基于模糊神经网络的自适应控制在pH中和过程控制中的应用
|
摘要
中和过程是一个典型的非线性、纯时延过程，用常规的线性控制方法不可能对其进行有效控制。在对中和过程自动控制算法的探索中，采用了非线性增益补偿控制和前馈控制，由于中和控制过程的反应过程变化复杂，模型难以确定，因此基于模型的前馈控制无法对pH值和流量值的波动进行有效的补偿。常规PID控制算法也无法对反应的变化过程做快速的、精确的反应，在控制过程中很容易引起振荡现象，很难对系统实现最优控制。
对此，本文提出了用一种基于模糊神经网络MRAC控制方案来实现中和过程的优化控制。模糊神经网络是用神经元网络来构造模糊系统，其即具有模糊系统善于表达人的经验性知识，又具有神经元网络的根据输入输出样本来自动设计和调整模糊系统的设计参数，实现模糊系统的自学习和自适应功能的特性。基于模糊神经网络的自适应控制方案可以在线学习和调整规则参数及隶属函数参数，其控制规则表是根据现场实际控制数据并通过计算得出的，再将之应用于现场的控制中，具有很高的适应性，并保证闭环控制系统的稳定性，有高度鲁棒性，适用于非线性、时变、时滞系统的控制。因此，模糊神经网络控制的应用能很好的克服这种缺陷，适合于中和过程的非线性变化过程。
本文的研究工作主要包括以下内容：
(1)综述了智能控制的研究发展，介绍了模糊控制技术和神经网络控制技术的各自优缺点，以及神经网络控制器的典型结构和几种常用的神经网络，自适应控制的特点及其应用。
(2)分析了中和过程控制的非线性时延特性和控制难度。介绍在实际应用中一般采用的控制方法：非线性增益补偿PID控制和前馈控制及其控制的效果，分析了存在的问题和困难。
(3)根据模糊系统和神经网络的优缺点，提出了用神经网络来构造模糊系统，一个基于模糊神经网络控制算法。设计了模糊神经网络控制器，以及把基于模糊神经网络的自适应控制应用到中和过程，建模并仿真。
(4)提出了基于PLC的pH中和过程自动控制系统，并把基于模糊神经网络自适应控制算法应用到某电化厂的pH中和过程自动控制系统中，最后分析了基于模糊神经网络自适应控制算法在中和过程的应用结果，以及PLC在pH中和自
摘要
动控制系统中的应用结果。
本文最后对全文所做的工作进行了总结，并提出了今后进一步研究所需要做
的工作。
Neutralization process is a typical nonlinear process with pure time delay. This process can't be controlled effectively with conventional linear control method. Feed back control base on Nonlinear PID with self-adjust parameter and feed forward control base on experience model are employed in the practical application. The realization of this control system and its result are introduced in this paper. Because the process of the reaction is complicated, it's difficult to model for it. The Nonlinear PID can't fit the change of the reaction. The feed forward control base on model also can't fit the change of pH value and flux of the original water. So the control method can't get a good result.
In terms of the above considerations, this thesis proposes a new nonlinear adaptive control method based on fuzzy neural network to realize optimization control of pH neutralization process. Fuzzy neural network not only has the characteristic of fuzzy system, but also has the characteristic of neural network. The nonlinear adaptive control method based on fuzzy neural network can learn and adjust rule parameter and subjection parameter on line. And the control rule list is calculated according to control data on the spot, which is also applied to the control on the spot. Therefore, it has good adaptive and robust and also can ensure the stability of closed loop control system, which is fit to handle both the nonlinearities and time-varying characteristics of the pH process.
The research work of this thesis consists of the following major parts:
Firstly, this thesis summarizes the development of intelligent control and also introduces the advantage and disadvantage of fuzzy control technique and neural control technique respectively. The characteristic and application of adaptive control are also presented in this thesis.
Secondly, the traditional control for pH process ?the compensation of nonlinearities plus PID control is presented in this thesis.
