Digital-Redesign Tracker for Unknown System with a Direct Feed-Through Term and Constraints: An Adaptive Mechanism for Tuning Weighting Matrices

摘要

This thesis proposes an OKID-based adaptive mechanism for tuning weighting matrices of muti-objective cost function: An application to the digital redesign tracker for the unknown system with a direct feed-through term and input/state/output constraints First the observer/Kalman filter identification (OKID) is used to identify the unknown and nonlinear system with a feed-through term into the equivalent mathematical model containing a feed-through term; this identified model is used for the design of the controller and observer By adding constraint terms into the linear quadratic performance index a new linear quadratic digital tracker is derived to track the desired reference and posses the effectiveness of constraints Further an OKID-based adaptive mechanism for tuning the weighting matrices is constructed in the cost function to update input state and output which exceed the limit bounds of saturation and make them update quickly and accurately Next the observer-based digital tracker is proposed for the system if the state is immeasurable Illustrative examples realize the effectiveness of the proposed design

獎項日期	2014 7月 29
原文	English
監督員	Jason Sheng-Hon Tsai (Supervisor)