 1998,86:1244～1272 Antsaklis P J. Special Section on Neural Networks for Control System.. IEEE Control
[56] Hant K J, Rltaus, R Murray Smith. Neural networks for control systems-A surrey. Automatica. 1992 (6)
[57] Hornik K,M Stinchcombe, H White. Multilayer feed-forward networks are universal approximators. Neural Networks, 1998(2)
[58] Wang N.,,J. Chen and J. Wang, Neuron Intelligent Control For Hydraulic Turbine Generators, Proc of IEEE international conference on industrial technology, Guangzhou, 94TH06559-3,1994,288～292
[59] Miller W T. Real-time Application of Neural Networks for Sensor-based Control of Robots. With Vision,IEEE, Trans. Syst.,Man, Cybern, 1989, (19):825～831
[60] BnlsariA. Some Analytical Solutions to the General Approximation Problem for Feedforward Neural Networks. Neural Networks 1993, (6):991～996
[61] Hertz J, et al. Introduction to Theory of Neural Compution. Sant Fee Complexity Science Series, 1991:156
[62] Atrom, K J, Anton J J, Arzen K E. Expert control. Automatica. 1986(2)
[63] G.N. Saridis. Analytic for mulation of the principle of increasing precision with decreasing intelligence for intelligent machines. Automatica, 1989(3)
[64] P.J. Anlsaklis. An introduction to intelligent and autonomous control. Kluwer Academic Publishers. 1993
[65] 周东华，叶银忠，现代故障诊断与容错控制，清华大学出版社，2000．6
[66] 周东华，孙优贤，控制系统的故障检测与诊断技术，清华大学出版社，1994．9
[67] 张育林，李东旭，动态系统故障诊断理论与应用，国防科技大学出版社，1997．12
[68] Handelman, D.A. andStengel, R.F, Combining Expert System and Analytical Redundancy Concepts for Fault-tolerant FlightControl. J.Guid, Control, Dynam, 1989,12(1):39-45
[69] Wehmuller, K. and Nguyen, B, Reconfigurable ControlLaws for Control Reconfigurable Combat Aircraft Subjected to Actuat or Failures and Surface Damage, 1989, WRDC-TR-89-3052
[70] Stengel, F.R. Toward intelligent Flight Control, IEEE Trans. Syst, Man, Cybem, 1993,23(6):1699-1717.
[71] Rauch, E.H. etal, Fault Detection, Isolation, and Reconfiguration for
Aircraft Using Neural Networks, Aug. 1993, Porc. 1993 AIAA Conf. Guidance, Navigatoin, and Control.
[72] Kim, H. and Shin, K. G. On the Maximum Feedback Delayina Linear/Nolinear Control System with Inprt Distur-bances Causedby Controler Computer Failures, IEEE Trans. Control Syst. Tech, 1994, 2(2) :110-122
[73] Groiserg,L.B, On Reliability Estimationof Fault-tolerant Systemswith Reserve Recovery, 1994, Autom. Remote Control, 55(3) :441-449．
[74] Gao,Z. and Antsaklis, P. J, Actuator and Sensor Failure Accominodation with Incomplet Information. First IEEEC onf. on Control Application, Dayton, OH, USA, 13-16, Sept. 1992, 2:890-892
[75] Eryurek, E. AParallel, Fault Tolerant Controland Diagnostics Ystem for Nuclear Power Plants. Ph. D. thesis,Nuclear Engineering Dept, University of Tennessee. Aug. 1994
[76] 胡泽新．多变量系统故障诊断和容错控制新方法及其在精馏过程中的应用． 控制与决策．1994，9(4) ：286-190
[77] Garcia, E. H. etal.A Reconfigurable Hybrid Systemand Its Applicationto Power Plant Control. IEEE Trans. Control Systems Tech, 1995,3
