A hybrid model for the hydrotreatment of gasoil

In this work neural networks are used to describe hydrogenation and hydrodesulfurization reactions in a gasoil hydrotreatment model. The resulting model has been called hybrid model, because it unifies first principle modeling (material balance in the reactor) and black box models (the neural network describing the kinetics). The proposed approach permits obtaining a simplified model of the hydroprocessing, which allows an accurate description of the system for a wide range of operating conditions and feed composition.