Observer-based Variable Structure Control for Mismatched Uncertain Systems and Time-Varying Delayed Systems with Unknown Time Delay

• Main Authors: NGUYEN CONG TRANG, 阮公壯
• Other Authors: Tsai, Yao-Wen
• Format: Others
• Language: en_US
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/z99963

博士 === 大葉大學 === 機械與自動化工程學系 === 105 === By using reduced-order observer (ROO) tool, which will estimate an un-measurable state variable, a novel finite time output feedback sliding mode controller (FTSMC) is proposed for both the matched and mismatched uncertain systems such that the state trajectories of the system reach the sliding surface in finite time and stay on its thereafter. Further, a new condition in terms of linear matrix inequalities (LMI) is derived by employing the Lyapunov stability theory such that the dynamic of the system in sliding mode is asymptotically stable. The developed variable structure control (VSC) theory, which includes the following important issues: 1.Two methods will be showed in this work for both matched and mismatched uncertain systems: The first method is the well-known eigenstructure assignment (ESA) methodology. The second is the Moore-Penrose inverse technique, which will be eliminated a limitation of the first method. 2.A suitable ROO is constructed to estimate unmeasured variables via sliding mode control technique. 3.A novel FTSMC, based on the estimated variables and output variables to stabilize a class of dynamic systems with both matched and mismatched uncertainties in finite reaching time, has been proposed. 4.A newly appropriate linear matrix inequality (LMI) condition is derived such that the system in the new sliding mode is not only completely invariant to matched uncertainties in finite time but also asymptotically stable. 5.A new Lemma is established for the aim of stable observer. 6.The methods will apply to the matched and mismatched uncertain systems with/without time delay and the large-scale systems. Keywords: variable structure control, reduced-order observer, linear matrix inequality, finite-time convergence, mismatched uncertainty.