Artificial Neural Network and Principal Component Analysis Study of Excess Molar Volumes and Excess Molar Enthalpies in Ionic Liquid Mixtures

This paper applies the model including back-propagation network (BPN) and principal component analysis (PCA) to estimate the excess molar volume and excess enthalpy of ionic liquid mixtures. The PCA was coupled with the BPN to optimize the BPN's parameters and improve the accuracy of proposed model. The excess molar volume and excess enthalpy of ionic liquid mixtures are examined as a function of the temperature (T), mole fractions of compounds (x(1) and x(2)), molar mass of pure ionic liquids (M-1 and M-2) and total molar mass (M-w) using artificial neural network. The obtained results by means of PCA-BPN model for excess molar volume and excess enthalpy have good agreement with the experimental data and absolute average deviations are 1.57 and 0.98%, respectively. Also, high coefficient of determination for excess molar volume and excess enthalpy are R-2 = 0.9983 and 0.9999, respectively.