Predicting the coefficients of the Daubert and Danner correlation using a neural network model

In the present research, three different architectures were investigated to predict the coefficients of the Daubert and Danner equation for calculation of saturated liquid density. The first architecture with 4 network input parameters including critical temperature, critical pressure, critical volume and molecular weight, the second architecture with 6 network input parameters including the ones in the first architecture with acentric factor and compressibility factor. The third architecture contains 12 network input parameters including 6 input parameters of the second architecture and 6 structural functional groups of different hydrocarbons. The three different architectures were trained and tested with the 160 sets of Daubert and Danner coefficients gathered from the literature. The trained neural networks were also applied to 15 un-known hydrocarbons and the outputs (Daubert and Danner coefficients) were used to predict the saturated liquid densities. The calculated liquid densities were compared with the experimental values. The Results indicated that the coefficients obtained from the second architecture produced more precise values for the liquid densities of the 15 selected hydrocarbons.