Performance assessment of constrained model predictive control systems

Abstract

Abstract A constrained minimum variance controller is derived based on a moving horizon approach that explicitly accounts for hard constraints on process variables. A procedure for the performance assessment of constrained model predictive control systems is then developed based on the constrained minimum variance controller. The performance bound computed using the proposed moving horizon approach converges to the unconstrained minimum variance performance bound when the constraints on process variables become inactive. The utility of the proposed method in the performance assessment of constrained model predictive control systems is demonstrated through a simulated example.