Study Of Lumped Kinetic Modeling Of Aromatic Type Catalytic Naphtha Reforming

Catalytic reforming who aims for producing high octane value and clean gasoline, benzene, methylbenzene and dimethylbenzene is an important process in petroleum processing. Due to considerable by-product of hydrogen, especially in the nowadays where crude oil has deteriorated and hydrogenation has progressed rapidly, the position of catalytic reforming in refining factory has become more important. From the background of one specific catalytic reforming unit in China and on the basis of the actual industrial data, a lumped kinetic model for catalytic reforming is studied and built up in order to guide the practical operation and optimization of reforming catalytic unit.An entire literature review of domestic and overseas situation and progress of technologies, catalysts and lumped kinetic models of catalytic reforming was made. What's more, a deep field trip and study to the specific catalytic reforming unit as the object of this research was taken. The study shows, this catalytic reforming unit is a classical aromatic type catalytic naphtha reforming unit whose feedstock is heavy naphtha from hydrocracking with aims for producing basic organic raw materials such as benzene, methylbenzene and dimethylbenzene. This unit adopted the first generation technology of catalytic reforming developed by UOP corp. in 1971. Through the research of the characteristics of catalytic reforming reactions and analysis of reaction complexity and comparison of various kinetic models, the lumped kinetic modeling is appropriate for this research purpose.Consequently, classified the countless reactions by the principle of dividing lumps-the similarity of reaction characteristics. Given the sources of the actual industrial data and customer's requirements of model's functions of optimization and prediction, divided the reaction feedstock and products to 27 lumps. Then built up the reaction network between these 27 lumps based on the mechanism of catalytic reforming reaction and simplified the reaction network by some assumptions:1. all reactions were considered as one order homogeneous phase reaction;2. the impact of hydrogen on reaction was expressed by hydrogen pressure;3. Every infinitesimal of axial cross section along moving bed was considered as one ideal plug flow reactor;Considering the impact of catalysts inactivation caused by coke deposition, etc. at last listed the equations of 27 lumps of lumped kinetic modeling of aromatic type catalytic naphtha reforming-the reaction rate equation of power function type.On account of 163 parameters in the 27 lumps of lumped kinetic modeling of aromatic type catalytic naphtha reforming, evaluating them all one time was difficulty, and the credit of the evaluated parameters were low. So a method of fractional steps was applied which split up the 27 lumps model to three sub-models then evaluated the parameters of these three sub-models and at last estimated all the parameters of the whole 27 lumps model. At the mean while, after the comparison of several multi parameters estimation method, a combination of local minimizing optimization method (BFGS) which has numerical stability and rapid convergence and global minimizing optimization method (simulated annealing) was applied to accomplish the estimation of 163 model parameters.All the data requested by the parameters estimation came from one specific catalytic reforming unit in China. For verifying the parameters obtained by every step, used the data under different periods or operation conditions to verify the model's reliability and extrapolation. The results show that 27 lumped kinetic modeling of aromatic type catalytic naphtha reforming is able to predict the products precisely, and can predict the details of products such as the compositions of products, hydrogen, coke formation, etc.Additionally, the distribution of products, the profile of reaction temperature and coke formation along the catalyst bed was simulated by the 27 lumped kinetic modeling of aromatic type catalytic naphtha reforming. Based on the success of validity of 27 lumped kinetic modeling of aromatic type catalytic naphtha reforming, in the final discussed the impact of industrial operation conditions (temperature, space velocity, ratio of hydrogen and hydrocarbon, residence time of catalyst, etc.) on the distribution of products. Then optimized the yield of the primary object-aromatic. The result shows:under the optimized operation conditions, the yield of aromatic increases by 0.98% and the yield of dimethylbenzene increases by 1.34% with the yield of hydrogen increasing a little.This research has certain research object whose all data came from the specific industrial unit. And this research is objective, practical and significant. The research result has significance of directly guiding the specific catalytic reforming unit, and other aromatic type catalytic reforming units could find inspiration from this research.