Real-Time Implementation of Improved State-Space MPC for Air Supply in a Coke Furnace

In an industrial coke process, the dynamic relationship between the input and output devices is complicated. Since performance is closely related to the control used, improved control alternatives are necessary. This paper deals with the oxygen content control loop and proposes an improved state-space model predictive control (MPC) aimed at satisfying steady air supply into the process. The details of the proposed MPC are first described and tested on a typical example and then implemented on an industrial coke furnace. Simulation examples and real-time implementation results both show the effectiveness of the proposed scheme.