[78] Willdk, A. S. A Survey of Design Methods for Failure Detectionin Dynamic Systemsm, Automatica, 1976, 12(6) :601-611
[79] Pazzani, M. J., failure Driven learning of Fault Diagnosis Heuistics, IEEETrans. Syst, Man, Cybern, 1987
[80] Frank, P. M, Engancement of Robustnessin Observer-based Fault Detection. Int, J. Control, 1994, 59(4) :955-982．
[81] Kinnaert, M. Hanus, R. andArte, Ph, Fault Detection and Iso lationfor Unstable Linear Systems. lEEETrans. Au-toma. Control. 1995, 40(4) : 740-743
[82] Massoumnia, M, Verghese, G, C, and Willsky, A. S, Failure Detection and Identification. IEEE Trans. Automa. Control. 1989． 34(3) :316-321
[83] Gertler, J. andSinger, D, Anew Structural Framework for Parity Equation-based Failure Detection and Isola-tion. Automatica. 1990, 26 (2) : 381-388．
[84] Prock,J., A New Technique for Fault Detection Using Petri Nets. Automatica. 1991,27(2) :239-245
[85] Wahner, J. and Shoureshi, R, A Robust Failure Diagnostic System for
Thermofluid Processes. Automatica. 1992, 28(2) :375-381．
[86] Nikiforov, I, Varavva, V, and Kireichikov, V, Applicanion of Statistical Fault Detection Algorithms to Navigation Systems Monitoring, Automatica. 1993,29(5) :1275-1290
[2] 打叶复烤理论与技术，江西科学技术出版社1995
[3] 打叶技术与主要设备北京长征高科技1992
[4] 金闻博 1995 打叶复烤理论与技术江西科学技术出版社
[5] 云南省烟草公司基建处 1996 云南省打叶复烤发展建设体会
[6] Cardwell昆明船舶公司秦皇岛烟机公司打叶复烤设备技术资料
[7] 《卷烟工艺规范》中国烟草总公司1994版
[8] 《三益烟草有限公司12000kg／h打叶复烤线设备交验报告》
[9] 陈铁军．链系统方法及其应用[M]．郑州：河南科学技术出版社，1993，5．
[10] 孙增圻．智能控制理论与技术．清华大学出版社，1997．4
[11] 李士勇．模糊控制·神经控制和智能控制论．哈尔滨工业大学出版社，1998．9
[12] 易继锴，候媛彬．智能控制技术．北京工业大学出版社，1999．9
[13] 冯冬青，谢宋和等．模糊智能控制[M]．北京：化学工业出版社．1998，9．
[14] 林小峰，廖志伟，方辉．隶属函数对模糊控制性能的作用与影响[J]．电机与控制学报，1998，2(4)：197～200．
[15] 周鸣争．基于遗传算法的模糊隶属函数的优化及应用[J]．安徽机电学院学报，1998，13(4)：13～17．
[16] 邓兵，梁文林．模糊控制系统中量化因子与比例因子的研究[J]．电光与控制，1999，(1)：6～12．
[17] 晏勇，杜继宏，冯元琨．模糊控制[J]．计算机自动测量与控制，1999，7(1)：55～57．
[18] 孙增圻．智能控制理论与技术[M]．清华大学出版社，广西科学技术出版社．1997．4．
[19] 刘向杰，柴天佑，刘红波．动力锅炉燃烧系统的模糊控制策略[J]．自动化学报，1998，24(4)：535～538．
[20] 张凤芝，任长明，宗润宽等．氢氮比预测模糊控制[J]．工业控制计算机，2000，13(3)：47～48．
[21] 王玉，张永明，吴瑞生等．模糊控制在核子秤系统中的应用研究[J]．核电子学与探测技术，2001，21(1)：37～39．
[22] 于微波，刘俊萍，何伟明．烟叶发酵温度的模糊控制[J]．吉林工学院学报，
1998，19(1)：34～38．
[23] 刘增良主编．模糊技术与应用选编[M]．北京：北京航空航天大学出版社，1997，5．
[24] Hunt K J, et al. Neural Networks for Control Systems-A Survey. Automatica, 1992,28(6).
[25] Hunt K J, Sharbaro D. Neural networks for nonlinear internal model control. IEE Proceedings-D, 1991,138(5):431～438.
[26] 张乃尧，栾天．基于模糊神经网络的模型参考自适应控制[J]．自动化学报，1996，22(4)：476～480．
[27] Berenji H R, Khedkar P. Learning and tuning fuzzy logic controllers through reinforcements. IEEE Trans. On Neural Networks, 1992,3(5):724～740.
[28] Astrm K. J., Anton J. J., Arzen K. E.. Expert control. Automatica, 1986,22:227～286.
[29] Zhou Q J., Bai J K.. An intelligent control of novel design. Proc of Multinational Instrumentational Coference, 1983,137～150.
