臺灣博碩士論文加值系統

在本論文中，由於在實際的應用中，原有的ANFIS反向學習法，無法有效地代表非線性受控體的反向動態模型，進而有效的設計出感應馬達的速度控制器，所以在本論文中，我們提出一種新的ANFIS反向學習方法，來對受控體進行有效的學習，達到利用ANFIS架構表示受控體反向動態模型，並且有效的設計出針對感應馬達的速度控制器，經過實驗證明我們所提出的學習法是一種簡單且有效的方法；接著再利用XSCALE PXA255嵌入式系統來實現我們所提出的控制法則，使用Embedded Linux系統當控制器的發展系統，成功的完成嵌入式智慧型控制器，不必再使用高成本的個人電腦或單板電腦以及昂貴的軟體來製作智慧型控制器。在實現嵌入式智慧型控制器方面，我們使用LINUX C語言撰寫此控制器，當我們收集完受控體的輸入、輸出資料，再利用TCP檔案傳輸程式將所採集的資料傳至遠端，在遠端我們使用off-line的訓練方法獲取受控體的反向動態模型，再將反向動態模型的參數集合以文字檔的形式透過TCP傳至嵌入式系統，在系統上我們有撰寫修改型的ANFIS回想架構，當獲取反向動態模型的參數集合即可對受控體進行控制，實驗結果證明我們所撰寫的ANFIS回想架構可以準確的控制我們的受控體，我們的嵌入式智慧型控制器不但擁有相當的適應性，而且具有優越的控制能力，在我們的實驗過程中，不但證明我們所提出的修改型反向學習法可以有效的表示受控體的反向動態模型，而且我們可以準確的對受控體進行有效的控制。

Because in real application, original ANFIS inverse learning method can not represent the inverse dynamic model of induction motor effectively. Thus in this thesis, a modified ANFIS inverse learning method is proposed. A modified training phase by introducing an overall transfer function and a modified application phase by introducing a feedback configuration which combines the ANFIS inverse model and the integral controller are proposed. This new design method gives us a simple and powerful way to design a speed controller for a servo-motor. An experimental result for a 3-phase induction motor is also given to illustrate the effectiveness of the proposed design method.
Moreover, in this thesis, a way to realize the modified ANFIS inverse learning method by XSCALE PXA255 embedded system is also proposed. In realizing the embedded intelligent controller, we use LINUX C language to implement the modified ANFIS control rules. The proposed program can gather the input and output data set of the plant, and send the data set to us via the TCP/IP protocol. In this way, we can easily get the required input and output data set using the host PC and obtain the inverse dynamic model using the off-line training method. While obtaining the inverse dynamic model, we also show a way to produce a text file which contains the parameter set of the model and transmit it to the embedded system via the TCP/IP protocol. Hence in this thesis, we have successful realized an embedded intelligent controller which combines the modified ANFIS control law and the TCP/IP communication program.

目 　錄
論 文 摘 要 I
ABSTRACT III
誌 　謝 IV
目 　錄 V
圖 目 錄 VII
表 目 錄 IX
第一章 前言 1
1-1研究背景與動機 1
1-2文獻探討 2
1-3研究目的 2
1-4研究範圍與限制 3
1-5研究假設 3
1-6論文架構 3
第二章 ANFIS理論與實驗 4
2-1 ANFIS架構 4
2-2 ANFIS 學習演算法 6
2-3反向學習法 11
2-4感應馬達速度控制 12
2-4-1感應馬達驅動設備 12
2-4-2使用反向學習法實作感應馬達速度控制 14
2-5 小結 18
第三章 修改型ANFIS理論與實驗 19
3-1修改型反向學習法 19
3-2驗證修改型反向學習法 20
3-3 使用修改型反向學習法實作感應馬達速度控制 25
3-4 小結 29
第四章 以嵌入式LINUX系統實作修改型ANFIS控制器 30
4-1以C語言實現ANFIS控制器 30
4-1-1撰寫ANFIS架構 30
4-1-2驗證ANFIS控制器 35
4-2嵌入式ARM LINUX系統 39
4-2-1 XScale PXA 255 39
4-2-2工具鏈(Toolchain) 41
4-2-3嵌入式 LINUX核心 42
4-2-4嵌入式LINLUX檔案系統 42
4-2-5 Bootloader 43
4-2-6下載嵌入式LINUX系統 45
4-2-7 FPGA驅動程式 47
4-3實現嵌入式ANFIS控制器 49
4-3-1 一階受控體製作 50
4-3-2 A/D、D/A資料採集 51
4-3-3嵌入式修改型ANFIS控制器 61
4-3-4製作嵌入式ANFIS系統 65
第五章 結論 66
參考文獻 67
作者簡介 70

