2008
EMSS 2008 Proceeding
Data driven adaptive model predictive control with constraint
Authors: Norhaliza Abdul Wahab, Jonas Balderud, Reza Katebi
Abstract
This paper proposes a Direct Adaptive Model Predictive Controller (DAMPC) with constraints that employs subspace identification techniques to directly identify and implement the controller. The direct identification of controller parameters is desired to reduce the design effort and computational load. The DAMPC method requires a single QR-decomposition for obtaining the controller parameters and uses a receding horizon approach to collect input-output data needed for the controller identification. The paper studies the effect of different horizon schemes, the stability robustness and compares the performance of the proposed control scheme when applied to a nonlinear process with that of a linear model predictive control scheme.
I3M Scientific Sponsors
I3M Industrial Sponsors
I3M Media Sponsors
