As an important petrochemical product,gasoline plays an irreplaceable role in the fields of economic production and national defense security.The molecular management technology of petroleum is gradually emerging.It recognizes and optimizes the petroleum processing process at the molecular level,which makes the development of refinery towards intelligence,integration and high efficiency become a hot research direction.In this paper,based on Matlab platform,from the molecular composition of gasoline,a mechanism-based prediction model of Reid vapor pressure and ASTM D86 distillation curve is developed.Compared with the traditional correlation model,the prediction result of the mechanism model is more accurate and universal,which can provide guidance for petroleum processing industry.Firstly,a phase equilibrium calculation engine for complex gasoline system is developed,and the activity-fugacity coefficient method is used to calculate the gas-liquid equilibrium.The calculation engine mainly includes a matrix module for calculating UNIFAC group coefficient,a thermodynamic property identification calling module for gasoline molecules,a bubble point and dew point calculation module and a liquid-liquid equilibrium calculation module.The model can calculate the phase equilibrium of gasoline system only by inputting the molecular composition information of gasoline.Compared with the calculation results of Aspen Plus,the relative errors of bubble point and dew point calculation module and liquid-liquid equilibrium calculation module are within 0.2%,0.8% and 1.5% respectively.Then,the prediction model of Reid vapor pressure based on e-T flash module is developed.E-T flash module is used to simulate the experimental measurement process of Reid vapor pressure.It is considered that the pressure of the system is Reid vapor pressure when the vaporization fraction is no longer changed.Firstly,consider the solubility of nitrogen in the oil sample in the experimental device.Then through the material balance analysis of flash evaporation,the vaporization fraction is expressed by the volume and density of the gas and liquid phase after flash evaporation,and the trial and error calculation of the vaporization fraction is carried out.The finished gasoline is used as the input of the model,and the relative errors between the predicted results of the model and the experimental values are all within 4%.Finally,the prediction model of ASTM D86 distillation curve is developed.The experimental device is simplified,and the functions of the evaporator,thermometer,condensation bath and measuring cylinder are kept.The batch distillation process of the evaporator is simulated by constantly removing the gas phase composition of the equilibrium bubble point,and a prediction model of the ASTM D86 distillation curve based on bubble point calculation is established.Then,considering the reflux of the inner wall of the distillation flask and the loss of light components before the initial distillation point,a prediction model of the ASTM D86 distillation curve based on the calculation of bubble dew point and the gas phase loss associated with the Reid vapor pressure is established.Finally,on this basis,the e-P flash module is used to simulate the reflux of the cooling pipe,and the prediction model of ASTM D86 distillation curve based on the mechanism of e-P flash is established.Using different types of actual gasoline to verify the accuracy of the model,the accuracy is high and the universality is good.Compared with the experimental values,the relative errors of the whole distillation range of gasoline are all within 3%. |