Two types of kinetic modeling, continuous and discrete lump model were studied and compared in this research in order to model the industrial scale VGO hydrocracking process. The experimental data obtained from a pilot-scale fixed bed reactor over Ni-Mo/Silica-Alumina catalyst in a wide range of operating conditions was used for prediction and tuning the kinetic parameters using Genetic algorithm. In this study, the discrete lump model with four parallel reactions to four lumped products showed more convergence to the experimental data than the continuous lump model. Afterward, the discrete kinetic model was used to simulate the vacuum gas oil (VGO) industrial hydrocracking reaction accompanied by catalyst deactivation. The activity of the catalyst was taken as a time dependent variable and the first year of operational data were used to derive the deactivation parameter. The refinery test runs spanning over the last two and half years of operation were used to validate the model and interpret the simulation results. A comparison between the industrial and the predicted data showed that there is a good agreement between them and the presented model provides a reasonable fit to estimate the product yields of LPG, naphtha, kerosene, diesel.
©2011 Walter de Gruyter GmbH & Co. KG, Berlin/Boston