臺灣博碩士論文加值系統

近年來，模糊規則式系統(fuzzy rule-based system)已經成功地應用在許多控制領域[2，，4]上。從理論的觀點來看，論文[11，17，20]已經證明模糊系統為任意非線性函數的萬能逼近器(universal approximators)基本上，一組語意式的控制規則(linguistic control rules)是構成模糊控制系統的主要部份，然而模糊規則庫的建立，卻是模糊控制器設計中最重要，也是最為困難的部份。大部份的模糊規則是透過專家的設計，亦即設計者本身必須是一個非常了解系統的專家，方能設計模糊規則庫，若設計者本身不是專家，則模糊規則的取得將會非常困難。

In recent years, fuzzy rule-based systems have been widely and successfully applied to many control domains [2, 3, 4]. From a theoretical point of view, fuzzy logic systems were proven as universal approximators of any nonlinear continuous function on a compact set to arbitrary accuracy [11, 17, 20]. Basically, a set of linguistic control rules is the kernel of a fuzzy inference system, which provides a approximate reasoning algorithm to convert expertise-oriented linguistic control strategies.