臺灣博碩士論文加值系統

如何設計一個低複雜度、快速建模、低輸出誤差的推論系統一直是學術與工程界的一個努力目標。模糊系統的設計過程一般包含兩個步驟：結構辨識及參數辨識。結構辨識之目的在於確定系統變數，建立模糊法則庫等。待結構確立後，如何最佳化歸屬函數及調整系統一些相關參數，則屬參數辨識之工作。
為了在不影響模型精確度的條件下建立模型，本研究利用灰色理論之灰關聯分析技術，計算對系統有影響之所有輸入變數與輸出變數之灰關聯度，再依其灰關聯度大小加以排序，以決定輸入與輸出變數的關係，據以選擇較重要的變數，以簡化所設計模型之複雜度。此外，更將此應用於彙整式搜尋引擎(meta-search engine)上，用以傳回排序過之網址，並改良灰關聯度為區間式灰關聯度，以提供較相關聯之關鍵字，預作個人化網頁或資訊搜尋之前置處理。
在參數辨識的過程中，不同歸屬函數間之動態調整會彼此相互影響，因此通常需要重複迭代或利用複雜調整方法，才能收斂得到較佳結果。故本學位論文將提出以模糊理論結合可拓學的物元變換思維方法，建立嶄新的可拓模糊推論機制。藉由經典域延伸至可拓域，歸屬函數擴展至可拓關聯函數，同時動態考量不同區域的歸屬函數所造成之相互影響關係，則可加速參數調整與推論的速度並可降低系統的輸出誤差。在本文中除了探討如何定義可拓關聯函數外，也對模型參數的調整及式子推導做有系統的介紹以符合系統規範的需求。
最後，本文除了將以數個常用案例模擬的結果來驗證低輸出誤差外，更以同等之平均平方誤差(mean squared error)來測試所需之迭代(iteration)次數，結果顯示，所提之可拓模糊推論機制優於其他傳統方法。

How to design an inferred system with low complexity, fast modeling, and low output error is an important issue for both the academy and industry. The identification of a fuzzy model includes both the structure and parameter identifications. The purposes of the structure identification are to select such as the system variables and to establish the fuzzy rule base. After identifying the structure, how to optimize the membership functions and to adjust some other related parameters belong to the job of parameter identification.
In order to maintain the requirement of good performance and possess the fast inference characteristics, the grey relational analysis of grey theory to calculate the grey relational degrees between the effective inputs and output variables is utilized by this dissertation. Then, we can order the degrees to determine the relative importance of the prospective factors to the main factor to simplify the structure identification. Besides, based on the grey relational method, the new orders of websites which returned from a meta-search engine is implemented. Furthermore, we improve the grey relational degree to regional relational degree to supply more related keywords for personal websites or preprocess of information searching.
During parameter identifications, dynamically adjusting the membership functions to satisfy one pattern may deteriorate the inference outcomes of the others. This may take many refining iterations or complicated method to achieve a satisfactory result. Thus, a novel extenics-based fuzzy inference mechanism, integrating the extended transformation method of extenics into the fuzzy inference model, is proposed in this dissertation. With the classic region being extended to extended region and membership function being extended to extended relational function, the inference speed will be improved and the output error will be reduced by dynamically considering the mutual effect of different zones of membership functions simultaneously. We will investigate how to define the extended relational functions and how to refine the roughly designed model. Besides, those equations will be systemically derived step by step to meet the system requirement.
Finally, not only some of commonly used cases will be evaluated for the low error outputs but also the iterations taken while given the same mean squared error will be counted. Simulation results from both single-input-single-output and double-input-single-output systems verified that the proposed extenics-based fuzzy inference model has better performance than the conventional methods.

