Simulations of vapor-liquid equilibrium (VLE) are widely used given their impact on the scale, design, and extrapolation of different operational units. However, due to a number of factors, it is almost impossible to experimentally study each of the VLE systems. VLE simulations can be developed using representations that are strongly dependent on the nature and interactions of the compounds forming mixtures. A model that helps in predicting these interactions would facilitate simulation processes. A Gray Box Neural Network Model (GNM) was created as Binary Interaction Parameters predictors (BIP), which are estimated using state variables and information from pure components. This information was used to predict VLE behavior in mixtures and ranges not used in the mathematical formulation. The GNM prediction capabilities (including temperature dependency) showed an error level lower than 5% and 20% for mixtures considered and not considered in the training data, respectively.Keywords: Acetone-Alcohol System; Peng-Robinson; Non-Linear Evaluations; ANN prediction. Modelamiento de equilibrio Líquido-Vapor usando redes neuronales grises como predictor de parámetros de interacción binariaResumen Las Simulaciones de Equilibrio Líquido Vapor (VLE) son ampliamente utilizadas dado su impacto en el escalamiento, diseño y extrapolación de diferentes operaciones unitarias. Sin embargo, dado considerable factores, es casi imposible experimentalmente estudiar cada uno de los sistemas de VLE. La simulación de VLE puede ser desarrollada utilizando representaciones que son fuertemente dependientes de la naturaleza e interacción de los compuestos que conforman la mezcla. Un modelo que ayude en la predicción de esas interacciones facilitará el proceso de simulación. Una Red Neuronal Gris (GNM) fue creada como un predictor de parámetros de interacción binaria, los que son estimados utilizando variables de estado e información de componentes puros. Esta información fue utilizada para predecir el comportamiento de VLE en mezclas y rangos no utilizados en la formulación matemática. Las capacidades predictivas del GNM (incluida la dependencia de temperatura) mostraron errores menores al 5% y al 20% para mezclas consideradas y no consideradas en los datos de entrenamiento, respectivamente.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.