Thirdly, the fuzzy neural network controller is designed and the adaptive control method based on fuzzy neural network is proposed according to the advantage and disadvantage of fuzzy system and neural network in this thesis. And the adaptive control method based on fuzzy neural network is applied to pH process.
Forth, Aimed at the process of pH value control of wastewater treatment in chemical factory, a PLC-based automatic control system of pH value of wastewater treatment is developed. In this thesis operation principle of this system, hardware configuration, and software configuration are all discussed in detail.
At last, a summary of the paper is given and further research interests are also proposed.
对此，本文提出了用一种基于模糊神经网络MRAC控制方案来实现中和过程的优化控制。模糊神经网络是用神经元网络来构造模糊系统，其即具有模糊系统善于表达人的经验性知识，又具有神经元网络的根据输入输出样本来自动设计和调整模糊系统的设计参数，实现模糊系统的自学习和自适应功能的特性。基于模糊神经网络的自适应控制方案可以在线学习和调整规则参数及隶属函数参数，其控制规则表是根据现场实际控制数据并通过计算得出的，再将之应用于现场的控制中，具有很高的适应性，并保证闭环控制系统的稳定性，有高度鲁棒性，适用于非线性、时变、时滞系统的控制。因此，模糊神经网络控制的应用能很好的克服这种缺陷，适合于中和过程的非线性变化过程。
本文的研究工作主要包括以下内容：
(1)综述了智能控制的研究发展，介绍了模糊控制技术和神经网络控制技术的各自优缺点，以及神经网络控制器的典型结构和几种常用的神经网络，自适应控制的特点及其应用。
(2)分析了中和过程控制的非线性时延特性和控制难度。介绍在实际应用中一般采用的控制方法：非线性增益补偿PID控制和前馈控制及其控制的效果，分析了存在的问题和困难。
(3)根据模糊系统和神经网络的优缺点，提出了用神经网络来构造模糊系统，一个基于模糊神经网络控制算法。设计了模糊神经网络控制器，以及把基于模糊神经网络的自适应控制应用到中和过程，建模并仿真。
(4)提出了基于PLC的pH中和过程自动控制系统，并把基于模糊神经网络自适应控制算法应用到某电化厂的pH中和过程自动控制系统中，最后分析了基于模糊神经网络自适应控制算法在中和过程的应用结果，以及PLC在pH中和自
摘要
动控制系统中的应用结果。
本文最后对全文所做的工作进行了总结，并提出了今后进一步研究所需要做
的工作。
Neutralization process is a typical nonlinear process with pure time delay. This process can't be controlled effectively with conventional linear control method. Feed back control base on Nonlinear PID with self-adjust parameter and feed forward control base on experience model are employed in the practical application. The realization of this control system and its result are introduced in this paper. Because the process of the reaction is complicated, it's difficult to model for it. The Nonlinear PID can't fit the change of the reaction. The feed forward control base on model also can't fit the change of pH value and flux of the original water. So the control method can't get a good result.
In terms of the above considerations, this thesis proposes a new nonlinear adaptive control method based on fuzzy neural network to realize optimization control of pH neutralization process. Fuzzy neural network not only has the characteristic of fuzzy system, but also has the characteristic of neural network. The nonlinear adaptive control method based on fuzzy neural network can learn and adjust rule parameter and subjection parameter on line. And the control rule list is calculated according to control data on the spot, which is also applied to the control on the spot. Therefore, it has good adaptive and robust and also can ensure the stability of closed loop control system, which is fit to handle both the nonlinearities and time-varying characteristics of the pH process.
The research work of this thesis consists of the following major parts:
Firstly, this thesis summarizes the development of intelligent control and also introduces the advantage and disadvantage of fuzzy control technique and neural control technique respectively. The characteristic and application of adaptive control are also presented in this thesis.
Secondly, the traditional control for pH process ?the compensation of nonlinearities plus PID control is presented in this thesis.