[30] 李淑兰主编．自然科学学科发展战略报告：自动化科学与技术．北京：科学出版社，1995．
[31] 张再兴，孙增圻．关于专家控制[J]．信息与控制，1994，24(3)：167～172．
[32] 黄山松，杜继宏，冯元琨．人工神经网络及其在控制中的应用[J]．计算机自动测量与控制，1999，7(2)：57～59．
[33] Special Issue on Application of Neural Networks. Proc. IEEE, 1996,84(10).
[34] Widrow B et al. Neural Networks, Application in Industry, Business and Science. Communication of the ACM, 37,1994:93～105.
[35] 冯慈璋，电磁场，人民教育出版社，1979，95
[36] 方志成，无损检测自动化与信息处理，国防工业出版社，1985，35～40
[37] 夏新民，烟叶复烤温湿度自动控制系统的分析《洛阳工学院学报》，1994年，第4期，72-76页
[38] 张建明，王宁等，PID自适应调整增益的神经元非模型控制，机电工程，1999(5)
[39] 万百五，大系统智能控制的进展，控制理论与应用，1995(1)
[40] 钟庆昌，谢剑英等，变参数PID控制器，信息与控制，1998(8)
[41]陶永华，尹怡欣等，《新型PID控制及其应用》，机械工业出版社，1999
[42]张峥 张嵘 Profibus现场总线在高效方坯连铸机控制系统中的应用，《世界仪表与自动化》，2002．1
[43]冯晓升，中国现场总线标准体系的形成与实施，《世界仪表与自动化》，2001．1
[44]郝赤强，温湿度微机自动测试系统的研制，《黑龙江烟草》，1995年，第8期，24-26页
[45]杨翠容，模糊解耦温湿度控制仪表及其应用，仪表技术与传感器，1998年第7期20-23
[46]李东辉，Fuzzy控制规则自调整和Fuzzy控制系统寻优及其仿真研究，模糊数学，1986，3：53-61．
[47]王金章等，模糊调节参数的在线自整定，控制与决策，1989，4：40-45．
[48]胡家耀等，模糊控制在退火炉燃烧过程控制中的应用，自动化学报，1989，6(15)：501-507．
[49]H Ying. Sufficient conditions on uniform approximation of multivariate functions by General Takagi-Sugeno fuzzy systems with linear rule consequent. IEEE Transactions on Systems, Man,& Cybernetics, 1998,28A:515-520.
[50]张乃尧，典型模糊控制器的结构分析[J]，模糊系统与数学，1997，11(2)：10～21．
[51]Tanaka K, Sugeno M, Stability analysis and design of fuzzy control systems. Fuzzy Sets and Systems, 1992,45(2):135-156.
[52]Holden, A.D.C. and Suddarth, S.C. Combined Netural-Net/knowledge-Based Adaptive Systems for large-Scale Dynamical Control. Internatimal Journal of Pattern Recognition and Artificial Intelligence. 1991(5)
[53]Widrow B er.al. Neural Networks, Application in Industry, Business and Science. Communication of the ACM, 37,1994:93～105
[54]Special Issue on Application of Neural Networks. Proc. IEEE, 1996,84(10)
[55]Kung S Y et al, Neural Networks for Intelligent Multimedia Processing ,Proc. IEEE, 1998,86:1244～1272 Antsaklis P J. Special Section on Neural Networks for Control System.. IEEE Control
[56] Hant K J, Rltaus, R Murray Smith. Neural networks for control systems-A surrey. Automatica. 1992 (6)
[57] Hornik K,M Stinchcombe, H White. Multilayer feed-forward networks are universal approximators. Neural Networks, 1998(2)
[58] Wang N.,,J. Chen and J. Wang, Neuron Intelligent Control For Hydraulic Turbine Generators, Proc of IEEE international conference on industrial technology, Guangzhou, 94TH06559-3,1994,288～292
[59] Miller W T. Real-time Application of Neural Networks for Sensor-based Control of Robots. With Vision,IEEE, Trans. Syst.,Man, Cybern, 1989, (19):825～831
[60] BnlsariA. Some Analytical Solutions to the General Approximation Problem for Feedforward Neural Networks. Neural Networks 1993, (6):991～996
[61] Hertz J, et al. Introduction to Theory of Neural Compution. Sant Fee Complexity Science Series, 1991:156
[62] Atrom, K J, Anton J J, Arzen K E. Expert control. Automatica. 1986(2)
[63] G.N. Saridis. Analytic for mulation of the principle of increasing precision with decreasing intelligence for intelligent machines. Automatica, 1989(3)