封面
摘要
Abstract
誌謝
Contents
List of Tables
List of Figures
Glossary of Symbols
Chapter 1 Introduction
1.1 Motivation
1.2 The convertional approaches
1.3 Summary
Chapter 2 Fuzzy System
2.1 Fuzzy sets and fuzzy logic
2.2 Fuzzy sets operations
2.3 The construction of fuzzy model
2.4 The fuzzy reasoning types
2.5 Summary
Chapter 3 Grey Relational Analysis
3.1 Grey Relational degree
3.2 Examples
3.3 Summary
Chapter 4 The Extension Set and Incompatible Problem
4.1 The definition of matter elements
4.2 The expansion ability of matter element
4.3 The evaluation of matter element expansion
4.4 The rhombus thinking method
4.5 The extension set
4.6 The extended relational function
4.7 The distance between a datum and an interval
4.8 The positional distance between a datum and two intervals
4.9 Summary
Chapter 5 The Proposed Extenics-Based Fuzzy Model
5.1 The single-input-single-output system
5.2 The double-input-single-output system
5.3 The data transformation methodology
5.4 Summary
Chapter 6 Simulation Results
6.1 Single-input-single-output case
6.2 Double-input-single-output case
6.3 The required iterations under different MSEs
6.4 SISO and DISO by transfornation methodology
6.5 Summary
Chapter 7 Conclusions
Appendix
References
Puublications
Referred papers
Conference papers
作者簡介(Vita)

[1] T. Takagi and M. Sugeno, “Fuzzy identification of systems and its applications to modeling and control,” IEEE Trans. Systems, Man, and Cybernetics, vol. 15, no. 1, pp.116-132, 1985.
[2] M. Sugeno and T. Yasukawa, “A fuzzy-logic-based approach to qualitative modeling,” IEEE Trans. Fuzzy Systems, vol. 1, no. 1, pp.7-31, 1993.
[3] H.-S. Hwang and K.-B. Woo, “Linguistic fuzzy model identification,” IEE Proc.-Control Theory Appl, vol. 142, no. 6, pp.537-544, 1995.
[4] J. L. Deng, “Introduction to grey system theory,” The Journal of Grey System, vol. 1, no. 1, pp.1-24, 1989.
[5] L. Ott, An introduction to statistical methods and data analysis, 3rd edition, Boston, MA: Pws-Kent, 1988.
[6] Y.-P. Huang and C.-C. Huang, “The integration and application of fuzzy and grey modeling methods,” Fuzzy Sets and Systems, vol. 78, no. 1, pp.107-119, 1996.
[7] W. Cai, “The extension set and incompatible problem,” J. of Scientific Exploration, vol. 1, pp.81-93, 1983.
[8] L. A. Zadeh, “Fuzzy sets,” Information and Control, vol. 8, pp.338-353, 1965.
[9] E. H. Mamdani, “Application of fuzzy algorithms for simple dynamic plant,” Proc. IEE, vol. 121, no. 12, pp.1585-1588, 1974.
[10] E. H. Mamdani, “Advances in linguistic synthesis of fuzzy controller,” Int. J. Man Machine Studies, vol. 8, no. 6, pp.669-678, 1976.
[11] E. H. Mamdani and S. Assilian, “An experiment in linguistic synthesis with a fuzzy logic control,” Int. J. Man Machine Studies, vol. 7, no. 1, pp.1-13, 1975.
[12] E. Kim, M. Park, S. Ji, and M. Park, “A new approach to fuzzy modeling,” IEEE Trans. Fuzzy Systems, vol. 5, pp.328-337, 1997.
[13] H. Nomura, I. Hayashi, and N. Wakami, “A learning method of fuzzy inference rules by descent method,” Proc. FUZZ-IEEE, pp.203-210, 1992.
[14] D. Nguyen and B. Widrow, “The truck backerupper: An example of self-learning in neural network,” IEEE Contr. Syst. Mag., vol. 10, no. 3, pp.18-23, 1990.
[15] L.-X. Wang and J.M. Mendel, “Generating fuzzy rules by learning from examples,” IEEE Trans. System, Man, and Cyber., vol. 22, no. 6, pp.1414-1427, Nov./Dec. 1992.
[16] L. Davis, Handbook of Genetic Algorithms, New York: Van Nostrand Reinhold, 1991.
[17] D. E. Goldberg, Genetic Algorithms in Search, Optimization and Machine Learning, Reading, MA: Addison-Wesley, 1989.
[18] J. L. Deng, “Control problems of grey systems,” Systems and Control Letters, vol. 5, pp.288-294, 1982.
[19] W. Cai, C.Y Yang, and W.C Lin, Methods of extension engineering, Science Press, Beijing, China, 1997. (in Chinese)
[20] J. Deng, Grey System Control, HUST Press, Wuhan, China, 1992. (in Chinese)
[21] G. -H. Tzeng and S. -H. Tsaur, “The multiple criteria evaluation of grey relation model,” The Journal of Grey System, vol. 3, 1994.
[22] M. -C. Liu and C. -M. Hong, “The study on the implement of evaluation criteria for computer skill instruction by grey relational analysis,” The Journal of Grey System, vol. 2, pp.117-130, 1997.
