A stabilizing predictive controller with implicit feedforward compensation for stable and time-delayed systems

Abstract

Predictive controllers stand out as a realizable solution concerning disturbance rejection in comparison to classical feedforward compensators. However, stability attributes are still scarce in studies of predictive model controllers with implicit disturbance compensation. Therefore, this paper proposes a nominal stable model predictive control with implicit feedforward action for stable and time-delayed systems. The control formulation is presented based on the state-space model representation, considering the control inputs and measured disturbances increments. The control law results in a one-layer optimization problem providing optimal input moves that lead the process output to the desired reference rejecting the measured disturbances in a stable closed-loop response. The nominal stability properties are guaranteed by proving that the control cost function is a Lyapunov function. A solar collector field is used as a case study to show the disturbance rejection and overall performance characteristics of the proposed control strategy. Simulation results demonstrate the advantages and progress of the novel control algorithm and its capabilities to circumvent common issues that arise from disturbance rejection with classical feedforward structures.