Tuning Guidelines of a Dynamic Matrix Controller for Integrating (Non-Self-Regulating) Processes

Designing a multivariable dynamic matrix conrtoller (DMC) controller for integrating processes is challenging because of the number of tuning parameters that affect the closed-loop performance. These tuning parameters required to implement DMC include the sample time; the prediction, model, and control horizons; the controlled variable weights; and the move suppression coefficients. The move suppression coefficients are used as the key tuning parameters to obtain a desirable DMC performance. This paper derives and demonstrates expressions for computing the complete set of tuning parameters for integrating processes. A novel contribution of this work is the derivation of an analytical expression for computing the move suppression coefficients based on the process model and the other DMC design parameters. The tuning rules are demonstrated on simulated processes including a constrained multivariable process simulation that displays integrating characteristics.