Thirdly, the fuzzy neural network controller is designed and the adaptive control method based on fuzzy neural network is proposed according to the advantage and disadvantage of fuzzy system and neural network in this thesis. And the adaptive control method based on fuzzy neural network is applied to pH process.
Forth, Aimed at the process of pH value control of wastewater treatment in chemical factory, a PLC-based automatic control system of pH value of wastewater treatment is developed. In this thesis operation principle of this system, hardware configuration, and software configuration are all discussed in detail.
At last, a summary of the paper is given and further research interests are also proposed.
引文
[1] Mamdani, E. H. Application of Fuzzy Algorithms for Control of Simple Dynamic Plant. IEEE Proc. Control and Science, 1974, 121(12): 1585-1588.
[2] Harald Wegmann. Fazzy-logik. Erste Anwendung mit SIMATIC S5. Atp.3(1993): 167-171.
[3] Astrom, K. J. and Anton, J.J. Expert Control. Proc. Of the IFAC 9~(th) Triennial World Congress. Budapest. Hungary. 1984.2579-2584.
[4] 商建东，陈康宁 专家智能pH控制器及应用，化工自动化及仪表，1998，25(3)：46-48．
[5] Psaltis, D. A. et al.. Neural Controllers. Proc. IEEE ICNN, San Diego, 1987, 551-558.
[6] Palm, R.. Sliding Mode Fuzzy Control. Proc. FUZZ-IEEE, San Diego, 1992, 519-526.
[7] Qin, X. F. et al. A Robust Self-Tuning Control Algorithm with Neural Network. IFAC-YAC, Beijing, 1995, 694-698.
[8] 费敏锐，陈伯时 专家模糊控制方法，控制与决策，1996，11(2)：256-260．
[9] Kuo, Y. H., Kao, C. Z. and Chan, J.J. A Fuzzy Neural Network Model and Its
[10] 张曾科 模糊数学在自动化技术中的应用，清华大学出版，1997，7(1)：131-133．
[11] Astrom K S. Intelligent Control IEEE Control Systems Magazine, 1991, 1: 39-40
[12] Takagi T., Sugeno M. Fuzzy Identification of Systems and Its Application to Modeling and control. IEEE Trans. On, Syst, Man Cybern, 1985. 15: 116-132
[13] Procyk T J et al. A Linguistic Self-Organizing Process Controller. Automatica, 1979, 15:15-30
[14] Lee C C. Fuzzy Logic in Control Systems: Fuzzy Logic. Controller-part Ⅰ, partⅡ. IEEE Trans. On Syst, Man, Cybern, 1990, 20(2): 404-435
[15] 孙增圻 智能控制理论与技术。北京：清华大学出版社，南宁：广西科学技术出版社，1997．172．191
[16] Antsaklis P J. Neural Networks in Control Systems. IEEE Control System Magazine (Special Section on Neural Networks for Systems and Control), 1990,3-5
[17] 焦李成 神经网络的应用与实现，西安：西安电子科技大学出版社，1993，264-287
[18] Ribeiro B M, Correia A D. Lime kiln simulation and control by neural networks. Neural networks for chemical engineers. Elsevier Science B. V. Amsterdam, Netherlands, 1995.163-192.
[19] 汤同奎，邵惠鹤 神经网络在控制中的应用，江苏石油化工学院学报，1998(3)，vol 10(1)：45-49．
[20] 舒迪前 预测控制系统及其应用 机械工业出版社 1995年
[21] 王献忠，杜维 模型预测控制发展概况 自动化与仪器仪表，1999，4(总84)：4-9．
[22] 席裕庚 预测控制[M]，北京：国防工业出版社，1993，12，38．
[23] 陈立定 具有预测控制功能的pH控制系统，自动化仪表，22(1)：36-37，48．
[24] 贾平，麻红昭，俞蒙槐 基于神经网络的多步预测控制算法及其在pH值控制中的应用，信息与控制，1999(8) vol 28 (增刊)．
[25] 陆燕，杜继宏，梁循 人工神经网络在控制中的应用，电子技术应用，1996，No．2：4-7．
[26] Berenji H R, Khedkar P. Learning and tuning fuzzy logic controllers through reinforcements.
