Modeling a non-ideal VGO hydrocracking reactor using selective overall effectiveness factors

Author(s): Sepehr Sadighi

The reported pilot experiments in the previous work were used to develop an optimizedmodel for a pilot scale hydrocracker of vacuumgas oil (VGO) under the following reaction conditions: temperature from380oC to 420oC, pressure of 156 bar, hydrogen-to-oil ratio of 1780 Nm3/Sm3 and liquid hourly space velocity (LHSV) ranging from0.5 to 2 hr-1.Acommercial dual functional amorphous catalyst was used in all experiments. The product of the reactor based on the most value added products was characterized to dry gas, light naphtha, heavy naphtha, kerosene, diesel and unconverted VGO. At first, a six-lump discrete lumping network including fifteen reactions was developed for the prediction of hydrocracking product yields. After applying the weighted lumping strategy, reducing the kinetic network and using selective effectiveness factors, the absolute average deviation (AAD%), R-squared and F-test of the model were reduced from 57.17%, 93.69%and 32.1 to 35.01%, 94.58%and 69.7, respectively. Then, to model the axial-dispersion effect through the catalytic bed, the N-tanks in series approach was used which could be shifted from plug ideal-flow regime (N=200) to completemixed flowregime (N=1), according to the Fibonacci series, to increase the accuracy of the model. Results confirmed that the best value for the number of series tanks, searched by Fibonacci golden numbers, was thirteen which reduced the final AAD%, R-squared and Ftest of the predicted yields to 31.59%, 95.15%and 73.6, respectively.Additionally, the estimated overall effectiveness factors by experimental data agree with the molecular diffusion and transport phenomena during hydrocracking process.

Share this