[64] P.J. Anlsaklis. An introduction to intelligent and autonomous control. Kluwer Academic Publishers. 1993
[65] 周东华，叶银忠，现代故障诊断与容错控制，清华大学出版社，2000．6
[66] 周东华，孙优贤，控制系统的故障检测与诊断技术，清华大学出版社，1994．9
[67] 张育林，李东旭，动态系统故障诊断理论与应用，国防科技大学出版社，1997．12
[68] Handelman, D.A. andStengel, R.F, Combining Expert System and Analytical Redundancy Concepts for Fault-tolerant FlightControl. J.Guid, Control, Dynam, 1989,12(1):39-45
[69] Wehmuller, K. and Nguyen, B, Reconfigurable ControlLaws for Control Reconfigurable Combat Aircraft Subjected to Actuat or Failures and Surface Damage, 1989, WRDC-TR-89-3052
[70] Stengel, F.R. Toward intelligent Flight Control, IEEE Trans. Syst, Man, Cybem, 1993,23(6):1699-1717.
[71] Rauch, E.H. etal, Fault Detection, Isolation, and Reconfiguration for
Aircraft Using Neural Networks, Aug. 1993, Porc. 1993 AIAA Conf. Guidance, Navigatoin, and Control.
[72] Kim, H. and Shin, K. G. On the Maximum Feedback Delayina Linear/Nolinear Control System with Inprt Distur-bances Causedby Controler Computer Failures, IEEE Trans. Control Syst. Tech, 1994, 2(2) :110-122
[73] Groiserg,L.B, On Reliability Estimationof Fault-tolerant Systemswith Reserve Recovery, 1994, Autom. Remote Control, 55(3) :441-449．
[74] Gao,Z. and Antsaklis, P. J, Actuator and Sensor Failure Accominodation with Incomplet Information. First IEEEC onf. on Control Application, Dayton, OH, USA, 13-16, Sept. 1992, 2:890-892
[75] Eryurek, E. AParallel, Fault Tolerant Controland Diagnostics Ystem for Nuclear Power Plants. Ph. D. thesis,Nuclear Engineering Dept, University of Tennessee. Aug. 1994
[76] 胡泽新．多变量系统故障诊断和容错控制新方法及其在精馏过程中的应用． 控制与决策．1994，9(4) ：286-190
[77] Garcia, E. H. etal.A Reconfigurable Hybrid Systemand Its Applicationto Power Plant Control. IEEE Trans. Control Systems Tech, 1995,3
[78] Willdk, A. S. A Survey of Design Methods for Failure Detectionin Dynamic Systemsm, Automatica, 1976, 12(6) :601-611
[79] Pazzani, M. J., failure Driven learning of Fault Diagnosis Heuistics, IEEETrans. Syst, Man, Cybern, 1987
[80] Frank, P. M, Engancement of Robustnessin Observer-based Fault Detection. Int, J. Control, 1994, 59(4) :955-982．
[81] Kinnaert, M. Hanus, R. andArte, Ph, Fault Detection and Iso lationfor Unstable Linear Systems. lEEETrans. Au-toma. Control. 1995, 40(4) : 740-743
[82] Massoumnia, M, Verghese, G, C, and Willsky, A. S, Failure Detection and Identification. IEEE Trans. Automa. Control. 1989． 34(3) :316-321
[83] Gertler, J. andSinger, D, Anew Structural Framework for Parity Equation-based Failure Detection and Isola-tion. Automatica. 1990, 26 (2) : 381-388．
[84] Prock,J., A New Technique for Fault Detection Using Petri Nets. Automatica. 1991,27(2) :239-245
[85] Wahner, J. and Shoureshi, R, A Robust Failure Diagnostic System for
Thermofluid Processes. Automatica. 1992, 28(2) :375-381．
[86] Nikiforov, I, Varavva, V, and Kireichikov, V, Applicanion of Statistical Fault Detection Algorithms to Navigation Systems Monitoring, Automatica. 1993,29(5) :1275-1290
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |