臺灣博碩士論文加值系統

本論文中，我們將討論動態灰色預測控制器的設計。我們嘗試著結合模糊理論、灰色預測控制以及基因演算法之優點來發展這樣的動態控制系統。我們將注意力集中在預測模式的變化轉換，不同於傳統方法以動態的預測步距選取為主。預測模式間的轉換及時機的規劃是我們設計的出發點，而且我們也會探討並非整個控制過程中都適合以預測的行為來控制系統，所以我們嘗試著去改變調整完整的控制器架構。最後由基因演算法來搜尋複雜的系統參數組合。藉由模擬的範例結果，我們可以發現所提出方法能使響應可以追蹤所期望的較佳之響應軌跡。

In this thesis, we will discuss the dynamic grey prediction systems. We try to integrate the advantages of fuzzy theory, grey prediction control, and genetic algorithm to develop a dynamical prediction control system. We pay our attention to the switch of the distinct grey prediction modes. It is different from the traditional grey prediction systems that usually focus on the selection of dynamic prediction steps. Furthermore, we also analyze the opportune moments for forecasting the system behaviors. We can know that there is not all of the control processes suitable to implement the prediction control. So we should modify the framework of the controller. Finally, we use genetic algorithms to help us for designing such a complex system. From the results of simulated experiments, we can see that the proposed methods make the performance response to track the desired trajectory well.

The Design and Application of Control Using Dynamic Grey Prediction System
Chapter 1 Introduction .................................. 1
1.1 Background .......................................... 1
1.2 Motivation........................................... 2
1.3 Organization ........................................ 3
Chapter 2 Grey Prediction .......................................... 4
2.1 Introduction .......................................... 4
2.2 The Mapped Generating and its Inverse Operator (MGO and IMGO) ........ 4
2.3 The Complete Prediction Process .......................................... 5
Chapter 3 Genetic Algorithm .......................................... 8
3.1 Introduction .......................................... 8
3.2 The Basic Operators in GA .......................................... 8
3.2.1 Reproduction .......................................... 9
3.2.2 Crossover .......................................... 9
3.2.3 Mutation .......................................... 10
3.3 Elite Method .......................................... 10
3.4 Reinforced Search Method .......................................... 11
Chapter 4 Application of a Dynamic Grey Prediction System Using Fuzzy Logic to
Switch the Prediction Modes ........................... 13
4.1 Introduction .......................................... 13
4.2 Grey Prediction Controller .......................................... 14
4.3 Genetic-based and Fuzzy-switching GPC ................................... 17
4.4 Simulation and Discussion .......................................... 20
4.5 Conclusion .......................................... 25
Chapter 5 An Alternative Approach for the Switching Grey Prediction Controller
.............. 26
5.1 Introduction .......................................... 26
5.2 Problem Formulation .......................................... 26
5.3 The Proposed Method .......................................... 29
5.4 Simulation .......................................... 31
5.5 Conclusion .......................................... 33
Chapter 6 Conclusions and Recommendations .............................. 34
References .......................................... 36
List of Figures
Fig. 2.1 The prediction procedure of a simple grey predictor ............. 6
Fig. 3.1 The complete flow chart of the genetic algorithm ................. 12
Fig. 4.1 The framework of a simple grey prediction controller ............ 14
Fig. 4.2 The output response of G(s) using a positive step in GPC ...... 15
Fig. 4.3 The output response of G(s) using a negative-step in GPC ....... 15
Fig. 4.4 The framework of the switching grey prediction PID controller ...... 16
Fig. 4.5 The Block Diagram of the Genctic-based and Fuzzy-switching GPC
controller .......................................... 17
Fig. 4.6 The fuzzy sets in the premise part of the fuzzy inference scheme ....18
Fig. 4.7 The fuzzy sets in the consequence part of the fuzzy inference
scheme .........18
Fig. 4.8 The step response of Gp(s) using GB-FSGPC method (9 parameters in
searching .......................................... 20
Fig. 4.9 The step response with kp in [5,15] ......................... 22
Fig. 4.10 The step response with wn=1 and wp=1 ............................ 23
Fig. 4.11 The system response with Ziegler-Nichols PID rules .............. 25
Fig. 5.1 The results of feeding Unit step signal into the stable plant Gp(s)
(Eq. 4.9 ) .......................................... 27
Fig. 5.2 The results of feeding Unit step signal into the unstable plant
Gu(s) ...................28
Fig. 5.3 The results of feeding an Inverse Unit step input signal into the
unstable plant .......................................... 29
Fig. 5.4 An Alternative Approach of the SGPC controller ................ 29
Fig. 5.5 The output response of controlling the Gp(s) by the proposed method
.......... 32
Fig. 5.6 The output response of controlling the unstable plant G2(s) ...... 33
List of Table
Table 4.1 The code length of the searched arguments in the proposed method
(GB-FSGPC) .......................................... 20
Table 4.2 The system parameters of Gp(s) using GB-FSGPC method .......... 21
Table 4.3 The performance indices of the system Gp(s) using GB-FSGPC design
.............. 21
Table 4.4 The searched results with kp in [5,15] ......................... 21
Table 4.5 The performance indices with kp in [5,15] ..................... 22
Table 4.6 The searching results with wn=1 and wp=1 ..................... 22
Table 4.7 The performance indices with wn=1 and wp=1 .................... 23
Table 4.8 The Ziegler-Nichols PID rules ................................ 24
Table 4.9 The system arguments with the Ziegler-Nichols rules .............. 24
Table 4.10 The performance indices with the Ziegler-Nichols rules ........... 25
Table 5.1 The system parameters of the design to control Gp(s) (a third-order
stable plant) .......................................... 31
Table 5.2 The performance indices of Gp(s) (a third-order stable plant) .... 31
Table 5.3 The system parameters of controlling the unstable plant G2(s)
( a third-order unstable plant ) ............................. 32
Table 5.4 The performance indices of controlling G2(s) ( a third-order unstable
plant ) .................................. 32

