Viskositas: Viscosity Prediction of Multicomponent Chemical Systems

TL;DR

This paper introduces Viskositas, a neural network model for predicting viscosity in multicomponent chemical systems, offering more accurate and reliable results than existing models, with potential benefits for industry and geophysics.

Contribution

The paper presents a novel neural network model for viscosity prediction that outperforms existing models in accuracy and reliability, based on a new database of chemical systems and temperatures.

Findings

Viskositas achieved lower mean absolute error than competing models.

The model demonstrated less variability and fewer outliers.

It provided more reliable viscosity predictions across diverse chemical systems.

Abstract

Viscosity in the metallurgical and glass industry plays a fundamental role in its production processes, also in the area of geophysics. As its experimental measurement is financially expensive, also in terms of time, several mathematical models were built to provide viscosity results as a function of several variables, such as chemical composition and temperature, in linear and nonlinear models. A database was built in order to produce a nonlinear model by artificial neural networks by variation of hyperparameters to provide reliable predictions of viscosity in relation to chemical systems and temperatures. The model produced named Viskositas demonstrated better statistical evaluations of mean absolute error, standard deviation and coefficient of determination in relation to the test database when compared to different models from literature and 1 commercial model, offering predictions with lower errors, less variability and less generation of outliers.