滑順模態與線性矩陣不等式理論在磁浮系統控制上的應用

Abstract

本篇論文強調以線性矩陣不等式理論為基礎的順滑模態控制器設計 ，應用於磁浮 系統實例。於設計順滑模態控制器時，極點分佈區域之限 制條件是以線性矩陣不等式 來表示。對於系統參數的變動，若是匹配的 ，順滑模態控制器本身具有抑制的能力， 但是對於不匹配的參數變動或 是干擾則不具備此能力。本論文所提的方法著眼於設計 一個狀態回授控 制器，既使有不匹配但是有上下限的參數估計誤差或是未知的參數如 負 載變動，仍能使系統的極點限制於一個 LMI的穩定區域。 This thesis addresses the design of an LMI-based sliding- mode control algorithm applied to control of hybrid magnetic levitation system. The pole placement constraints in designing a sliding-mode controller are expressed in terms of linear matrix inequality (LMI). Sliding-mode controllers have the ability to resist certain matching uncertainties. The proposed method in this thesis is focusing on shifting the poles of the system into the desired LMI stable region by using state- feedback control even though the system contains mismatching but bounded uncertainties due to modeling error or payload variation.