針對不確定非線性系統之適應式自我建構模糊類神經網路之研究

STUDY OF ADAPTIVE SELF-CONSTRUCTING FUZZY NEURAL NETWORK FOR UNCERTAIN NONLINEAR SYSTEMS

趙志家(Chih-Chia Chao)

指導教授 ： 呂虹慶

大同大學/電機資訊學院/電機工程學系所/碩士(2009年)

摘要

在這篇論文裡針對一種不確定非線性的系統提出了具有適應式自我建構模糊類神經網路控制器。所提出的適應式自我建構模糊類神經網路控制器是由適應式自我建構模糊類神經網路辨識器、計算控制器和強健控制器所組成。這辨識器是用來估測不確定非線性系統的參數。計算控制器主要設計用來計算的這辨識器的輸出。強健控制器的設計是為了補償系統參數的不確定性及不確定的外部干擾，以達成系統的強建穩定性。適應式自我建構模糊類神經網路辨識器採用結構學習和參數學習，來達到有效的近似效能和估測非線性系統的未知參數。在結構學習中利用馬氏距離來決定神經元的產生或消除與否。同時，參數學習中適應法則是基於里亞普諾夫推導而來,因此能夠保證系統的穩定度。最後,模擬結果及整合線性感應馬達和倒單擺架構的實驗執行，證實了所提出的適應式自我建構模糊類神經網路控制器之有效性。

關鍵字

模糊類神經網路 ； 估測器 ； 強健控制 ； 倒單擺

並列摘要

The adaptive self-constructing fuzzy neural network (ASCFNN) controller is proposed for the uncertain nonlinear systems in this thesis. The ASCFNN control system is composed of an ASCFNN identifier, a computation controller and a robust controller. The ASCFNN identifier is used to estimate parameters of the uncertain nonlinear system. The computation controller is designed to sum up the output of the ASCFNN identifier. The robust controller is designed to compensate the uncertainties of the system parameters and uncertain external disturbance, and achieve robust stability of the system. The structure and parameter learnings are adopted in the ASCFNN identifier to achieve favorable approximation performance. The Mahalanobis distance (M-distance) method in the structure learning is employed to determine if the fuzzy rules are generated/ eliminated or not. Concurrently, the adaptive laws are derived based on the sense of Lyapunov so that the stability of the system can be guaranteed. Finally, the simulation results and the experiment which integrates the linear induction motor (LIM) and an inverted pendulum (IP) are implemented to verify the effectiveness of the proposed ASCFNN controller.

並列關鍵字

fuzzy neural network ； estimator ； robust control ； inverted pendulum

參考文獻

[1] M. B. McFarland and A. J. Calise, ”Adaptive nonlinear control of agile antiair missiles using neural networks,” IEEE Transactions on Control Systems Technology, vol. 8, no. 5, pp. 749–756, September 2000.

[2] F. J. Lin, C. H. Lin, and P. H. Shen, “Self-constructing fuzzy neural network speed controller for permanent-magnet synchronous motor drive,” IEEE Transactions on Fuzzy Systems, vol. 9, no. 5, pp. 751–759, October 2001.

[3] H. Hu and P. Y. Woo, “Fuzzy supervisory sliding-mode and neural-network control for robotic manipulators,” IEEE Transactions on Industrial Electronics, vol. 53, no. 3, pp. 929–940, June 2006.

[5] F. J. Lin, W. J. Huang, and R. J. Wai, “A supervisory fuzzy neural network control system for tracking periodic inputs,” IEEE Transactions on Fuzzy Systems, vol. 1, pp. 41–52, July 1999.

[6] P. S. Sastry, G. Santharam, and K. P. Unnikrishnan, “Memory neuron networks for identification and control of dynamical systems,” IEEE Transactions on Neural Networks, vol. 2, pp. 306–319, May 1994.