Analysis of Data on Vapor–Liquid Equilibrium in Multicomponent Systems Using Artificial Neural Networks

Abstract

Abstract: A brief analysis of the possibilities of using the method of artificial neural networks (ANNs) for assessing and correlating data on vapor–liquid equilibrium is presented. The advantages of the Focke method are considered in the case of a limited amount of data, in this case, the parameters of vapor–liquid equilibrium. Six binary and four ternary systems are considered using a modified Argatov–Kocherbitov technique. The estimation of the correctness of the ANN method is presented for the values of excess Gibbs energy calculated from the data on the vapor–liquid equilibrium.