A Systematic Min–Max Optimization Design of Constrained Model Predictive Tracking Control for Industrial Processes against Uncertainty

Abstract

A systematic min-max optimization design of model predictive tracking control (MPC) for industrial processes under partial actuator uncertainty and unknown disturbances is proposed in this brief. A comprehensive state-space model with state variables and tracking error combined together and tuned separately is first proposed from the input-output process data; then, a subsequent min-max optimization-based MPC controller is designed, where two-step optimization is adopted to further enhance the system performance. In addition, the proposed min-max MPC offers more degrees of freedom to tune the closed-loop system performance and improved control performance will be obtained compared with conventional min-max MPC tracking approaches. Case studies on an injection molding process and a nonlinear batch reactor show the effectiveness of the proposed approach.