Molecular Modeling Of Composition Based On Detailed Evaluation Data Of Crude Oil

In chemical process design,process simulation is a crucial step.With the increasing precision and range of simulation,and the need to accurately predict the nature and composition of products,the traditional lumped model can’t meet many requirements,so the development of molecular level model is particularly critical.Based on the hybrid frame of Structural Unit and Bond-Electron Matrix(SU-BEM),the mathematical expression of oil and the generation of molecular set are presented in this paper,based on the macroscopic properties,the composition and content of the characteristic molecules in different distillate oils are calculated by inverse algorithm,and the optimization method in the process of construction is explored.The first step of the paper’s work is how to optimize the model by selecting the right combination of Probability Density Functions(PDF)to describe the composition of oil molecules.Taking straight-run gasoline as an example,this paper compares the difference between the predicted results and the experimental results of its composition retrieved by macro-properties under different combination of distribution functions,among which the simulation results based on structure constraint function combination are the best,and select it as the function combination of the subsequent model construction.Then,taking diesel as an example,the molecular composition model of diesel was established,and the predicted results were compared with the experimental results,which proved that the optimized model could be extended to heavier fractions.Based on this,the molecular level modeling of diesel mixing process is also carried out.The predicted values of the model are in good agreement with the experimental values,and the detailed information on the molecular distribution and macroscopic properties of the oil is obtained,the accuracy and applicability of the molecular level mixing model are proved.Finally,based on the detailed evaluation data of crude oil,the representative molecular core structures of crude oil fractions and crude oil,including gasoline,diesel,vacuum gas oil and vacuum residuum were determined,the pre-defined molecular sets of fractions and crude oils were constructed.Taking Saudi Arabian Light crude oil as an example,the molecular composition models of the fractions of sand light crude oil were constructed respectively,and then the crude composition models were obtained by mixing,the predicted values of the model properties are in good agreement with the experimental results of the crude oil detailed evaluation data,indicating that the crude oil molecular composition model constructed can represent the crude oil,and the information of boiling point distribution and carbon number distribution are also obtained,the correctness and validity of the method are verified.