臺灣博碩士論文加值系統

化工程序中多為複雜且常常呈現時變非線性的型態，而且對於外在影響有極高的敏感度，在正常操作狀態下稍有一些變動則產物將會有很大的變異性，所以在作控制器設計時必須格外重視即時系統的掌控特性。本研究即在發展出一套利用瞬間線性化類神經網路的模式，以改善傳統PID控制器於非線性系統的控制策略。在模式建立方面，採用Levenberg-Marquardt演算法來學習類神經網路ARX模式，以代表系統輸入輸出之行為。當其應用於控制迴路下時，基於對控制策略簡化及加速運算的實際需求，使用瞬間線性化類神經網路的技術抽取出線性化模式，以提供控制參數的即時運算。在控制器設計部分，運用GMV(General Minimum Variance)最小化目標函數的概念，及存在限制條件下找尋即時最適的PID控制器調諧參數。此控制策略與一般的gain scheduling同樣擁有各操作區段的線性化動態預測模式，但根據非線性類神經網路結構所推算出的控制參數擁有無限區段的結果，所以在系統動態擷取上具有更高的可信度。為了應用在實際的操作過程，本研究也提出了動量過濾、更新控制參數的參考準則及控制作動步伐的修正法，期能在實際運用上有大幅的改善。故此研究有兩大特點：(1)簡化複雜的計算模式，(2)將具有非線性學習特性的類神經網路融入控制器的設計。最後將此所發展的控制架構應用於非線性差分方程式、pH中和反應槽與批次反應器之模擬控制，以驗證此方法的效用。

The inherent time-varying nonlinearity and complexity usually exist in chemical processes. The design process would be significantly deviated from the normal operating condition when only a slight disturbance occurs. Accordingly, the design of control structure should be properly adapted based on the instantaneous state. In this research, an improved conventional PID control scheme using linearization through a specified neural network is developed to control nonlinear processes. An input-driven output neural network ARX model trained by Levenberg-Marquardt algorithm is introduced in the model design. The linearization of the neural network model will be proposed to extract the linear model for updating the controller parameters. In the scheme of the optimal tuning PID controller, the concept of general minimum variance is presented and a constrained criterion is also considered. Like gain scheduling, the control system of the proposed method at each time interval is chosen from a set of predefined linearized dynamic model. Unlike gain scheduling control, the control parameters based on the neural network model have an infinite scheduling resolution when using a neural network for updating parameters. To apply the proposed method to most of the practical application problems, several variations of the proposed method, including the momentum filter, the updating criterion and the step size of the control action, are presented to provide significant improvement and make the proposed algorithm more practical. The proposed method has two advantages. First, less computation of linear adaptive control scheme is used. Second, the nonlinear characteristics of neural networks can be incorporated into the control design. To demonstrate the potential applications of the proposed strategies, several problems, including batch reactors and pH neutralization processes, are applied.

摘要I
AbstractII
目錄IV
圖目錄VI
表目錄XIII
第一章 前言1
1.1自我調整PID控制器的演進1
1.1.1以線性模式為基礎的PID控制策略2
1.1.2以非線性模式為基礎的PID控制策略3
1.2研究動機7
第二章 自我調整PID控制器理論9
2.1可調PID控制器的演進9
2.2即時調整的修正Ziegler-Nichols法則12
2.3廣義的最小變異量控制與PID控制器結合理論16
2.3.1廣義的最小變異量控制理論(GMVC)16
2.3.2 GMVC轉換為PID控制器之策略20
2.4最適化參數調諧的延伸22
第三章 利用線性化類神經網路模式與廣義的最小變異量控制策略之自我調整PID控制器設計24
3.1類神經網路模式建立與學習25
3.2線性化類神經網路模式29
3.2.1瞬間線性化類神經網路模式29
3.2.2瞬間線性化類神經網路模式與線性模式的比較34
3.3以瞬間線性化模式預測模式為基礎的自我調諧PID控制器39
3.4在限制條件下以瞬間線性化模式之自我調整PID控制器設計43
3.5以瞬間線性化模式預測誤差修正控制作動的經驗式法則45
第四章 模擬與比較測試54
4.1非線性數學模式的控制54
4.1.1類神經網路模式的學習54
4.1.2各種自我調整PID控制的比較及說明57
4.1.2.1修正的即時調整Ziegler-Nichols法則57
4.1.2.2傳統GMVC的線上PID即時調諧策略58
4.1.2.3線性模式與GMVC結合理論進行線上PID即時調諧63
4.1.2.4瞬間線性化類神經網路模式與GMVC理論結合進行線上PID即時調諧66
4.2連續程序非線性化工的控制72
4.2.1程序說明72
4.2.2類神經網路模式建立74
4.2.3設定點改變測試77
4.2.4未知擾動與測試83
4.2.5模式預測誤差測試87
4.3批次反應器的控制90
4.3.1程序說明90
4.3.2類神經網路模式建立93
4.3.3雜訊干擾下之設定點改變測試95
4.3.3.1無雜訊干擾下之設定點改變測試95
4.3.3.2有雜訊干擾下之設定點改變測試98
第五章 結論與未來工作103
5.1結論103
5.2未來工作104
5.2.1預測控制104
5.2.2多變數控制105
符號說明107
參考文獻110

