臺灣博碩士論文加值系統

本篇論文提出分別以進化演算法結合最速下降法以及線性規劃結合最速下降法為基礎之兩種用於未知非線性多輸入多輸出系統之適應性多變數PID容錯控制器。對於具有單調非線性的非線性系統，一個使用延伸型卡爾曼濾波器來調整權值的類神經網路被用來線上估測最精確的等價系統模型，之後使用最速下降法找出一個最佳控制，進化演算法或線性規劃則用來線上調整PID容錯控制器。

Based on the evolutionary programming (EP) combined with the steepest descent method and the linear programming (LP) combined with the steepest descent method respectively, two adaptive multivariable PID fault tolerant controller schemes for the unknown nonlinear multi-input multi-output (MIMO) systems are proposed in this thesis. In the case of nonlinear dynamic system, and for monotonic nonlinearity, a neural network adapted with the extended Kalman filter is created to estimate the most accurate equivalent system model online, then the EP or LP is used for on-line tuning the well-performed PID fault tolerant controller after using the steepest descent method to find an optimal control.

Chinese Abstract I
Abstract II
Acknowledgements III
List of Contents IV
List of Tables VI
List of Figures VII

Content
Chapter 1 Introduction 1-1
Chapter 2 EP-Based Adaptive PID Fault Tolerant Controllers for Unknown Nonlinear MIMO Systems 2-1
2.1 Introduction 2-2
2.2 Adaptive Neural Network Model 2-3
2.2.1 Neural Network Model 2-3
2.2.2 Neural Network Identification 2-4
2.2.3 Extended Kalman Filter Algorithm 2-7
2.3 Evolutionary Programming 2-9
2.3.1 Quasi-random Sequences (QRS) 2-10
2.4 EP-Based Adaptive Algorithm for PID Fault Tolerant Controllers 2-15
2.4.1 Control Structure and Convergence Analysis with Steepest Descent Method 2-16
2.4.2 Gains Tuning for PID Controller with EP 2-18
2.5 An Illustrative Example 2-23
Chapter 3 LP-based Adaptive PID Fault Tolerant Controllers for Unknown Nonlinear MIMO Systems 3-1
3.1 Linear Programming 3-2
3.2 Gains Tuning for PID Controller with Linear Programming 3-3
3.3 An Illustrative Examples 3-4
Chapter 4 Conclusion 4-1
Reference R-1

[1] M. Blanke, M. Kinnaert, J. Lunze, M. Staroswiecki, and J. Schrder. Diagnosis and Fault-Tolerant Control. Germany, Berlin: Springer-Verlag, 2004.
[2] R. J. Patton, P. M. Frank and R. Clark. Fault Diagnosis in Dynamic Systems: Theory and Application. Englewood Cliffs, NJ: Prentice-Hall, 1989.
[3] B. M. Badreddine and F. Lin, “Adaptive PID controller for stable/unstable linear and non-linear systems,” Proceedings of the 2001 IEEE International Conference on Control Applications, pp.1031. 5-7, Sept. 2001
[4] S. Chand, “on-line, self-monitoring tuner for proportional integral derivative controllers,” Proceedings of the 30th IEEE Conference on Decision and Control, vol.2, pp.1905–1916, 11-13 Dec. 1991.
[5] J. Litt, “An expert system to perform on-line controller tuning,” IEEE Control Systems Magazine, vol.11, issue 3, pp.18-23, April 1991.
[6] K Takao, T. Yamamoto and T. Hinamoto, “Memory-based on-line tuning of PID controllers for nonlinear systems,” Proceedings of the 2004 American Control Conference, vol. 5, pp. 4016-4021, 30 June-2 July, 2004.
[7] Y. Ding, “A nonlinear PID controller based on fuzzy-tuned immune feedback law,” Proceedings of the 3rd World Congress on Intelligent Control and Automation, vol. 3, pp.1576-1580, 28 June-2 July 2000.
[8] V. Kecman, Learning and Soft Computing. Cambridge, Massachusetts: MIT Press, 2001.
[9] S. Haykin. Neural Networks. Second Edition. New Jersey: Prentice-Hall, 1999.
[10] O. Nelles. Nonlinear System Identification. Germany, Berlin: Springer-Verlag, 2000.
[11] K. S. Narendra and K. Parthasarathy. Identification and control of dynamical systems using neural networks. IEEE Trans. on Neural Networks, vol. 1, no.1, pp. 4-27, 1990.
[12] K. S. Narendra, and A. M. Annaswamy. Stable adaptive systems. Engelwood Cliffs, NJ: Prentice Hall, 1989.
[13] S. Haykin. Kalman Filtering and Neural Networks. Wiley-Interscience, 2001.
[14] D. L. Yu, T. K. Chang and D. W. Yu, “Fault tolerant control of multivariable processes using auto-tuning PID controller,” IEEE Trans. on Systems, Man and Cybernetics, Part B, vol. 35, issue 1, pp. 32-43, Feb. 2005.
[15] Y. Iiguni, H. Sakai and H. Tokumaru, “A real-time learning algorithm for a multilayered neural network based on the extended Kalman filter,” IEEE Trans. on Signal Processing, vol. 40, issue 4 pp. 959-966, 1992.
[16] L. Ljung and T. Sderstrm, Theory and Practice of Recursive Identification. Cambridge, MA: MIT Press, 1983.
[17]Z. Michalewicz. Genetic Algorithms + Data Structures = Evolution Programs. Germany, Berlin: Springer-Verlag, 1996.
[18] J. H. Halton, “On the efficiency of certain quasi-random sequences of points in evaluating multidimensional integrals,” Numerische Mathematik, vol. 2, pp. 84-90 with Corrigenda on p. 196, 1960.
[19] J. C. Van Der Corput, “Verteilungsfunktionen,” Proc. Kon Akad. Wet., Amsterdam, vol. 38, pp. A13-A21, pp. 1058-1066, 1935.
[20] T. Rommel, Neural networks based adaptive control, Report no. 97-30, The University of Auckland, New Zealand, Diploma Thesis, 1997.
[21] V. Kecman and T. Rommel, Performance of neural networks based adaptive backthrough control, Proc. ICONIPANZIIS/ANNES, Dunedin, New Zealand, 1997.
[22] W. D. Chang, R.C. Hwang, and J.G. Hsieh, “A multivariable on-line adaptive PID controller using auto-tuning neurons,” Engineering Applications of Artificial Intelligence, vol. 16, pp. 57-63, 2003.
[23] R. Salman, “Neural networks of adaptive inverse control systems,” Applied Mathematics and Computation 163 pp. 931–939, 2005.
[24] G. G. Yen and L. W. Ho, “Online multiple-model-based fault diagnosis and accommodation,” IEEE Trans. on Industrial Electronics, vol. 50, issue 2, pp. 296-312, 2003.
[25] G. B. Dantzig, M. N, Thapa. Linear Programming 1: Introduction. Germany, Berlin: Springer-Verlag, 1997.