ViscNet: Neural network for predicting the fragility index and the temperature-dependency of viscosity

Abstract

Viscosity ($\eta$) is one of the most important properties of disordered matter. The temperature-dependence of viscosity is used to adjust process variables for glass-making, from melting to annealing. The aim of this work was to develop a physics-informed machine learning model capable of predicting $\eta (T)$ of oxide liquids. Instead of predicting the viscosity itself, the NN predicts the parameters of the MYEGA viscosity equation: the liquid's fragility index, the glass transition temperature, and the asymptotic viscosity. With these parameters, $\eta$ can be computed at any temperature of interest, with the advantage of good extrapolation capabilities inherent to the MYEGA equation. The dataset was collected from the SciGlass database; only oxide liquids with enough data points in the high and low viscosity regions were selected, resulting in a final dataset with 17,584 data points containing 847 different liquids. About 600 features were engineered from the liquids' chemical composition and 35 of these features were selected using a feature selection protocol. The hyperparameter (HP) tuning of the NN was performed in a set of experiments using both random search and Bayesian strategies, where a total of 700 HP sets were tested. The most successful HP sets were further tested using 10-fold cross-validation, and the one with the lowest average validation loss was selected as the best set. The final trained NN was tested with a test dataset of 85 liquids with different compositions than those used for training and validating the NN. The $R^2$ for the test dataset's prediction was 0.97. This work introduces three advantages: the model can predict viscosity as well as the liquids' glass transition temperature and fragility index; the model is designed and trained with a focus on extrapolation; finally, the model is available as free and open-source software licensed under the GPL3.

arXiv comment31 pages, 27 figures. Layout change. Added journal reference

References in corpus (11)

• Scikit-learn: Machine Learning in Python
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Decoupled Weight Decay Regularization
• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
• A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials
• Tune: A Research Platform for Distributed Model Selection and Training
• Predicting the dissolution kinetics of silicate glasses using machine learning
• Predicting Young's Modulus of Glasses with Sparse Datasets using Machine Learning
• $C_6$ coefficients and dipole polarizabilities for all atoms and many ions in rows 1-6 of the periodic table
• Designing optical glasses by machine learning coupled with a genetic algorithm

Cited by in corpus (6)

• Designing optical glasses by machine learning coupled with a genetic algorithm
• Predicting and interpreting oxide glass properties by machine learning using large datasets
• GlassNet: a multitask deep neural network for predicting many glass properties
• Machine learning modeling of the atomic structure and physical properties of alkali and alkaline-earth aluminosilicate glasses and melts
• Predicting Oxide Glass Properties with Low Complexity Neural Network and Physical and Chemical Descriptors
• Revealing the Compositional Control of Electrical, Mechanical, Optical, and Physical Properties of Inorganic Glasses