Modified Molecular Matrix Model For Predicting Molecular Composition Of Naphtha

To achieve maximum utilization of resources is an important development concept in the oil refining industry.Recently,stricter and ever increasing environmental legislations and demand for higher quality products are the driving forces for the refiner in an attempt to put more emphasis on molecules of refining products and characterize the refining streams on the molecular basis.Molecular management involves keeping the right molecules at the right place,at the right time,and at the right price.For naphtha,the paraffins in its composition can be used for hydrocracking to ethylene,and the naphthenes can be used to catalytically reformate aromatics.To accomplish the maximum utilization of naphtha,we should make management of naphtha processing at the molecular level.To fulfill this,molecular characterization of refining streams is the first step.Previous researchers developed a methodology to interrelate the bulk properties and molecular information of the refining streams based on MTHS(molecular type homologous series)matrix representation.The objective of this work is to enhance the accuracy and usability of the method,and establish a new MTHS matrix model for predicting the composition of naphtha.The main works were carried on as follows:(1)Build the new MTHS matrix to characterize naphtha composition.The new representation matrix consisted of 202 real molecules,and the corresponding properties matrixes are established by collecting 17 kinds of physical properties;(2)Put forward the molecular composition within each homologous series follows normal distribution against boiling point.According to the bulk properties of samples,using matlab software to build a prediction model.By forecasting naphtha and gasoline samples verify the normal distribution assumption have better applicability for prediction model than gamma distribution assumption;(3)Using new MTHS prediction model which have new molecular matrix and modified distribution assumption to predict a naphtha sample.The result shows new model can improved the prediction accuracy and have better forecast accuracy.