Model predictive control with state estimation for heavy oil hydroprocessing in a slurry-phase reactor Academic Article in Scopus uri icon

abstract

  • A Model Predictive Control (MPC) strategy combined with a state estimator is presented for heavy oil hydroprocessing in a slurry-phase reactor. The state estimator is based on the Extended Kalman Filter (EKF) algorithm for nonlinear systems and accounts for uncertainties in the model parameters and measurement noise. The system states are estimated using only temperature measurements and the reactor model, while the control strategy stabilizes the system at the desired set point and minimizes the energy consumption within the reactor. The reactor and kinetic models are taken from the literature and describe the axial dispersion of the hydrocracking of Mexican heavy crude oil (11.97¿ API) based on a lumping approach with four pseudo-components (vacuum residue, vacuum gasoil, middle distillates, and naphtha) and SARA fractions (saturates, aromatics, resins, and asphaltenes) in the presence of a mineral catalyst. A traditional PI controller is introduced to compare with the MPC performance. Simulations are conducted under input and set point disturbances while addressing uncertainties arising from process estimations or measurement devices. Furthermore, the proposed approach showed improved performance in maintaining the temperature at the desired set points while avoiding the energy oscillations observed with the PI controller, making it suitable for real-time implementation in industrial settings. The improved estimation of process states allows for better decision-making, potentially reducing hydrogen consumption and optimizing energy usage. These results highlight the potential of MPC-based strategies in advancing digitalization and automation in hydroprocessing units, aligning with Industry 4.0 principles. © 2025 Elsevier Ltd.

publication date

  • May 15, 2026

published in