臺灣博碩士論文加值系統

摘要
SO-NFS (Self-Organization Neuro-Fuzzy System)為一具有自我建構與學習能力的智慧型系統，結合模糊理論與類神經網路的技術SO-NFS擁有與人類相似的推論與學習的能力。起初SO-NFS中並無任何的控制策略或控制法則，但依據所收集受控系統的部分資料SO-NFS經由學習的過程可以自動的建構出控制該系統的控制法則與合適的參數集合。假使受控系統為一未知系統，SO-NFS依然可以經由自我建構與學習來達到控制該未知系統的目的。SO-NFS的建構過程中可以分為兩個階段（1）結構學習及（2）參數學習；結構學習的目的在建立控制的法則庫，該控制的規則是以IF-THEN的型態來表示，在結構學習的過程中，叢集（clustering）的方法被運用來決定控制法則（即IF-THEN法則）的數目，以避免控制法則的數目太多而增加設計的困難與運算資源的浪費；參數學習則利用隨機最佳化演算法（Random Optimization algorithm），針對不同的應用問題設計適當的代價函數（cost function）來引導SO-NFS學習出適合的參數集合。SO-NFS所具備的法則庫自動建構與系統參數自我調整的能力，使的SO-NFS不只可以應用在線性系統的控制也可以應用在非線性系統的控制上。

Abstract
A cluster-based self-organization neuro-fuzzy system (SO-NFS) is proposed for control of unknown plants. The neuro-fuzzy system can learn its rule-based structure and parameters from input/output training data. There is no fuzzy IF-THEN rule in the system initially. The fuzzy control policy is constructed automatically during learning process when the system is simulated by input/output training data. There are two phases in the self-organization learning process, which are so-called structure learning and parameter learning. Using a cluster-based algorithm, the neuro-fuzzy system in its genesis can be generated to have its initial control policy (IF-THEN rules) for application by the stimulation of input/output training data. With the well-known random optimization (RO) method, the generated neuro-fuzzy system can learn its parameters for specific applications. The proposed approach for self-organization of system structure and self-adjustment of system parameter can be applied on problems of both linear and nonlinear control for unknown plants, and can levitate the problem of curse of dimensionality in the traditional fuzzy systems. The pseudo-error learning control is proposed for closed-loop control applications, and inverse learning control is performed as well.

CONTENTS
CHAPTER PAGE
1 INTRODUCTION………………………………………………………1
1.1 Feedback Control Systems……………………………………1
1.2 Problem Statement……………………………………………2
2 CRISP SET, FUZZY SET AND FUZZY LOGIC………………………5
2.1 Classical Sets…………………………………………………5
2.2 Fuzzy Sets………………………………………………………6
2.2.1 Membership Function…………………………………………6
2.2.2 Fuzzy Set………………………………………………………8
2.2.3 Support, Crossover Point and Fuzzy Singleton………8
2.2.4 Height, Normal, Subnormal, Convex and Fuzzy
Number…………………………………………………………9
2.2.5 Base Variable and Linguistic Variable…………………9
2.3 Theoretic Operations of Fuzzy Sets………………………10
2.3.1 Union, Intersection, Complement and Cartesian
Product………………………………………………………12
2.3.2 Fuzzy Relation and Sup-Star Composition………………13
2.3.3 Triangular Notms (t-norm) and Triangular Co-
Norms (t-conorm)………………………………14
3 FUZZY LOGIC CONTROLLER…………………………………………16
3.1 Introduction…………………………………………………16
3.2 Fuzzification Interface……………………………………18
3.3 Knowledge Base………………………………………………20
3.3.1 Rule Base………………………………………………………20
3.3.2 Data Base………………………………………………………22
3.3.2.1 Selection of Membership Function for Fuzzy Set……22
3.3.2.2 Fuzzy Partition of Input/Output Space………………22
3.4 Fuzzy Inference Engine……………………………………25
3.4.1 Fuzzy Relation…………………………………25
3.4.2 Fuzzy Implication………………………………27
3.4.3 Approximate Reasoning…………………………29
3.4.3.1 Tautologies and Compositional Rule of Inference…29
3.4.3.2 Generalized Modus Ponens (GMP) and
Generalized Modus Tollens (GMT)……………30
3.4.4 Multiconditional Approximate Reasoning………31
3.4.5 Fuzzy Inference System…………………………35
3.4.5.1 Mamdani’s Fuzzy Model………………..36
3.4.5.2 Takagi and Sugeno’s Fuzzy Mode………37
3.5 Defuzzification Interface……………………………………39
3.5.1 Center of Area (COA) Defuzzificatio……………40
3.5.2 Center of Sum (COS) Defuzzification……………40
4 FUZZY CONTROL OF AUTO-WAREHOUSING SYSTEM…………………42