Reference[1] J. L. Deng, “Introduction to Grey System Theory”, The Journal of Grey System, Vol. 1, pp. 1-24, 1989.[2] C. C. Wong and W. C. Liang, “Design of Switching Grey Prediction Controller”, The Journal of Grey System, Vol. 1, pp. 47-60, 1997.[3] C. C. Wong, W. C. Liang, H. M. Feng, and D. A. Chiang, “Grey Prediction Controller Design”, The Journal of Grey System, Vol. 2, pp. 123-131, 1998.[4] C. C. Wong and C. C. Chen, “Switching Grey Prediction PID Controller Design”, The Journal of Grey System, Vol. 4, pp. 323-333, 1997.[5] Z. M. Chen, H. X. Lin, and Q. M. Hong, “The Design and Application of a Genetic-Based Fuzzy Gray Prediction Controller”, The Journal of Grey System (in Chinese), Vol. 1, No. 1, pp. 33-45, 1998.[6] J. L. Deng, “Control problems of grey system”, Systems and Control Letters, Vol. 5, pp. 288-294, 1982.[7] B. Cheng, “The grey control on industrial process”, Journal of Huangshi College (in Chinese), Vol. 1, pp. 11-23, 1986.[8] D. E. Goldberg, “Engineering Optimization via Genetic Algorithm”, Ninth Conference on Electronic Computation, pp. 471-482, 1986.[9] J. J. Zhao, “Design of Optimal Fuzzy Controllers Based on Genetic Algorithms”, Master Thesis, NCU, 1996.[10] J. H. Holland, “Outline for a logical theory of adaptive system”, J. ACM, Vol. 3, pp. 297-314, 1962.[11] K. A. DeJong, “An analysis of the behavior of a class of genetic adaptive system”, Ph.D. dissertation (CCS), Univ. Mech., Ann Arbor, 1975.[12] J. Mitzpatrick, J. J. Grefenstette, and D. Van Gucht, “Image registration by genetic search”, in Proc. IEEE southeastcon’84, pp. 460-464, 1984.[13] J. J. Grefenstette, R. Gopal, B. Rosmaita, and D. Van Gucht, “Genetic algorithm for the traveling salesman problem”, in Proc. Int. Conf. Genetic Algorithms and their Applications, pp. 160-165, 1985.[14] K. Kristinsson and G. A. Dumont, “Genetic algorithms in system identification”, Third IEEE Int. symp. Intelligent contr., Arlington, VA, pp. 597-602, 1988.[15] J. J. Grefenstette, Ed., Pro. Int. Conf. Genetic Algorithm and their applications. Hillsdale, NJ: Lawrene Erlbaum Associates, 1985.[16] D. J. Scheffer, Ed., Proc. 3rd Int. Conf. Genetic Algorithms. Palo Alto, GA: Morgan Kaufmann, 1989.[17] L. Davis, Ed. Genetic Algorithms and Simulated Annealing. London: Pitman, 1987.[18] D. E. Goldberg, Genetic Algorithms in Search, Optimization and Machine Learning. Reading, MA: Addison-Wesley, 1989.[19] Z. Y. Zhao, M. Tomizuka, and S. Isaka, “Fuzzy Gain Scheduling of PID Controllers”, IEEE Trans. on Systems, Man, and Cybernetics, Vol. 23, No. 5, pp. 1392-1398, 1993.[20] J. H. Chen, H. L. Chen, H. Y. Chung, “Study on Fuzzy Controller and Grey Prediction Model for the Motion of a Robot”, Proc. 23nd National Symposium on Automatic Control, pp. 379-383, Taiwan, R.O.C., March 2001.[21] J. H. Chen, “Fuzzy Controller and Grey Prediction Model Applied to the Motion Control of a Robot Arm under the Windows”, Master Thesis, NCU, 2001.[22] 王文俊, “認識Fuzzy”, 全華科技圖書股份有限公司 出版, 1997.[23] 翁慶昌 等, “灰色系統基本方法及其應用”, 高立圖書有限公司 出版, 2001.[24] 江金山 等, “灰色理論入門”, 高立圖書有限公司 出版, 1998.[25] C. C. Hang, K. J. Astrom, and W. K. Ho, “Refinements of the Ziegler- Nichols tuning formula.” Proc. IEE, Pt. D., vol. 138, pp. 111-118, 1991.