Nonlinear model predictive controller for the riser reactor of a FCC unit

A nonlinear dynamic model for the riser reactor of a fluid catalytic cracking (FCC) was established to simulate and analyze the dynamic process. The stochastic noise associated with the feed-in process has been considered in the modeling. As the process behaves much of the steady nonlinear characteristics, a Hammerstein model is thus established accordingly. The paper presents a modification to the general predictive control strategy of the Hammerstein model, leading to the control equation directly while eliminating the need for calculating the nonlinear part of the Hammerstein model. The simulation result for the FCC riser reactor shows the effectiveness of the proposed method at both set point tracking and of the disturbances rejection.