1.Ahmed, M. S. and Tasadduq, I. A., “Neural-net Controller for Nonlinear Plants：Design Approach Linearisation,” IEE Proc-Control Theory Appl., 5, 315 (1994).2.Ali, E., “Control of Nonlinear Chemical Processes Using Adaptive Proportional-Integral Algorithms,” Ind. Eng. Chem. Res., 39, 1980 (2000).3.Ali, E., “pH Control Using PI Control Algorithm with Automatic Tuning Method,” Trans. IChemE., 79, 611 (2001).4.Arthur, J., “A Nonlinear PI(D) Controller,” Canadian J. Chem. Eng., 67, 485 (1989).5.Astrom, K. J., “Theory and Applications of Adaptive Control — A Survey,” Automatica, 19, 471 (1983).6.Astrom, K. J. and McAvoy, T. J., “Intelligent Control,” J. Proc. Contr., 2, 115 (1992).7.Astrom, K. J. and Wittenmark, B., “Online Self-Tuning Regulators,” Automatica, 9, 185 (1973).8.Bhat, N. and McAvoy, T. J., “Use of Neural Nets for Dynamic Modeling and Control of Chemical Process Systems,” Compu. Chem. Engng, 14, 573 (1990).9.Bobal, V., Bohm, Josef and Prokop, R., “Practical Aspects of Self-tuning Controllers,” Int. J. Adapt. Contr. Proc., 13, 671 (1999).10.Chen, C. T., “Direct Adaptive Control of Chemical Process Systems,” Ind. Eng. Chem. Res., 40, 4121 (2001).11.Chen, C. T and Peng S. T., “A Simple Adaptive Control Strategy for Temperature Trajectory Tracking in Batch Processes,” Canadian J. Chem. Eng., 76, 1118 (1998).12.Chen, J., “Systematic Derivations of Model Predictive Control Based on Artificial Neural Network,” Chem. Eng. Comm., 164, 35 (1998).13.Chen, Q. and Weigand, W. A., “Dynamic Optimization of Nonlinear Process by Combining Neural Net Model with UDMC,” AIChE J., 14, 1488 (1994).14.Clarke, D and Gawthrop, P., “A Self-Tuning Controller,” Proc. IEEE, 122, 929 (1979).15.Demetri P.; Athanasios S. and Alan A. Y., “A Multilayered Neural Network Controller,” IEEE Contr. Sys., 8(2), 17 (1988).16.Dong J. W. and Brosilow B. C., “Design of Robust Multivariable PID Controllers via Internal Model Control (IMC),” Proc. of the American Contr. Conf., 3380 (1997).17.Edgar, T. F. and Himmelblau, D. M., Optimization of Chemical Processes, McGraw-Hill International Editions (1989).18.Gao, F. R, Wang, F. L. and Li, M. Z., “Neural Network-Based Optimal Iterative Controller for Nonlinear Processes,” Canadian J. Chem. Eng., 78, 363 (2000).19.Gheorghe P. and Vasile P., “An Improved Inverse Neural Control Structure for Nonlinear Dynamic Systems,” Electronics, Circuits and Systems, 1999. Proceedings of ICECS ''99. The 6th IEEE International Conference, 2, 985 (1999).20.Gong, X. F.; Gao, J. C. and Zhou, C. H., “Extension of IMC Tuning to Improve Controller Performance,” IEEE Contr. Sys., 3, 1770 (1996).21.Hagan, M. T.; Demuth, H. B. and Beale, M., Neural Network Design, PWS Publishing Company (1996).22.Ho, H. L., Rad, A. B., Chan, C. C. and Wong, Y. K. “Comparative Studies of There Adaptive Controllers,” ISA Trans, 38, 43. (1999).23.Hornik, K.; Stinchcombe, M. and White, H., “Mutliplayer Feedforward Neural Networks are Universal Approximators,” Neural Networks, 2, 359 (1989).24.Hornik, K.; Stinchcombe, M. and White, H., “Universal Approximation of an Unknown Mapping and its Derivatives Using Mutliplayer Feedforward Networks,” Neural Networks, 3, 551 (1990).25.Irwin, G. W., Warwick, K. and Hunt, K. J., Neural Network Applications in Control, The Institution of Electrical Engineering, London (1995).26.Kenzo F.; Toru Y. and Masahiro K., “Self-Tuning PID Control of a Polybutene Process,” IFAC Advanced Control of Chemical Processes, 103 (1997).27.Kravaris, C. and C. B. Chung, “Nonlinear State Feedback Synthesis by Global Input/Output Linearization,” AICHE J., 33, 592 (1987).28.Kumpati, S. and Kannan, P., “Identification and Control of Dynamic Systems Using Neural Networks,” IEEE Trans. on Neural Networks, 1, 4 (1990).29.Lakshminarayanan, S.; Shah, S. L. and Nandakumar, K., “Modeling and Control of Multivariable Process：Dynamic PLS Approach,” AIChE J., 43, 