在本論文裡主要分成三大部分來討論，第一部份介紹了Mamdani模糊系統應用在模糊PI、PD、PI+PD控制器上，藉由專家知識庫的取得與應用，並且搭配了線上型的參數自我調整機制，使系統不僅達到穩定化的狀態而且符合我們要求的暫態響應，還可以使系統存在一定的強健性質，以及克服真實世界物理系統的時間延遲現象。模糊邏輯控制可以有效因應外擾或雜訊所導致系統參數變化的影響，受控系統可針對所遭遇的狀況來選擇最適合的設定排程，以達到最終的目的，因此特別適用於動態特性隨操作狀況改變的受控系統。 在第二部分介紹了T-S模糊系統應用在動態輸出回饋控制器上，藉由模糊集合的觀念，將非線性系統線性化之後，利用了李亞普諾夫函數，使系統達到穩定化的狀態和H∞的性能指標，克服系統外的雜訊干擾，並且搭配了線性矩陣不等式的數學工具，如此不僅可以有效率的求得最佳解之外還可以使系統達到我們要求的時間響應。 在第三部份裡則透過了電腦實驗模擬，包括了線性和非線性時延系統的響應控制和倒單擺的平衡控制來驗證所設計出的控制器是否符合當初所要求的目標響應。 In this thesis, mainly divide into three major parts and discuss, the first part is introduced Mamdani fuzzy system to be applied to fuzzy PI, PD, PI+PD controller. One that is by expert's knowledge base obtains and application, using the on-line self parameter adjusted make system not only stability but also achieve we request the transient response. The system has certain robustly, and overcome true world physics system’s time-delay. Fuzzy logic control can overcome noise effectively and the systematic parameter changes, especially suitable for the controlled system that the state changes. In the second part is introduced T-S fuzzy system to apply to the dynamic output feedback controller. We proposes a systematic design methodology for T-S model based fuzzy control systems with guaranteed H∞ performance and additional constraints on the closed-loop pole locations. These objectives are formulated in terms of a convex optimization problem involving linear matrix inequality. To demonstrate its usefulness, the proposed design methodology is applied to the problem of robust regulation of a inverted pendulum. Experiment and simulation results show that the proposed fuzzy controller yields the optimal disturbance rejection performance and the desired transient response characteristics.
