Machine Learning for the Prediction of Thermodynamic Properties in Amorphous Silicon DOI Creative Commons
Nicolás Amigó

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(10), P. 5574 - 5574

Published: May 16, 2025

This study integrated molecular dynamics (MD) simulations with machine learning techniques, specifically Linear, Ridge, and Support Vector Regression, to predict the thermodynamic properties of amorphous silicon (a-Si) under varying conditions. The MD provided a detailed dataset that captured atomic-level behavior a-Si, which enabled exploration how factors, such as cooling rate, temperature, pressure, affect material’s density, internal energy, enthalpy. Machine models were trained on this demonstrated exceptional predictive accuracy R2 values exceeded 0.95 minimal root-mean-square errors. results reveal temperature pressure significantly influenced while rate had minor effect. generated isobaric isothermal curves, offered deeper insights into a-Si complemented traditional by providing more efficient means explore states. work highlights potential accelerate materials enabling faster generation additional data. approach enhances understanding equation state opens new avenues for applying hybrid modeling technique other materials.

Language: Английский

Machine Learning for the Prediction of Thermodynamic Properties in Amorphous Silicon DOI Creative Commons
Nicolás Amigó

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(10), P. 5574 - 5574

Published: May 16, 2025

This study integrated molecular dynamics (MD) simulations with machine learning techniques, specifically Linear, Ridge, and Support Vector Regression, to predict the thermodynamic properties of amorphous silicon (a-Si) under varying conditions. The MD provided a detailed dataset that captured atomic-level behavior a-Si, which enabled exploration how factors, such as cooling rate, temperature, pressure, affect material’s density, internal energy, enthalpy. Machine models were trained on this demonstrated exceptional predictive accuracy R2 values exceeded 0.95 minimal root-mean-square errors. results reveal temperature pressure significantly influenced while rate had minor effect. generated isobaric isothermal curves, offered deeper insights into a-Si complemented traditional by providing more efficient means explore states. work highlights potential accelerate materials enabling faster generation additional data. approach enhances understanding equation state opens new avenues for applying hybrid modeling technique other materials.

Language: Английский

Citations

0