4.1 Introduction………………………………………………42
4.2 Design of Fuzzy Logic Controller…………………………44
4.3 Evaluation…………………………………………………52
4.3.1 Simulation Study…………………………………52
4.3.2 Design of Reference Curve………………………53
4.3.3 The Control Scheme………………………………55
4.3.4 Experimental Results……………………………56
4.4 Discussion…………………………………………………58
4.5 Conclusion…………………………………………………63
5 FUNDAMENTAL OF NEURAL NETWORKS………………66
5.1 Introduction………………………………………………66
5.2 Model of a Neuron…………………………………………67
5.3 Feedforward Neural Networks……………………………69
5.3.1 Single-Layer Perceptron Network………………69
5.3.2 Multi-Layer Feedforward Neural Network………72
5.4 Back-Propagation Learning Process……………………74
5.4.1 Back-Propagation Networks (BPN)………………74
5.4.2 Derivation of the Back-Propagation Learning
Algorithm…………………………………………76
6 SELF-ORGANIZATION NEURO-FUZZY SYSTEM………82
6.1 Introduction………………………………………………82
6.2 Cluster-Based Self-Organization Process………………84
6.2.1 Cluster-Based Algorithm for Input Space artitioning…84
6.2.2 Rule Base Constructing…………………………87
6.3 Neural-Fuzzy Architecture of SO-NFS……………………89
6.4 Learning Control Schemes…………………………………94
6.4.1 Inverse Learning (General Learning) Scheme…95
6.4.2 Pseudo-error Learning Scheme…………………97
6.4.3 Random Optimization (RO) for Pseudo-Error Learning…99
7 LEARNING CONTROL APPLICATIONS…………………105
7.1 Water Bath Temperature Control…………………………105
7.2 Inverted Pendulum Control………………………………114
7.3 Crane System of Auto-warehousing System………………125
7.4 Discussion…...……………………………………………133
8 Conclusion………………………………………………………135
REFERENCES……………………………………………………137
VITA…………………………………………………………………141

REFERENCES
Baba, N., 〝A new approach for finding the global minimum of error function of neural networks,〞Neural Networks, vol. 2, no3, pp. 367-373, 1989.
Baraldi, A., Blonda, P., 〝A survey of fuzzy clustering algorithms for pattern recognition  Part Ⅰ&Ⅱ,〞IEEE Transactions on Systems, Man, and Cybernetics, Part B, vol. 29, no. 6, pp. 778-801, 1999.
Bojadziev, G. and Bojadziev, M., Fuzzy sets, fuzzy logic, applications, Singapore World Scientific, 1995.
Brown, M. and Harris, C., Neurofuzzy adaptive modeling and control, Prentice Hall, New York, 1994.
Buckley, J. J., 〝Sugeno type controller, are universal controllers,〞Fuzzy Sets and Systems, vol. 52, pp. 299-303, 1993.
Franklin, G. F., Powell, J. D., and Emani-Naeini, A., Feedback control of dynamic system, 3nd Edition, Addison-Wesley, Reading, Massachusetts, 1995.
Hagan, M. T., Demuth, H. B., and Beale, M., Neural network design, PWS publishing company, Boston, 1995.
Haykin, S., Neural Networks, Macmillan college publishing company, New York, 1994.
Hornik, K., Stinchcombe, M., and White, H., 〝Multilayer feedforward networks are universal approximators,〞 Neural Networks, vol. 2, pp. 359-366, 1989.
Jang, J. S. R., Sun, C. T., and Mizutani, E., 〝Neuro-fuzzy and soft Computing,” Prentice Hall, 1997.
Jang, J. R.,〝Self-learning fuzzy controllers based on temporal back propagation,〞 IEEE Transactions on Neural Networks, vol. 3, no. 5, pp. 714-723, 1992.
Juang, C. F. and Lin, C. T., 〝An on-line self-constructing neural fuzzy inference network and its application,〞IEEE Transactions on Fuzzy Systems, vol. 6. no. 1, 1998.
Kim, H. R. and Lee, H. S.,〝Postprocessor using fuzzy vector quantiser in HMM-based speech recognition,〞Electronics Letters, vol. 17, no. 22, pp. 1998-2000, 1991.
Klir, G. J. and Yuan, B., Fuzzy sets and fuzzy logic, Prentice Hall RTP, Upper Saddle River, 1995.
Kuo, B.C., Automatic control system, 7th Edition, Prentice Hall, Englewood Cliffs, 1995.
Lee, C. C., 〝Fuzzy logic in control systems: Fuzzy logic controller  Part Ⅰ,〞IEEE Transactions on Systems, Man, and Cybernetics, vol. 20, pp. 404-418, 1990.
