Machine learning for glass science and engineering: A review

Liu, Han; Fu, Zipeng; Yang, Kai; Xu, Xinyi; Bauchy, Mathieu

doi:10.1016/j.nocx.2019.100036

Cited by 47 publications

(40 citation statements)

References 78 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Machine learning algorithms We assess the performance of three archetypical ML algorithms (PR, ANN, and RF) as a function of the number of training data points. These methods are chosen as they belong to three distinct families of ML models, i.e., polynomial, network-based, and tree-based [25,26]. All the hyperparameters of the ML models considered herein were optimized in a previous study so as to achieve an optimal balance between under-and overfitting [16].…”

Section: 1mentioning

confidence: 99%

Using machine learning to predict concrete’s strength: learning from small datasets

Ouyang¹,

Shen²,

Li³

et al. 2021

Eng. Res. Express

Self Cite

View full text Add to dashboard Cite

Despite previous efforts to relate concrete proportioning and strength, a robust knowledgebased model for accurate concrete strength predictions is still lacking. As an alternative to physical or chemical-based models, machine learning (ML) methods offer a new solution to this problem. Although ML can handle the complex, non-linear, non-additive relationship between concrete mixture proportions and strength, it requires large datasets. This is a concern as reliable strength data is rather limited, especially for industrial concretes. Here, based on a large dataset (>10,000 observations) of measured compressive strengths from industrially-produced concretes, we compare the ability of select ML algorithms to "learn" how to reliably predict concrete strength as a function of the size of the dataset. Based on these results, we discuss the competition between how accurate a given model can eventually be (when trained on a large dataset) and how much data is actually required to train this model.

show abstract

Section: 1mentioning

confidence: 99%

Using machine learning to predict concrete’s strength: learning from small datasets

Ouyang¹,

Shen²,

Li³

et al. 2021

Eng. Res. Express

Self Cite

View full text Add to dashboard Cite

show abstract

“…With the help of MD simulation, we could calculate the constraint strength and bond angle distributions, which makes the modeling more powerful and precise. Besides MD simulation, machine learning offers a unique way to design glassy materials through quantitative structure‐property relationship (QSPR) . Combining machine learning with topological constraint theory is promising for future research.…”

Section: Conclusion Perspectives and Challengesmentioning

confidence: 99%

Progress in modeling of glass properties using topological constraint theory

Zheng

Zeng

2020

Int J of Appl Glass Sci

View full text Add to dashboard Cite

Prediction of glass properties as a function of composition has been a long-standing problem. Topological constraint theory has built the quantitative relationships between composition and macroscopic properties, which can be used to predict compositional dependence of glass properties. In this review, we will summarize the recent important progress for the development of the theory. The topological model for glass transition temperature has been proposed from temperature-dependent constraint theory. We will also discuss the effect of water on topological constraint, how to account for mixed alkali effect and mixed alkaline earth effect, and the application of the model to new glass family besides inorganic glasses. Topological models for hardness and thermal expansion have also been developed. Topological models for predicting elasticity have been recently proposed. We also show some example on the application of the theory in designing new glasses and we suggest some future directions of the theory.

show abstract

“…This has inspired recent efforts to model the viscosity of ionic liquids with neural networks (Paduszyński and Domańska, 2014;Beckner et al, 2018). For silicate melts, there have been several attempts to predict melt and glass properties using machine learning (see the reviews of Tandia et al, 2019;Liu et al, 2019), stretching back to the work of Dreyfus and Dreyfus (2003) on the prediction of liquidus temperature with artificial neural networks. Regarding melt viscosity, Hwang et al (2020) recently proposed an approach based on machine learning and the use of "cationic fingerprinting" to predict the temperatures of three reference viscosity points (10 1.5 , 10 6.6 and 10 12 Pa•s) for Na 2 O-SiO 2 -Al 2 O 3 -CaO melts with an error of ± 33 °C.…”

Section: Introductionmentioning

confidence: 99%

Structure and properties of alkali aluminosilicate glasses and melts: Insights from deep learning

Losq

Valentine

Mysen

et al. 2021

Geochimica et Cosmochimica Acta

View full text Add to dashboard Cite

The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.

show abstract

Machine learning for glass science and engineering: A review

Cited by 47 publications

References 78 publications

Using machine learning to predict concrete’s strength: learning from small datasets

Using machine learning to predict concrete’s strength: learning from small datasets

Progress in modeling of glass properties using topological constraint theory

Structure and properties of alkali aluminosilicate glasses and melts: Insights from deep learning

Contact Info

Product

Resources

About