Title :
Generalized predictive control of multivariable linear systems
Author_Institution :
Universit?? Libre de Bruxelles, Bruxelles, Belgium
Abstract :
In this paper, the generalized predictive control (GPC) algorithm developed by Clarke {2} is extended in order to control multiple-input-multiple-output (MIMO) linear systems. This algorithm is based on the minimization of a multistep criterion involving the future output errors and future control signals. It requires the design of a long range predictor for the controlled process. The theory is developed for a CARIMA model and for the corresponding state space description in innovation form. This last approach is used to analyse the closed-loop system obtained with the GPC.
Keywords :
Control systems; Error correction; Linear systems; MIMO; Minimization methods; Prediction algorithms; Predictive control; Process control; State-space methods; Technological innovation;
Conference_Titel :
Decision and Control, 1987. 26th IEEE Conference on
Conference_Location :
Los Angeles, California, USA
DOI :
10.1109/CDC.1987.272612
