- 期刊
以系統化方法設計模糊控制器的模糊規則庫
A Systematic Approach for Designing Fuzzy Rule Base of Fuzzy Controller
摘要
模糊控制器已被廣泛的應用。然而，模糊控制規則的實現傳統上仍然是由繁雜的嘗試錯誤過程，並根據實際的經驗或一些實驗來實現模糊控制規則。本文提出利用直接型可適性模糊控制器來建構模糊控制器的模糊規則庫且在T-S模糊模型的形式裡不需受控系統的動態特性。選擇多組包含所有可控狀態空間範圍的初始輸入變數，針對完整的模糊規則集合，利用了不同的初始輸入變數來訓練直接型可適性模糊控制器，儲存最終的控制參數向量來建構模糊規則庫。因為此規則包含了所有可控狀態空間範圍，因此套用此模糊規則庫的模糊控制器不需再修改模糊規則即可達到不錯的逼近特性。從兩個系統的模擬結果可知，以建構的模糊規則集合可以達成在模糊控制設計的有效性。
並列摘要
The fuzzy control has been in a wide variety of applications. However, it is known that the fuzzy control rules are traditionally achieved by a tedious trial and error process and based on their experience or some experiments. This paper proposes that used direct adaptive fuzzy controller to construct the fuzzy rule base of fuzzy controller and does not require representation of the plant dynamics in the form of Takagi-Sugenos's fuzzy model. First, we choose many set of initial input variables from all of the controllable state space domain, and the different initial input variables is used to train the direct adaptive fuzzy controller for a complete fuzzy rule set, then stored the final parameter vector of controller to construct the fuzzy rule base. Because the rules contained all of the controllable state space domain, used this fuzzy controller of fuzzy rule base will never to modify the fuzzy rules and can achieve the approximate property very well. Simulation results from the two systems have shown, by constructing the fuzzy rule sets, the effectiveness of the proposed method on the fuzzy control design.