Prioritized constraint handling NMPC using Volterra series models

This paper presents a nonsquare Nonlinear Model Predictive Control (NMPC) formulation that uses prioritized soft output constraints and hard input constraints. The proposed method uses empirical Volterra series models to represent the plant dynamics and is a closed loop tested on multivariable refinery facility simulation. The process exhibits significant open-loop nonlinearity such as input steady-state multiplicity and nonlinear dynamics near the nominal operating point. The results show improvement when compared with traditional linear control methods. Copyright © 2010 John Wiley & Sons, Ltd.