APPARATUS AND METHOD FOR MODEL PREDICTIVE CONTROL (MPC) OF A NONLINEAR PROCESS

摘要

A method includes obtaining (402) a nonlinear process model (142) modeling a nonlinear process to be controlled. The method also includes obtaining (404) an objective function (144) defining how the process is controlled. The method further includes obtaining (406) a control model (146) defining a dynamic feasible region associated with a controlled variable, where the controlled variable is associated with the process. In addition, the method includes controlling (408-418) the nonlinear process by solving a control problem that includes the process model, control model, and objective function. The dynamic feasible region defined by the control model could represent a funnel region (302). The objective function could include terms for minimizing and optimizing adjustments to one or more manipulated variables associated with the process. Controlling the nonlinear process could include performing simultaneous control and optimization, where adjustments to the one or more manipulated variables are chosen to meet the control objectives and possibly to optimize and minimize the adjustments.