IEEE Trans. on Neural Networks, 1992, 3(5): 724-740.
[27] 汤同奎，邵惠鹤 神经网络在控制中的应用，江苏石油化工学院学报，1998(3)，vol 10(1)：45-49.
[28] Philip D Wasserman. Neural Computing Theory and Practice. New York, Van Nostrand Reinhold. 1989. 124-129.
[29] F. G. Shinskey. Process-Control System. Second Edition. McGRAW-HILL BOOK COMPANY(New York 1979).
[30] Gustafsson T. K. and K. V. Waller, Dynamic modeling and reaction invariant control of pH. Chem. Engng Sci, 1983, 38: 389-398.
[31] F. G. Shinskey. PH and pION Control in Process and Waste Streams. The Foxboro Company John Wiley & Sons. 1973.
[32] 王骥程，祝和云．化工过程控制工程(第二版). 化学工业出版社，1991．
[33] Culter C R, Ramaker B L Dynamic Matrix Control---A Cmoputer Control Algorithm. Proc, JACC. San Francisco, 1980.
[34] Clarke D W, Mohtadi C Tuffs P S. Generalized Predictive Control. pt. 1. The Basic Algorithm, pt. 2. Extensions and Interpreations. Automatica. 1987, 23(2): 137-160.
[35] 王锦标，方崇智．过程计算机控制．清华大学出版社，1992．
[36] 金以慧．过程控制. 清华大学出版社，1993．
[37] 周琼，汪洋，褚健 自动化理论、技术与应用(第四卷) 浙江大学出版社 1997，453-456
[38] 韩京清。非线性PID控制器。自动化报，1994(4)：487-490．
[39] A. Haalman, Adjusting Controllers for a Deadtime Process, Control Engineering, Vol. 12, July-65, 1965, 71-73
[40] 金晓明，荣冈，王骥程．自适应模糊控制的新进展．信息与控制，1996，25(4)：217～223
[41] Kong S G, Kosko B. Adaptive fuzzy systems for backing up a truck-and-trailer. IEEE Trans. On Neural Networks, 1992, 3(2): 211～233
[42] Chen Y C, Teng C C. A model refence control structure using a fuzzy neural network. Fuzzy Sets and Systems, 1995, 73: 291～312
[43] J. G. Ziegler and N. B. Nichols, Optimun Setting for Automatic Controller Trans. ASME, Vol. 64, No. 11, 1942, 759-768.
[44] D. Nauck and R. Kruse, A neural fuzzy controller learning by fuzzy error propagation, Proc. NAFIPS 1992, Puerto Vallarta, Mexico, Vol. Ⅱ, 1992.
[45] Sing C H, Postlethwaite B. pH control: handling nonlinearity and deadtime with fuzzy relational mode-based control[J]. IEE Proc Control Theory Appl, 1997, 144(3): 263-268.
[46] Rouhani R, Mehra R. K Model Algorithmic Control(MAC), Basic Theoretical Properties. Automatica, 1982, 18(4): 401-414.
[47] Culter C R, Ramaker B L Dynamic Matrix Control---A Cmoputer Control Algorithm. Proc, JACC. San Francisco, 1980.
[48] Clarke D W, Mohtadi C Tuffs P S. Generalized Predictive Control. pt.1. The Basic Algorithm, pt. 2. Extensions and Interpreations. Automatica. 1987, 23(2): 137-160.
[49] Garcia C E, Morari M. Internal Model Control. 1. A Unifying Review and Some New Results. Ind. Eng. Chem. Pro. Des. Dev. 1982, 21(2): 308-323.
[50] X. Z. Gao, X.. M. Gao and S. J. Ovaska, Trajectory Control Based on a Modified Elman Neural
Network. IEEE Trans. on Neural Network, 1997: 2505-2510.
