以個人電腦為基礎之模糊控制磁浮軸承系統

Abstract

近幾年來, 有關磁浮的技術被廣泛的發展並已應用到現代化的工業中, 例 如磁浮軸承、磁浮運輸系統等等. 由於磁浮系統的幾項特性, 諸如低噪音 、低功率消耗（無摩擦）、高速、潤滑問題的減少等, 使得人們開始注意 到此技術並產生極大的興趣, 更進一步從事學術上的研究以及工業上的實 際應用.本論文主要研究的是可控永磁式電磁鐵在磁浮軸承系統上的應用. 從磁浮原理, 機械結構, 終至於控制器的設計與實現, 本論文提出一有系 統的設計程序,並將研究重點擺在『近零功率消耗』及『準確的位置控制 』上, 最後並計劃使磁浮轉軸能在高速中穩定運轉.在控制技術上, 首先 先以一傳統的PID控制器作模擬, 並將其實現到真實系統中. 本論文的控 制器是以模糊控制器（FUZZY Controller）為設計的重點, 並利用遺傳演 算法（Genetic Algorithm）於歸屬函數及模糊規則參數的選取, 以進一 步改善控制性能. 最後並將以上所提透過一部586個人電腦及一塊12-bits 之AD/DA介面卡實現於真實的磁浮軸承系統上. Magnetic levitation(Maglev) technology has been largely developed and applied to modern industry in recent years, suchas magnetic bearing system and transport application. Possessingthe qualities of noiselessness, low power wastage, high speed andsafety, etc. It has prevailed the field for people's attention and interests in developing and researching. This thesis presents an investigation of the controlled-PM electromagnet Maglev system for bearing application. A systematic design procedure is developed to achieve successful integration ofall the subsystems including magnetic suspension, mechanical structure, and controller designing. The research goal focuses on an active magnetic bearing(AMB) with extremely zero-power consumption and well gap tracking, and finally rotating in high speeds. In control techniques, a traditional PID cintroller is firstused in simulation and applied to the magnetic bearing system in practice.For comparison, this thesis focuses on a FUZZY control technique, and thenapplied genetic algorithm(GA) on it for performance improvement. All of above is finally based on a 586-PC(personal computer) with a 12-bits AD/DAcard in practice.