Experience-based identification and model predictive control for a methanol recovery distillation column

In process industries, the plant testing method, which is used to obtain a model for the design of a MPC controller, is a kind of commonly used system identification method. However, this method has the disadvantages of time-consuming and production disruption. For this reason, in this paper, the bottom stage temperature control system of a methanol recovery distillation column is taken as the research object, and the model predictive effects based on step tests and experiences from operators are simulated and compared. Finally, a MPC controller based on an experience model is designed and put into operation. The actual plant application shows that the MPC controller could effectively improve the stability of the bottom stage temperature.