[23] J. Deng, The Course of Grey System Theory, HUST Press, Wuhan, China, 1990. (in Chinese)
[24] C.-Y. Pai, “An overall description of grey relation analysis,” Theory and Practice of Grey System, vol. 3, no. 1, pp.119-122, 1993. (in Chinese)
[25] C.-Y. Lo, “Tool steel material selection using grey relation methods,” The National Symposium on Grey System Theory and Applications, Taiwan, pp.261-265, 1997.
[26] M.-S. Hsieh, Y.-H Huang and D.-C. Tseng, “A study on grey choice of function criteria for marking information systems,” Proc. of Third National Conf. on Grey Theory and Applications, Taiwan, pp.315-317, 1998.
[27] Y. Lin and G. A. Cunningham III, “A new approach to fuzzy-neural system modeling,” IEEE Trans. Fuzzy Systems, vol. 3, no. 2, pp.190-198. May 1995.
[28] J. Shue and S. Wu, “The design and implementation of a distributed search engine for WWW homepages,” Master Thesis of National Chung Chen University, Taiwan, pp.14-23, 1997.
[29] J. A. Dickerson and B. Kosko, “Fuzzy function approximation with ellipsoidal rules,” IEEE Trans. Systems, Man, and Cybernetics. part B: Cybernetics, vol. 26, no. 4, pp.542-560, 1996.
[30] G. J. Klir and B. Yuan, “Fuzzy sets and fuzzy logic theory and applications,” Prentice-Hall press, New Jersey, pp.327-367, 1995.
[31] Y.-P. Huang, H.-J. Chen, and H.-C. Chu, "Identifying a fuzzy model by using the bipartite membership functions," Fuzzy Sets and Systems, vol. 118, no. 2, pp.199-214, 2001.
[32] E. Kim, M. Park, S. Kim, and M. Park, “A transformed input-domain approach to fuzzy modeling,” IEEE Trans. Fuzzy Systems, vol. 6, no. 4, pp.596-604, Nov. 1998.
[33] K. Nozaki, H. Ishibuchi, and H. Tanaka, “A simple but powerful heuristic method for generating fuzzy rules from numerical data,” Fuzzy Sets and Systems, vol. 86, pp.251-270, 1997.
[34] Y.-P. Huang and H.-C. Chu, “Simplifying fuzzy modeling by both gray relational analysis and data transformation methods,” Fuzzy Sets and Systems, vol. 104, pp.183-197, 1999.
[35] Y.-P. Huang and H.-J. Chen, “The extenics-based fuzzy modeling method and its applications,” in Proc. IEEE Canadian Conf. on Electrical and Computer Engineering, Edmonton, Alberta, Canada, pp.977-982, May 1999.
[36] F. Guely and P. Siarry, “Gradient descent method for optimizing various fuzzy rule bases,” Proc. FUZZ-IEEE, pp.1241-1246, 1993.
[37] R.M. Tong, ”The evaluation of fuzzy models derived from experimental data,” Fuzzy Sets and Systems, vol. 1, pp.1-6, Feb. 1993.
[38] W. Pedrycz, “An identification algorithm in fuzzy relational systems,” Fuzzy Sets and Systems, vol. 13, pp.153-167, 1984.
[39] C. W. Xu and Y. Z. Lu, “Fuzzy model identification and self-learning for dynamics systems,” IEEE Trans. Systems, Man, and Cybernetics, vol. 17, pp.683-689, July 1987.
[40] G. E. .P. Box and G. M. Jenkins, Time Series Analysis, Forecasting and Control, San Francisco, CA: Holden Day, 1970.
[41] L. Wang and R. Langari, “Building Sugeno-type models using fuzzy discretization and orthogonal parameter estimation techniques,” IEEE Trans. Fuzzy Systems, vol. 3, pp.454-458, Nov. 1995.
[42] M. Sugeno and K. Tanaka, “Successive identification of a fuzzy model and its applications to prediction of a complex system,” Fuzzy Sets and Systems, vol. 42, pp.315-334, 1991.
[43] R. R. Yager and D. P. Filev, Essentials of fuzzy modeling and control, Interscience Press, New York, NY, 1998.
[44] R. R. Yager, and D. P. Filev, “Approximate clustering via the mountain method,” IEEE Trans. on Systems, Man, and Cybernetics.
[45] J. M. Barone, D. P. Filev, and R. R. Yager, “Mountain method-based fuzzy clustering: Methodological considerations,” Technical Report #MII-1320, Machine Intelligence Institute, Iona College, New Rochelle, NY, 1993.