| Summary: | 博士 === 國立成功大學 === 電機工程學系碩博士班 === 92 === A complete study of the robust control of time-delay systems via the linear matrix inequality approach is presented in this dissertation. This includes the developments of the stability analysis/controller design/filtering problem/tracking control for a series of uncertain systems with time delay, two-dimensional (2-D) discrete state-delayed systems described by Roesser model, a class of stochastic systems with time delay and a class of neutral stochastic systems. A novel state transformation matrix has been proposed for the stability analysis. Based on the Lyapunov-Krasovskii functionals combining with linear matrix inequality (LMI) techniques, simple and improved delay-dependent robust stability criteria are derived. For the filtering design, asymptotically stabilizing Kalman filtering and filtering approaches are presented for the system with time delay. For the robust tracking control, the proposed tracker obtained in terms of LMIs guarantees the stability of closed-loop systems and makes the output to approach the command reference input with a specified performance. For the two-dimensional discrete state-delayed system, a 2-D dynamic output feedback controller is designed to achieve the closed-loop system asymptotic stability and a specified performance using the LMI approach. Also, we develop a 2-D filtering approach with a specified performance measure. For the stochastic system with time delay, a state feedback controller is designed to guarantee the closed-loop system remaining stable for all admissible uncertainties. Finally, we propose a state feedback controller to stabilize the class of neutral stochastic nonlinear systems with a specified performance.
|
|---|
