臺灣博碩士論文加值系統

在本文中，我們提出了遞迴式函數鏈結類神經模糊網路結合改良式差分進化演算法和文化改良式差分進化演算法來解決預測和控制的問題。提出的遞迴式函數鏈結類神經模糊網路在歸屬函數層中加入了迴饋訊號的連結，以用來解決時間性的問題。除此之外，兩個有效的學習演算法，稱為改良式差分進化演算法和文化改良式差分進化演算法作為遞迴式函數鏈結類神經模糊網路的參數調整。在差分進化演算法中為了能夠增加突變的差異性，我們從母體中選擇了四條個體進行突變，有助於搜尋解能力的提升。而在文化改良式差分進化演算法中，它是結合了文化演算法和改良式差分進化演算法，在進化的過程中，利用文化演算法中的信仰空間來粹取並使用這些資訊，能夠有助於效能上的提升。模擬結果顯示，我們所提出的方法在收斂速度以及均方根值誤差都要比其它的方法擁有更好的效能。

In this thesis, a recurrent functional neural fuzzy network (RFNFN) with modified differential evolution (MDE) and cultural-based modified differential evolution (CMDE) is proposed for solving prediction and control problems. The proposed RFNFN model has feedback connections added in the membership function layer that can solve temporal problems. Moreover, two efficient learning algorithms, called modified differential evolution (MDE) and cultural-based modified differential evolution (CMDE) for tuning parameters of the RFNFN. In order to increase the diversity of mutations in differential evolution, we randomly choose four individual from the population to mutation. The solution can search capacity more efficiently. In cultural based modified differential evolution (CMDE) combined the cultural algorithm and modified differential evolution. It during the evolutionary process, the belief spaces extraction and use of the information is very effective in increase the performance. Simulation results show that the converging speed and root mean square error (RMSE) of the proposed method has a better performance than those of other methods.

Contexts
摘要 I
Abstract III
Acknowledgment V
Contexts VII
List of Figures IX
List of Tables XI
Chapter 1: Introduction 1
1.1 Motivation 1
1.2 Thesis Organization 6
Chapter 2: The Modified Differential Evolution (MDE) 7
2.1 Review of Differential Evolution 8
2.2 The Proposed Modified Differential Evolution (MDE) 10
2.2.1 Initialization 11
2.2.2 Mutation 11
2.2.3 Crossover 12
2.2.4 Selection 13
2.3 Functions Optimization 14
Chapter 3: The Proposed Recurrent Functional Neural Fuzzy Network (RFNFN) 18
3.1 Functional Link Neural Networks 19
3.2 Structure of the RFNFN Model 21
3.3 Learning Algorithm 25
3.4 Illustrative Examples 25
3.5 Discussion 33
Chapter 4: The Cultural-Based Modified Differential Evolution (CMDE) 34
4.1 The Cultural Algorithm 35
4.2 The Proposed Cultural-Based Modified Differential Evolution (CMDE) 36
4.3 Illustrative Examples 41
Chapter 5: Conclusion 51
Bibliography 53
個人簡歷 59
List of Figures
Figure 1: Pseudo-code of DE/1. 9
Figure 2: Pseudo-code of DE/2. 9
Figure 3: Flowchart of the MDE algorithm. 10
Figure 4: Illustration of the ranking process for 7 individuals. 12
Figure 5: Illustration of the crossover process for D=10 parameters. 13
Figure 6: Structure of FLNN. 20
Figure 7: Structure of the proposed RFNFN model. 22
Figure 8: (a) The prediction results of the proposed MDE method. (b) The prediction errors of the proposed MDE method. (c) The prediction results of the DE2 method. (d) The prediction errors of the DE2 method. (e) The prediction results of the DE1 method. (f) The prediction errors of the DE1 method. 29
Figure 9: The learning curves of the MDE, DE2 and DE1 methods. 30
Figure 10: The learning curves of the MDE, DE2 and DE1 methods. 32
Figure 11: The framework of cultural algorithm. 36
Figure 12: The framework of the proposed CMDE learning method. 37
Figure 13: Flowchart of the CMDE algorithm. 37
Figure 14: (a) The prediction results of the proposed CMDE method. (b) The prediction errors of the proposed CMDE method. 43
Figure 15: The learning curves of the CMDE, MDE, DE2 and DE1 methods. 43
Figure 16: Sphere and coil arrangement of the magnetic levitation system. 45
