This paper extends the range of Generalized Predictive Control (GPC) algorithm proposed by Clark et al. based on a CARIMA model, and shows that it is also applicable for an ARMAX model and a CARMA model. Unified formulas are given to provide direct approach to GPC using model parameters without solving Diophantine equations. The extended GPC is applied for the control of a pneumatic actuator based on its ARX model built by neural network. The performance of designed GPC controller is impressive for its quick response and accurate tracking.
Citation:
Qiang Song, Fang Liu, "The Direct Approach to Unified GPC Based on ARMAX/CARIMA/CARMA Model and Application for Pneumatic Actuator Control," icicic, vol. 1, pp.336-339, First International Conference on Innovative Computing, Information and Control - Volume I (ICICIC'06), 2006