[51] D. P. Kwork, P. Tam and K. Zhou Control of a pH Neutralization Process Using a Modified Elman Neural Net, IEEE 1994, 71-76.
[52] 朱群雄，孙峰 RNN神经网络的应用研究，北京化工大学学报，1998，25(1)：86-90
[53] KUMPATI S. NARENDRA and KANNAN PARTHASARATHY Identification and Control of Dynamical Using Neural Networks IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL 1. NO 1. MARCH 1990: 4-27.
[54] Sastry P S, Santharam G, Unnikrishnan K P. Memory neuron networks for indentification and control of dynamical systems. IEEE Trans. On Neural Networks, 1994, 5(2): 306-319.
[55] Pham D T and Liu X. Dynamic System Modelling using Partially Recurrent Neural Networks. Journal of System Engineering, 1992, 2: 90-97.
[56] 何玉彬 李新忠 神经网络控制技术及其应用 科学出版社 2000年
[57] GE Cimplicity Automation Software使用手册
[58] GE VersaMax系统设计使用手册 1999
[59] Jianbin Hao, Shaohua Tan, Joos Vandewalle One step ahead predictive control of nonlinear systems by neural networks. Proceedings of 1993 International joint conference on neural networks: 2761-2764
[60] Hall R. C. and D. E. Seborg, Modeling multiloop control. Proc. Am. Control Conf., Pitsburgh, 1989: 1822-1827.
[61] Wei, jianping Li, huade Sun, ming Sun, shaoyuan Modeling for a Control Industrial Object Based on Recurrent Neural Network. Proceedings of the 3~(rd) world congress on intellingent control and automation, June 28-July 2,2000, Hefei, R. R. China.
[62] Cheng Y, Karjala T W and Himmelblau D M. Identification of Nonlinear Dynamic Process with Unknown and Variable Dead Time Using an Internal Recurrent Neural Network. Ind. Eng. Chem. Res, 1995, 34: 1753-1742.
[63] 郑宏兴，姚纪欢，张成 MATLAB 5.x 工具箱使用技巧与实例，华中科技大学出版社，2000．
[64] S. I. Mistry and S. S. Nair Iderntification and control experiments using neural designs, IEEE Control Systems Magazine, vol. 14, pp. 48-57, June 1994.
[65] 冯纯伯，史维，王明君 自适应控制 电子工业出版社，1986
[2] Harald Wegmann. Fazzy-logik. Erste Anwendung mit SIMATIC S5. Atp.3(1993): 167-171.
[3] Astrom, K. J. and Anton, J.J. Expert Control. Proc. Of the IFAC 9~(th) Triennial World Congress. Budapest. Hungary. 1984.2579-2584.
[4] 商建东，陈康宁 专家智能pH控制器及应用，化工自动化及仪表，1998，25(3)：46-48．
[5] Psaltis, D. A. et al.. Neural Controllers. Proc. IEEE ICNN, San Diego, 1987, 551-558.
[6] Palm, R.. Sliding Mode Fuzzy Control. Proc. FUZZ-IEEE, San Diego, 1992, 519-526.
[7] Qin, X. F. et al. A Robust Self-Tuning Control Algorithm with Neural Network. IFAC-YAC, Beijing, 1995, 694-698.
