區間第二類模糊類神經網路控制器與其在馬達上之應用

Interval Type-2 Fuzzy Neural network Controller and Its Application in DC Motors

潘建宏

指導教授 ： 洪欽銘 ； 王偉彥

國立臺灣師範大學/科技學院/工業教育學系/碩士(2010年)

摘要

在論文中，提出一個結合區間第二類模糊類神經網路的倒階控制器，並分別針對典型非線性系統和非典型非線性系統來做控制器的設計。主要的控制器為區間第二類模糊類神經網路近似器，其設計可以調整內部參數，包括平均值、標準差等，為了線上調整這些內部參數，本文設計適應律來調整，並使用均值定理的方法來取代傳統的泰勒線性化展開，雖然泰勒線性化展開可以將非線性的函數轉換成部分線性形式，但是會導致高階微分項帶入到近似誤差的模型裡，為了避免產生高階微分項的問題，所以使用均值定理來取代。閉迴路系統的穩定性也可以透過李亞普諾夫方程式來分析說明，以保證該系統是漸近穩定的。最後以模擬結果，來論證本文所提出的方法在受控體的輸出及想要的參考訊號兩者之間的追蹤誤差可以達到較好的追蹤效能。

關鍵字

區間第二類模糊 ； 適應控制 ； 倒階控制 ； 模糊類神經 ； 非線性控制

並列摘要

In this thesis, an adaptive backstepping interval Type-2 fuzzy neural network (IT2FNN) controller is proposed for a class of nonlinear system. We designed the controllers for affine and nonaffine nonlinear systems, respectively. The IT2FNN identifier is the main controller. The design of the controller can adjust its inside parameters, including mean and standard deviation. In order to adjust these parameters, we use adaptive law. We also use mean value theory to replace Taylor linearization expansion. Although Taylor linearization expansion, which can transform the nonlinear function into partially linear form. However, the linearization expansion method results in the fact that the higher-order derivative terms introduced into approximation model may produce the unpredictable and unfavorable influence on control performance. In addition, the stability of the closed-loop system is analyzed by mean of Lyapuniv function. Finally, simulation results use one example to demonstrate the output tracking error between the plant output and the desired reference command can achieve favorable tracking performance of the proposed scheme.

參考文獻

[3] W. Y. Wang, Y.H. Chien, and I.H. Li, “An On-Line Robust and Adaptive T-S Fuzzy-Neural Controller for More General Unknown Systems,” International Journal of Fuzzy Systems, vol. 10, no. 1, pp. 33-43, 2008.

[4] C. T. Lin, and L. Siana, “An Efficient Human Detection System Using Adaptive Neural Fuzzy Networks,” International Journal of Fuzzy Systems, vol. 10, no. 3, pp. 150-160, 2008.

[1] K. Hornik, M. Stinchcombe, and H. White, “Multilayer feedforward networks are universal approximators,” Neural Networks, no. 2, pp. 359-366, 1989.

[5] S. Wu, M. J. Er, and Y. Gao, “A fast approach for automatic generation of fuzzy rules by generalized dynamic fuzzy neural networks,” IEEE Transactions on Fuzzy Systems, vol. 9, pp.578-594, 2001.

[7] L. X. Wang and J. M. Mendel, “Fuzzy basis functions, universal approximation, and orthogonal least squares learning,” IEEE Trans. Neural Netw., vol. 3, no. 5, pp. 807–814, Sep. 1992.