Simultaneous prediction of the thermodynamic properties of aqueous solution of ethylene glycol monoethyl ether using artificial neural network

Abstract

In this work, an artificial neural network (ANN) model for simultaneous prediction of the density, viscosity, thermal expansion coefficient, excess molar volume, and viscosity deviation of the aqueous solution of ethylene glycol monoethyl ether (EGMEE) covering the whole mole fractions at temperatures from 293.15 to 333.15K under atmospheric pressure has been used. A total of 75 data points of thermodynamic properties at several temperatures and mole fractions, have been applied to train and test the model. This study reveals that the ANN model shows an excellent alternative for simultaneous prediction of the thermodynamic properties of the aqueous solution of EGMEE with mean square error (MSE<0.0051%) and high coefficient of determination (R2≥0.9913). The results also represent that the ANN model performs better than the Redlich–Kister type equations for estimating of the thermodynamic properties of the aqueous solution of EGMEE with the overall improvement of 99% for more cases. Comparison between the ANN model results and those of achieving from some previous methods indicates that this work can prepare a simple method for simultaneous prediction of the thermodynamic properties of the aqueous solution of EGMEE in a better accord with experimental data.