[8] 费敏锐，陈伯时 专家模糊控制方法，控制与决策，1996，11(2)：256-260．
[9] Kuo, Y. H., Kao, C. Z. and Chan, J.J. A Fuzzy Neural Network Model and Its
[10] 张曾科 模糊数学在自动化技术中的应用，清华大学出版，1997，7(1)：131-133．
[11] Astrom K S. Intelligent Control IEEE Control Systems Magazine, 1991, 1: 39-40
[12] Takagi T., Sugeno M. Fuzzy Identification of Systems and Its Application to Modeling and control. IEEE Trans. On, Syst, Man Cybern, 1985. 15: 116-132
[13] Procyk T J et al. A Linguistic Self-Organizing Process Controller. Automatica, 1979, 15:15-30
[14] Lee C C. Fuzzy Logic in Control Systems: Fuzzy Logic. Controller-part Ⅰ, partⅡ. IEEE Trans. On Syst, Man, Cybern, 1990, 20(2): 404-435
[15] 孙增圻 智能控制理论与技术。北京：清华大学出版社，南宁：广西科学技术出版社，1997．172．191
[16] Antsaklis P J. Neural Networks in Control Systems. IEEE Control System Magazine (Special Section on Neural Networks for Systems and Control), 1990,3-5
[17] 焦李成 神经网络的应用与实现，西安：西安电子科技大学出版社，1993，264-287
[18] Ribeiro B M, Correia A D. Lime kiln simulation and control by neural networks. Neural networks for chemical engineers. Elsevier Science B. V. Amsterdam, Netherlands, 1995.163-192.
[19] 汤同奎，邵惠鹤 神经网络在控制中的应用，江苏石油化工学院学报，1998(3)，vol 10(1)：45-49．
[20] 舒迪前 预测控制系统及其应用 机械工业出版社 1995年
[21] 王献忠，杜维 模型预测控制发展概况 自动化与仪器仪表，1999，4(总84)：4-9．
[22] 席裕庚 预测控制[M]，北京：国防工业出版社，1993，12，38．
[23] 陈立定 具有预测控制功能的pH控制系统，自动化仪表，22(1)：36-37，48．
[24] 贾平，麻红昭，俞蒙槐 基于神经网络的多步预测控制算法及其在pH值控制中的应用，信息与控制，1999(8) vol 28 (增刊)．
[25] 陆燕，杜继宏，梁循 人工神经网络在控制中的应用，电子技术应用，1996，No．2：4-7．
[26] Berenji H R, Khedkar P. Learning and tuning fuzzy logic controllers through reinforcements.
IEEE Trans. on Neural Networks, 1992, 3(5): 724-740.
[27] 汤同奎，邵惠鹤 神经网络在控制中的应用，江苏石油化工学院学报，1998(3)，vol 10(1)：45-49.
[28] Philip D Wasserman. Neural Computing Theory and Practice. New York, Van Nostrand Reinhold. 1989. 124-129.
[29] F. G. Shinskey. Process-Control System. Second Edition. McGRAW-HILL BOOK COMPANY(New York 1979).
[30] Gustafsson T. K. and K. V. Waller, Dynamic modeling and reaction invariant control of pH. Chem. Engng Sci, 1983, 38: 389-398.
[31] F. G. Shinskey. PH and pION Control in Process and Waste Streams. The Foxboro Company John Wiley & Sons. 1973.
[32] 王骥程，祝和云．化工过程控制工程(第二版). 化学工业出版社，1991．
[33] Culter C R, Ramaker B L Dynamic Matrix Control---A Cmoputer Control Algorithm. Proc, JACC. San Francisco, 1980.
[34] Clarke D W, Mohtadi C Tuffs P S. Generalized Predictive Control. pt. 1. The Basic Algorithm, pt. 2. Extensions and Interpreations. Automatica. 1987, 23(2): 137-160.
[35] 王锦标，方崇智．过程计算机控制．清华大学出版社，1992．
[36] 金以慧．过程控制. 清华大学出版社，1993．
[37] 周琼，汪洋，褚健 自动化理论、技术与应用(第四卷) 浙江大学出版社 1997，453-456
[38] 韩京清。非线性PID控制器。自动化报，1994(4)：487-490．
[39] A. Haalman, Adjusting Controllers for a Deadtime Process, Control Engineering, Vol. 12, July-65, 1965, 71-73
[40] 金晓明，荣冈，王骥程．自适应模糊控制的新进展．信息与控制，1996，25(4)：217～223
[41] Kong S G, Kosko B. Adaptive fuzzy systems for backing up a truck-and-trailer. IEEE Trans. On Neural Networks, 1992, 3(2): 211～233
[42] Chen Y C, Teng C C. A model refence control structure using a fuzzy neural network. Fuzzy Sets and Systems, 1995, 73: 291～312
[43] J. G. Ziegler and N. B. Nichols, Optimun Setting for Automatic Controller Trans. ASME, Vol. 64, No. 11, 1942, 759-768.
