Research On Hydrocracking Catalysts Grading Technology and Kinetic Model

Research on Hydrocracking Catalysts Grading Technology and Kinetic Model

ABSTRACT

Peng Chong,Fang Xiangchen,Zeng Ronghui

Fushun research institute of petroleum and petrochemicals,SINOPEC

In recent years, with the growing demands for clean fuels and the increasing production of heavy, inferior crude oils have imposed hydrocracking as one of the most important secondary petroleum refinery processes, which get more and more attention by refinery. The hydrocracking process is versatile, flexible and with strong adaptation to inferior feed but yielding high liquid yield and superior products commonly used to convert the heavy, inferior vacuum gas oil to smaller, lower-boiling ones like qualified chemical naphtha, kerosene, diesel, lubrication base oil and etc.. The production of more and more hydrocracking devices prompts the importance and urgency of hydrocrackers' simulation and optimization.

Hydrocracking takes place over a dual-function catalyst in a hydrogen-rich, high temperature atmosphere, and other reactions, such as hydrodesulfurization (HDS), hydrodenitrogenation (HDN), hydrodearomatization (HDA) and other hydrotreating reactions occur simultaneously. So the hytrotreating kinetic model should be established before the hydrocracking kinetic model. In this research, the hydrogenation of vacuum gas oil (VGO) on hydrotreated catalyst C was studied, and macro kinetic model of VGO hydrodesulfurization, kinetic model of VGO hydronitrogenation involving adsorption of nitrides and three lump kinetic model of VGO hydrodearomatization based on hydrocarbon composition (C_A, C_N, C_P ) were established respectively for total sulfur, total nitrogen and hydrocarbon composition. Through data analysis and kinetic model parameters optimization, the established hydrotreating kinetic model had a good ability to predict the total sulfur content, nitrogen content and hydrocarbon composition in hydrotreated oil.

HDS: 

HDN:

HDA: 

Based on previous studies, the high-pressure hydrocracking lumped kinetic model was established. The principle of lump division was as follows: the crude oil (feedstock and hydrogenated oil) was divided into six lumps according to the fixed distillation interval, namely as feedstock-hydrogenated residue oil, diesel fraction, kerosene fraction, heavy naphtha fraction, light naphtha fraction and refinery gas lump. In order to predict the hydrocracking product distribution, the six lumped kinetic model was established as follows: the kinetic model parameters were regressed by hydrocracking experiments at different conversion degree for two different hydrocracking catalysts, and kinetic model parameters mainly included exponential factor and activation energy. For the prediction of hydrocracking product properties, the function was established by product properties and cracking conversions, and the relative coefficient (R²) was high enough to predict the properties.

The lumped kinetic model parameters were regressed from the hydrocracking experiment data of two different catalysts by Fmincon nonlinear least squares algorithm in Matlab 2011b. The optimized model parameters were used as the initial value to predict the product distribution of hydrocracking catalysts grading experiment through modulating partial parameters. The similar method was used to predict the product properties. The results showed the calculation was in good aggrement with the experimental product yields and product properties so that it had strong guidance for industrial process.