遺傳演算模糊類神經與其在直流伺服馬達上之應用

Abstract

本文提出一個使用小型的遺傳演算法來調整模糊類神經網路的參數，並將其應用於函數近似與非線性系統之適應控制器設計。此小型的遺傳演算法應用於適應控制器設計，不需要事先離線學習的程序和複雜的數學運算。相較於傳統非線性系統的適應控制器，可有效減少適應控制器所需複雜的數學運算。在非線性系統之適應控制過程中，模糊類神經控制器的權重値是經由遺傳演算法來即時調整，以產生適當的控制輸入。為了即時評估閉迴路系統穩定的趨勢，本文從里亞布諾夫（Lyapunov）函數的穩定性分析推導過程中，提出一個能量適應函數於小型的遺傳最佳演算法中，藉此獲得較佳的閉迴路系統的穩定度。此外，由於小型的遺傳演算法可能在即時控制過程中使系統狀態進入不安全的區域。因此，加入安全控制器以限制閉迴路系統的狀態進入不安全的區域。 本文藉由電腦模擬結果驗證所提出方法的可行性與效能。最後，將此模糊類神經控制器應用在直流伺服馬達追蹤控制實驗。

In this thesis, a compact genetic algorithm used to tune the parameters of fuzzy neural networks is proposed for function approximation and adaptive control of nonlinear systems. For the design of adaptive controller, the compact genetic algorithm does not require the procedure of off-line learning and the complicated mathematical computation. Compared with traditional adaptive controllers, computation loading can be effectively alleviated. In adaptive control procedure for nonlinear systems, the weights of the fuzzy neural controller are online adjusted by the compact genetic algorithm in order to generate appropriate control input. For the purpose of on-line evaluating the stability of the closed-loop systems, an energy fitness function derived from Lyapunov function is involved in the compact genetic algorithm. In addition, the system states may go into the unsafe region if the compact genetic algorithm can not instantaneously generate the appropriate weights. In order to guarantee the stability of the closed-loop nonlinear system, a safe controller is incorporated into the fuzzy neural controller. Finally, some computer simulation examples and a servo motor experiment are provided to demonstrate the feasibility and effectiveness of the proposed method.