[44] D. Nauck and R. Kruse, A neural fuzzy controller learning by fuzzy error propagation, Proc. NAFIPS 1992, Puerto Vallarta, Mexico, Vol. Ⅱ, 1992.
[45] Sing C H, Postlethwaite B. pH control: handling nonlinearity and deadtime with fuzzy relational mode-based control[J]. IEE Proc Control Theory Appl, 1997, 144(3): 263-268.
[46] Rouhani R, Mehra R. K Model Algorithmic Control(MAC), Basic Theoretical Properties. Automatica, 1982, 18(4): 401-414.
[47] Culter C R, Ramaker B L Dynamic Matrix Control---A Cmoputer Control Algorithm. Proc, JACC. San Francisco, 1980.
[48] Clarke D W, Mohtadi C Tuffs P S. Generalized Predictive Control. pt.1. The Basic Algorithm, pt. 2. Extensions and Interpreations. Automatica. 1987, 23(2): 137-160.
[49] Garcia C E, Morari M. Internal Model Control. 1. A Unifying Review and Some New Results. Ind. Eng. Chem. Pro. Des. Dev. 1982, 21(2): 308-323.
[50] X. Z. Gao, X.. M. Gao and S. J. Ovaska, Trajectory Control Based on a Modified Elman Neural
Network. IEEE Trans. on Neural Network, 1997: 2505-2510.
[51] D. P. Kwork, P. Tam and K. Zhou Control of a pH Neutralization Process Using a Modified Elman Neural Net, IEEE 1994, 71-76.
[52] 朱群雄，孙峰 RNN神经网络的应用研究，北京化工大学学报，1998，25(1)：86-90
[53] KUMPATI S. NARENDRA and KANNAN PARTHASARATHY Identification and Control of Dynamical Using Neural Networks IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL 1. NO 1. MARCH 1990: 4-27.
[54] Sastry P S, Santharam G, Unnikrishnan K P. Memory neuron networks for indentification and control of dynamical systems. IEEE Trans. On Neural Networks, 1994, 5(2): 306-319.
[55] Pham D T and Liu X. Dynamic System Modelling using Partially Recurrent Neural Networks. Journal of System Engineering, 1992, 2: 90-97.
[56] 何玉彬 李新忠 神经网络控制技术及其应用 科学出版社 2000年
[57] GE Cimplicity Automation Software使用手册
[58] GE VersaMax系统设计使用手册 1999
[59] Jianbin Hao, Shaohua Tan, Joos Vandewalle One step ahead predictive control of nonlinear systems by neural networks. Proceedings of 1993 International joint conference on neural networks: 2761-2764
[60] Hall R. C. and D. E. Seborg, Modeling multiloop control. Proc. Am. Control Conf., Pitsburgh, 1989: 1822-1827.
[61] Wei, jianping Li, huade Sun, ming Sun, shaoyuan Modeling for a Control Industrial Object Based on Recurrent Neural Network. Proceedings of the 3~(rd) world congress on intellingent control and automation, June 28-July 2,2000, Hefei, R. R. China.
[62] Cheng Y, Karjala T W and Himmelblau D M. Identification of Nonlinear Dynamic Process with Unknown and Variable Dead Time Using an Internal Recurrent Neural Network. Ind. Eng. Chem. Res, 1995, 34: 1753-1742.
[63] 郑宏兴，姚纪欢，张成 MATLAB 5.x 工具箱使用技巧与实例，华中科技大学出版社，2000．
[64] S. I. Mistry and S. S. Nair Iderntification and control experiments using neural designs, IEEE Control Systems Magazine, vol. 14, pp. 48-57, June 1994.
[65] 冯纯伯，史维，王明君 自适应控制 电子工业出版社，1986