Modeling And Simulation For Hydrocracking Reactor

With the growing demand for oil quality and the increasing production of heavy crude oils which contain sulphur in recent years, more and more attention has been focused on the hydrocracking process which has become one of the most important secondary petroleum refinery processes. Now hydrocracking unit is built in most new petroleum refinery item. Then the demands for the simulation optimization and simulation training of practical hydrocracking are increasing accordingly. Therefore, modeling and simulation for hydrocracking reactor is becoming increasingly important.Thus, modeling and simulation for the hydrocracking reactor of Sinopec Guangzhou Petrochemical Complex was studied in this paper.Firstly, the characteristics of hydrocracking reaction were studied and it was quite reasonable to assume that for all the cracking reactions that participated by the same lump, the reaction mechanism and activation energies are the same. Then by giving out the reaction rate equation of each reaction, a simplified kinetics model was proposed for the hydrocracking reaction, where the six-lump of inlet and outlet, including feedstock, diesel oil, aviation kerosene, heavy naphtha, light naphtha and gas were adopted based on the principle of lump for kinetics. Then, based on the analysis of characteristics of hydrocracking process, a six-lump dynamic mechanics model for hydrocracking reactor is proposed. The model which was composed of 42 partial differential equations and 5 algebra equations took both material balances and energy balances into consideration on the basis of kinetics model.Secondly, the reaction kinetics parameters of the lump model were regressed from the operation data in an industrial plant by Nelder-Mead minimization algorithm of Matlab 7.0. The minimization algorithm took the residual error between calculated value and measured value of production yields and outlet temperature as object function. Then, the estimated kinetic parameters were inserted into steady-state model, the calculation results are in a good agreement with the industrial data, so the model's reaction kinetics parameters are credible. Then, the data under some different steady-state operational conditions for the hydrocracking reactor were used to verify the model's steady-state prediction precision. The simulation is implemented by the ODE15s of MATLAB7.0. The results of prediction show the good prediction precision of the model, and the distribution of the yields and the temperatures along the reactor can also be simulated and predicted. Moreover, the time-consuming for one simulation is less than 1.5 seconds.Lastly, this paper took one operating mode as an example to simulate the dynamic impact of operating modes on the distribution of reactor temperature and production yields. The dynamic simulation is also implemented by MATLAB7.0. Because of the shortages such as long calculation time, big error and so on for those popular algorithms such as inite element method, finite difference method, etc., the partial differential equation model is solved by orthogonal collocation. The dynamic model simulation results show that the impact of the input variables (feed flux, feed temperature and cold hydrogen flux) on the raw material conversion, the production yields and the temperature in reactor can be better investigated.