Figure 17: Control block diagram for the magnetic levitation system. 48
Figure 18: Experimental magnetic levitation system. 49
Figure 19: (a)-(c) Experimental results of RFNFN-CMDE controller, the RFNFN-DE2 controller, and the RFNFN-DE1 controller due to periodic sinusoidal command for reference position and actual position. 50
List of Tables
Table 1: Benchmark function. 14
Table 2: The function value of best, worst, average and standard deviations of MDE, DE2 and DE1 on the test function 1. 15
Table 3: The function value of best, worst, average and standard deviations of MDE, DE2 and DE1 on the test function 2. 15
Table 4: The function value of best, worst, average and standard deviations of MDE, DE2 and DE1 on the test function 3. 16
Table 5: Comparison of proposed performance MDE, DE2 and DE1 methods. 30
Table 6: Comparison of proposed performance MDE, DE2 and DE1 methods. 32
Table 7: Comparison of proposed performance CMDE, MDE, DE2 and DE1 methods. 44

[1] C. H. Chen, C. J. Lin, and C. T. Lin, “A functional-link-based neuro-fuzzy network for nonlinear system control,” IEEE Trans. on Fuzzy Systems, Oct. 2008, Vol. 16, No. 5, pp. 1362-1378.
[2] J. S. R. Jang, “ANFIS: Adaptive-network-based fuzzy inference system,” IEEE Trans. on Syst., Man, and Cybern., 1993, Vol. 23, pp. 665-685.
[3] T. Takagi and M. Sugeno, “Fuzzy identification of systems and its applications to modeling and control,” IEEE Trans. on Syst., Man, and Cybern., 1985, Vol. 15, pp. 116-132.
[4] C. J. Lin, C. H. Chen, and C. Y. Lee, “Efficient immune-based particle swarm optimization learning for neuro-fuzzy networks design,” Journal of Information Science and Engineering, Sep. 2008, Vol. 24, No. 5, pp. 1505-1520.
[5] H. Takagi, N. Suzuki, T. Koda, and Y. Kojima, “Neural networks designed on approximate reasoning architecture and their application,” IEEE Trans. On Neural Networks, Sep. 1992, Vol. 3, pp. 752-759.
[6] E. Mizutani and J. S. R. Jang, “Coactive neural fuzzy modeling,” in Proc. Int. Conf. Neural Networks, 1995, pp. 760-765.
[7] C. H. Chen, C. T. Lin, and C. J. Lin, “A functional-link-based fuzzy neural network for temperature control,” Proceedings of the 2007 IEEE Symposium on Foundations of Computational Intelligence, Honolulu, Hawaii, USA, April, 2007 pp. 53-58.
[8] Y. H. Pao, Adaptive Pattern Recognition and Neural Networks, MA: Addison-Wesley, 1989.
[9] J. C. Patra, R. N. Pal, B. N. Chatterji, and G. Panda, “Identification of nonlinear dynamic systems using functional link artificial neural networks,” IEEE Trans. on Syst., Man, and Cybern., Apr. 1999, Vol. 29, No. 2, pp. 254-262.
[10] C. J. Lin and C. H. Chen, “Identification and prediction using recurrent compensatory neuro-fuzzy systems,” Fuzzy Sets and Systems, March 2005, Vol. 150, No. 2, pp. 307-330.
[11] C. J. Lin and C. C. Chin, “Prediction and identification using wavelet-based recurrent fuzzy neural networks,” IEEE Trans. on Systems, Man, and Cybernetics (Part:B), Oct. 2004, Vol. 34, No. 5, pp. 2144-2154.
[12] D. Nauck and R. Kruse, “A fuzzy neural network learning fuzzy control rules and membership functions by fuzzy error backpropagetion,” IEEE Int. Conf. Neural Networks, 1993, pp.1022-1027.
[13] J. S. Qu, C. Wang and Z. Z. Wang, “Structure-context based fuzzy neural network approach for automatic target detection,” IEEE Internation, Geoscience and Remote Sensing Symposium, IGARSS’03, Proceedings, July 2003, Vol. 2, pp. 767-769.
[14] W. A. Farag, V. H. Quintana, and G. Lambert-Torres, “A genetic-based neural fuzzy approach for modeling and control of dynamical systems,” IEEE Trans. on Neural Networks, 1998, Vol. 9, No. 5, pp. 756-767.
