Research On The Kinetic Model For Residual Oil Hydroprocessing

With poorer quality of crude oil and the increasing demand for cleaning oil products,it is more and more urgent to convert heavy oil into light one.Hence,residue hydrogenation technology has being received extensive attention as one of the important means to obtain light oil products.In the residue hydrogenation process,the reaction conditions,different types of catalysts loading schemes,and feedstock properties all have effects on the final oil properties.Generally,the reaction parameters were set up through extensive experimentation and experience,which required a lot of manpower,time and money.However,by establishing a mathematical model,process parameters can be optimized objectively and the time and cost would be saved.Therefore,this paper has proposed a kinetic model of residue hydrodesulfurization(HDS)to describe the quantitative relationship between reaction conditions,catalyst deactivation,feedstock properties and HDS conversion of residual oil.The model reflects the basic characteristics of the residue HDS reaction and has high accuracy.At the same time,the model was constructed from the representative two-step catalytic reaction process,which makes the model have high universality and is suitable for other hydrogenation reactions with similar reaction process to residue HDS reaction.In addition,according to the modification based on this model,the grading model of the residue hydrotreating catalyst and heating model can be obtained,which provide a basis for optimizing the catalyst loading ratio,predicting the bed temperature and increasing rate.Firstly,based on the two-step reaction process of residue HDS and combined with the characteristics of residue hydrogenation reaction,the kinetic model of residue-catalyzed HDS was obtained by logical derivation under reasonable assumptions.The model is simple in form,calculated conveniently and the parameters have physical meaning.After using the model to fit the residual HDS reaction pilot data and industrial operation data,the average relative deviation(ARD)obtained were 1.83 8%and 0.523%respectively,indicating that the model has higher fitting accuracy.After determining the value of the parameter by data fitting,the model can calculate(predict)the HDS reaction conversion rate according to giving reaction conditions.Compared Predicted conversion rate with the pilot and industrial HDS operational data that did not participate in the fitting,the average relative deviations were 2.000%and 0.730%respectively,indicating that the model has higher prediction accuracy.Since the residue HDS reaction process is similar to most hydrogenation reactions,the model is also applicable to these hydrogenation reactions.In order to verify the universality of the model,the residue hydrodemetallization(HDM)pilot data,the residue hydrodehydration carbon(HDC)industrial data and the o-dichlorobenzene isomerization reaction data were substituted into the model.The average relative deviations of the calculated conversion rate and experimental conversion rate were 2.608%,2.792%and 2.042%respectively,indicating that the model has high application universality.Based on the residue hydrogenation catalytic model,the single catalyst volume was introduced into the model by series superposition and averaging key parameter to obtain the residue hydrogenation catalyst grading model.Among them,the key parameter averaging method is to average the two key parameters-the pre-exponential factor and the deactivation rate constant in the hydrogenation model.The average relative deviations of HDS,HDC and HDM model obtained by averaging key parameter were 1.202%,2.962%and 1.242%respectively,which shows the high accuracy and universality of the model.Through the above catalyst grading equation of each reaction and the constraint conditions,the optimal assembly ratio of the single catalyst can be obtained,therefore optimizing the catalysts filling scheme.Finally,in order to solve the problem of error amplification when using the grading model to draw the calibration-heating curve,the temperature-related term in the grading model obtained by averaging key parameter was transformed by non-Taylor expansion to get the residue hydrotreating catalyst grading heating model-an explicit function of the average temperature of the bed.Substituting the residual HDS experimental data into the model fitting,the obtained model parameters were reasonable,and the relative deviation of predicted bed average temperature(CAT)and the actual CAT average was only 0.246%,which proves the rationality of the model.Using the obtained equation to draw a standard heating curve can get the required CAT is 407 ? when the reaction was carried out for 8 months,which is consistent with the actual industrial device operation.