Lee, C. C., 〝Fuzzy logic in control systems: Fuzzy logic controller  Part Ⅱ,〞IEEE Transactions on Systems, Man, and Cybernetics, vol. 20, pp. 419-435, 1990.
Li, Y. F. and Lan, C. C., 〝Development of fuzzy algorithms for servo systems,〞IEEE Control Systems Magazine, pp. 65-72, 1989.
Li, Chunshien and Priemer, Roland, 〝Fuzzy control of unknown multiple-input-output plants,〞Fuzzy Sets and Systems, vol. 104, no. 2, pp.245-267, 1999.
Li, Chunshien, Ph.D. thesis, 〝Self-learning fuzzy control of unknown plants,〞Department of Electrical Engineering and Computer Science, University of Illinois at Chicago, 1996.
Li, Chunshien and Priemier, Roland,〝Self-learning general purpose PID controller,〞 Journal of the Franklin Institute vol. 334B, no. 2, pp. 167-189, 1997.
Lin, C. T., Juang, C. F., and Huang, J. C., 〝Temperature control of rapid thermal processing system using adaptive fuzzy network,〞Fuzzy Setes and Systems, vol. 103, pp.49-65, 1999.
Lin, C. T. and Lee, C. S. G., Neural fuzzy systems, Prentice Hall PTR, Upper Saddle River, 1996.
Matyas, J., 〝Random optimization,〞Automatic and Remote Control, Vol. 26, pp. 243-253, 1965.
Nguyen, D. H. and Widrow, B., 〝Neural networks for self-learning control syetems,〞IEEE Control Systems Magazine, pp. 18-23, 1990.
Ogata, K., Modern Control Engineering, 2nd Edition, Prentice Hall, Englewood Cliffs, New Jersey, 1990.
Peacock, A. M., Renshaw, D., and Hannah, J., 〝Fuzzy data fusion approach for image processing,〞Electronics Letters, vol. 35, no. 18, pp.1527-1529, 1999.
Plataniotis, K. N., Androutsos, D., and Venetsanopoulos, A. N. 〝 Adaptive fuzzy systems for multichannel signal processing,〞Proceeding of the IEEE, vol. 87, no. 9, pp. 1601-1622, 1999.
Psaltis, D., Sideris, A., and Yamamura, A., 〝A multilaryed neural network controller,〞 IEEE Control Systems Magazine, vol. 8, no. 4, pp. 17-21, 1988.
Solis, F. J. and Wets, J. B., 〝Minimization by random research techniques,〞Mathematics of Operations Research, vol. 6, pp. 367-373, 1989.
Sugeno, M. and Nishida, M., 〝Fuzzy control of model car,〞Fuzzy Sets and Systems, vol.16, pp. 103-113, 1985.
Sugeno, M. and Yasukawa, T., 〝A fuzzy-logic-based approach to qualitative modeling,〞IEEE Transaction on Fuzzy Systems, vol. 1, pp. 7-31, 1993.
Takagi, T. and Sugeno, M.,〝Fuzzy identification of systems and its applications to modeling and control,〞IEEE Transactions on Systems, Man, and Cybernetics, SMC-15(1), pp. 116-132, 1985.
Tanomaru, J. and Omatu, S., 〝Process control by on-line trained neuural controller,〞IEEE Transactions on Industrial Electrons, vol. 39, pp. 511-521, 1992.
Tong, R. M., Beck, M. B. and Latten, A., 〝Fuzzy control of the activated sludge wastewater treatment process,〞Automatica, vol. 16, no. 6, pp. 695-701, 1980.
Tsoukalas, L. H. and Uhrig, R. E., Fuzzy and neural approaches in Engineering, Wiley, New York, 1997.
Wang, L. and Langari, R., 〝Building Sugeno-type models using fuzzy discretization and orthogonal parameter estimation techniques,〞IEEE Transactions on Fuzzy Systems, vol. 3, no. 5, pp. 724-740, 1992.
Wang, L. X., 〝Fuzzy system are universal approximators,〞IEEE Internetinal Conference on Fuzzy Systems pp. 1163-1169, 1992.
Zadeh, L. A., 〝Fuzzy Sets,〞Inforation and Control, 8, pp. 338-353, 1965.
Zadeh, L. A., 〝Fuzzy algorithms,〞Inforation and Control, 12, pp. 94-102, 1968.
Zhang, W, R., 〝NPN fuzzy sets and NPN qualitative algebra: a computational framwork for bipolar cognitive modeling and multiagent decision analysis,〞IEEE Transactions on Systems, Man, and Cybernetics, Part B, vol. 26, no. 4, pp. 561-574, 1996.