[15] C. J. Lin and Y. J. Xu, “A self-adaptive neural fuzzy network with group-based symbiotic evolution and its prediction applications”, Fuzzy Sets and Systems, Apr. 2006, Vol. 157, No. 8, pp. 1036-1056.
[16] X. Q. Zuo, S. Y. Li and X. J. Ban, “An immunity-based optimization algorithm for tuning neural fuzzy controller,” Proceedings of the Second International Conference on Machine Learning and Cybernetics, Nov. 2003, Vol.2, pp.666- 671.
[17] X. Wen and A. Song, “An immune evolutionary algorithm for sphericity error evaluation,” International Journal of Machine Tools & Manufacture, 2004, Vol. 44, pp. 1077-0184.
[18] A. Kalinli and N. Karabogab, “Artificial immune algorithm for IIR filter design,” Engineering Applications of Artificial Intelligence, 2005, Vol. 18, pp.919-929.
[19] J. Kennedy and R. Eberhart, “Particle swarm optimization,” Proc. IEEE Int. Conf. on Neural Networks, 1995, pp. 1942-1948.
[20] R. Storn and K. Price, “Differential evolution: a simple and efficient adaptive scheme for global optimization over continuous spaces”, Technical Report TR-95-012, International Computer Science Institute, Berkeley, USA, 1995.
[21] R. Storn, “System Design by Constraint Adaptation and Differential Evolution,” IEEE Trans. Evolutionary Computation, Apr. 1999, Vol. 3, no. 1, pp.22-34.
[22] T. Robiˇc and B. Filipiˇc, “DEMO: Differential Evolution for Multiobjective
Optimization,” Proceedings of the Third International Conference on Evolutionary Multi-Criterion Optimization – EMO 2005, ser. Lecture Notes in Computer Science, 2005, Vol. 3410, pp. 520–533.
[23] J. Brest, S. Greiner, B. Boskovic, M. Mernik and V. Zumer, “Self-adapting control parameters in differential evolution: a comparative study on numerical benchmark problems,” IEEE Transactions on Evolutionary Computation, 2006, Vol. 10. No. 6, pp. 646-657.
[24] J. T. Tsai, T. K. Liu, J. H. Chou, “Hybrid Taguchi-genetic algorithm for global numerical optimization”, IEEE Trans. on Evolutionary Computation, 2004, Vol. 8, No.4, pp.365~377.
[25] P. P. Angelov and D. P. Filev, “An approach to online identification of Takagi-Sugeno fuzzy models,” IEEE Trans. on Syst., Man, and Cybern., Feb. 2004, Vol. 34, No. 1, pp. 484-498.
[26] Z. L. Gaing, “A particle swarm optimization approach for optimum design of PID controller in AVR system,” IEEE Trans. on Energy Conversion, June 2004, Vol.19, No. 2, pp. 384-391.
[27] R. S. Cowder, “Predicting the Mackey-glass time series with cascade-correlation learning,” Proceedings of the 1990 Connectionist Models Summer School, 1990, pp.117-123.
[28] S. H. Ling, F. H. F. Leung, H. K. Lam, Y. S. Lee and P. K. S. Tam, “A novel genetic-algorithm-based neural network for short-term load forecasting,” IEEE Trans. Industrial Electornic., Aug. 2003, Vol. 50, No. 4, pp. 793-799.
[29] R. G. Reynolds, “An introduction to cultural algorithms,” Proc. 3rd Annual Conference on Evolutionary Programming, A.V. Sebald and L.J. Fogel (eds.), World Scientific, River Edge, NJ, USA, 1994, pp. 131- 139.
[30] X. Jin and R. G. Reynolds, “Using knowledge-based evolutionary computation to solve nonlinear constraint optimization problems: a cultural algorithm approach,” Proc. IEEE Congress on Evolutionary Computation, Washington, DC, USA, 1999, pp. 1672-1678.
[31] D. E. Goldberg. Genetic Algorithm in search, Optimization and Machine Learning. Addison-Wesley Publishing Company, Reading, Massachusetts, 1989.
[32] S. M. Saleem, Knowledge-based solution to dynamic optimization problems using cultural algorithms, PhD thesis, Wayne State University, Detroit, Michigan, 2001.
[33] Feedback Instruments Limited http://www.fbk.com/.