參考文獻
[1] L.A.Zadeh, ”Fuzzy Set” ,Information and Control,Vol.8,pp.338-353, 1965.
[2] C.C.Lee, “Fuzzy logic in control systems : Fuzzy logic controller-part I” IEEE Tra-ns.System,Man,Cybern.,Vol.20,No.2,PP.404-418,Mar./Apr.1990.
[3] C.C.Lee, “Fuzzy logic in control systems : Fuzzy logic controller-part Ⅱ” IEEE Tra-ns.System,Man,Cybern.,Vol.20,No.2,PP.419-435,Mar./Apr.1990.
[4] W.R.Hwang and W.E.Thompson,”Design of intelligent fuzzy logic controllers using genetic algorithms”, Proc IEEE Conf.,Fuzzy Systems,IEEE World Congress on Computational Intelligence.,pp.1383-1388.1994.
[5] J. S. Jang, “Self-learning fuzzy controllers based on temporal back propagation”,IE-EE Transactions Neural Networks, Vol. 3, pp. 714-724, 1992.
[6] J.S.R.Jang, C.T. Sun, and E.Mizutani, Neuro-fuzzy and Soft Computing, Prentice, Hall, New Jersey, 1997.
[7] J.S.R. Jang, "ANFIS: adaptive-network-based fuzzy inference system", IEEE Trans. Syst., Man, Cybern. 23, pp. 665-685, 1993.
[8] M. A. Denai, F. Palls, and A. Zeghbib, "ANFIS Based Modelling and Control of Non-linear System: A tutorial", Systems, Man and Cybern., IEEE International Con-ference on Vol. 4, pp. 3433-3438, 10-13 Oct, 2004.
[9] H.Bersini, J. P. Nordvik and A. Bonarini, “A simple direct adaptive fuzzy controll-er derived from its neural equivalent”, IEEE International Conference on Fuzzy S-ystem, Vol. 1, pp. 345-350, 1993.
[10] L.Shaoyuan and X. Yugeng, “A fuzzy- neural network for adaptive control of no-nlinear dynamic systems”, IEEE International Fuzzy Systems Conference Proceedi-ngs, Vol. 1, pp. 449-453, 1999.
[11] J.Kim and N.Kasabov, “HyFIS: adaptive neuro-fuzzy inference systems and their application to nonlinear dynamical systems”, Neural Networks, Vol. 12, pp.1301-1319, 1999.
[12] Y. C. Wang, C. J. Chien and C. C. Teng, “Takagi-Sugeno recurrent fuzzy neural networks for identification and control of dynamic systems”, IEEE International Conference on Fuzzy Systems, Vol. 1, pp. 537-540, 2001.
[13] M. Vasudevan, and R. Arumugam, "A Robust Torque Control of Induction Motor for Electric Vehicle Applications Using ANFIS", TENCON, IEEE Region 10 Co-nference on Vol. D, pp. 17-20, 21-24 Nov., 2004.
[14] M.V. Aware, A.G. Kothari, and S.O. Choube, "Application of Adaptive Neuro-Fuz-zy Controller (ANFIS) for Voltage Source Induction Motor Drive", Power Electro-nics and Motion Control Conference, Vol. 2, pp. 935-939,15-18 Aug., 2000.
[15] T.C. Ahn, Y.W. Kwon, H.S. and Hwang, P. W, "Design of Neuro-Fuzzy Controlle-r on DSP for Real-Time Control of Induction Motors", IFSA World Congress an-d 20th NAFIPS International Conference, pp. 3038-3043, 25-28 July, 2001.
[16] S.Paramasivam, R. Arumugam, and B. Umamaheswari, "Indirect Rotor Position Es-timation of Switched Reluctance Motor Using ANFIS", Power Electronics and Dr-ive Systems Conference, pp. 921-926, 17-20 Nov, 2003.
[17] J.H. Park, D.H. Kim, S.S. Kim, D.J. and Lee, M.G. Chun, "C-ANFIS Based Faul-t Diagnosis for Voltage-Fed PWM Motor Drive Systems", IEEE Annual Meeting of the Fuzzy Information, Vol. 1, pp. 379-383, 27-30 June, 2004.
[18] W.A.Kwong and K.M.Passino,”Dynamically Focused Fuzzy Learning Control”, Sys-tem , Man and Cybernetics,Part B,IEEE Transactions on. , Vol.26 , Issue : 1 , F-eb .1996,PP.53-74.
[19] C. T. Lin, C. J. Lin, and C. S. Lee, “Fuzzy Adaptive Learning Control Network with on-line Neural Learning”, Fuzzy Sets and Systems, Vol. 71, pp. 25-45, 1995.
[20] Z. Li, D. Xiao and S. He, “A fuzzy adaptive PID controller based on neural net-works”, Control and Decision, Vol. 8, No. 3, pp. 340-345, May 1996.(in Chines-e)
[21] L.Peng and P. Y. Woo, “Neural-fuzzy control system for robotic manipulators”, I-EEE Control Systems Magazine, Vol. 22, pp. 53-63, 2002.
[22] Michael K.Johson and Erik W.Troan,”Linux Application Development”,Addison W-esley,2004.
[23] Karim Yaghmour,”Building Embedded Linux System”,O’REILLY,2005.
[24] 張斐章、張麗秋、黃浩倫，”類神經網路理論與實務”，東華書局，2003。
[25] 王進德，”嵌入式 Linux 程式設計”，全華圖書，2005。
[26] 黃國勝，”嵌入式系統 I/O 界面軟硬體實務”，全華科技，2004。
[27] Craig Hollabaugh and Ph.D，陳清豪 譯，”內嵌式 Linux 系統”，培生教育出版集團，2002。
[28] 徐千洋，”LINUX C 函式庫參考手冊”，旗標出版股份有限公司，2003。