9, (1997).30.Landau, I. D., Lozano, R. and M'Saad, M., Adaptive Control, Springer-Verlag London Limited (1997).31.Liu, G. P. and Daley, S., “Optimal-Tuning Nonlinear PID Control of Hydraulic Systems,” Contr. Eng. Practice, 8, 1045 (2000).32.Lopez, A. M.; Murrill P. W. and Smith, C. L., “Controller Tuning Relationships Based on Integral Performance Criteria,” Instrumentation Technology, 57, 14 (1967)33.Luyben, W. L., Process Modeling, Simulation and Control for Chemical Engineers, McGraw-Hill International Editions (1990)34.Marzuki, K. and Rubiyah, Y., Neuro-Control and it's Applications Advances in Industrial Control, Springer-Verlag London Limited (1996).35.Nahas, E. P.; Henson, M. A. and Seborg, D. E., “Nonlinear Internal Model Control Strategy for Neural Network Models,” Computers chem. Engng, 12, 1039 (1992).36.Ng, T.C.T; Leung, F. H. F. and Tam, P. K. S., “A simple gain scheduled PID controller with stability consideration based on a grid-point concept,” Industrial Electronics, 1997. ISIE ''97., Proceedings of the IEEE International Symposium, 3, 1090 (1997).37.Norio, M., Masao, I., Kyoji, H., R. K. Wood, Hirofumi, H. and Mituyoshi, O., “Auto-Tuning PID Controller Based on Generalized Miniumum Variance Control for a PVC Reactor,” J. Chem. Eng. Jap., 31(4), 626 (1998).38.Psichogios, D. C. and Ungar, L. H., “Direct and in Direct Model Based Control Using Artificial Neural Networks,” Ind. Eng. Chem. Res., 30, 2564 (1991).39.Ravichandran, T., and Karray, F., “Knowledge Based Approch for Online Self-Tuning of PID-Control,” Proce. of the American Contr. Conf., 25 (2001).40.Rovira, A. A., Control Algorithm Tuning, and Adaptive Gain Tuning in Process Control, Ph. D. dissertation of Department of Chemical Engineering, Louisiana State University, Baton Rouge (1981).41.Ruano A. E. and Azevedo A B., “B-spline Neural Netwrok Assisted PID Autotuning,” Int. J. Adapt. Contr. Proc., 13, 291 (1999).42.Saiful, A. and Omatu, S., “Neuromorphic Self-tuning PID controller,” Proc. Of 1993 IEEE ICNN, San Francisco, 552 (1993).43.Sigeru, O., Takeshi, I. and Michifumi., Y., “Skill-Based PID Control by Using Neural Networks,” IEEE Contr. Sys., 2, 1972 (1998).44.Sigeru, O., Toru, F., and Michifumi, Y., “Neuro-PID Control for Inverted Single and Double Pendulums,” IEEE Contr. Sys., 4, 2685 (2000).45.Smith, C. A., and Corripio, A. B., Principles and Practice of Automatic Process Control, John Wiley & Sons, Inc (1997).46.Tan, Y. H., and Dang, X. J., “Generalised Nonlinear PID Controller Based on Neural NetWorks,” Information, Decision and Control, 1999. IDC 99. Proceedings. 1999, 519 (1999).47.Toru, Y., Toshitaka, O., and Sirish, L. S., “Design of a Multivariable Neural-Net Based PID Controller,” Neural Information Processing, 1999. Proceedings. ICONIP ''99. 6th International Conference, 3, 1051 (1999).48.Toru, Y., Yoshiyuki, K., Michifumi, Y., and Sigeru, O., “Stablization of Double Inverted Pendulum with Self-Tuning Neuro-PID,” Neural Networks, 2000. IJCNN 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference, 4, 345 (2000).49.Toshiharu, M., Atsushi, F., Munehiro, N., and Yoshihisa, Ishida, “Non-Linear PID Controller Using Neural Networks,” Neural Networks,1997., International Conference, 2, 811 (1997).50.Wang, B. Z., Lin, T. and Fan, Z., “Neural Network Based Online Self-Learning Adaptive PID Control,” Intelligent Control and Automation, 2000. Proceedings of the 3rd World Congress, 2, 908 (2000).51.Wright R. A.; Kravaris C. and Kazantzis N., “Model-Based Synthesis of Nonlinear PI and PID Controllers,” AICHE J., 47, 8, 1805 (2001).52.Yeong-Koo Yeo and Tae-In Kwon, “A Neural PID Controller for the pH Netralization Process,” Ind. Eng. Chem. Res., 978 (1999).