High-Precision Prediction of Thermal Conductivity of Metals by Molecular Dynamics Simulation in Combination with Machine Learning Approach

Molecular dynamics (MD) simulation is a powerful tool t o estimate materials properties from atom istic viewpoint. However, the scope of application of MD simulations is limited to problems where the Newton's equation of motion for atoms i s dominant. Therefore, i t i s inherently insufficient to estimate thermal conductivity of metallic materials, which consists of phonon ic and electronic components. In this study, machine learning (ML) based regression model is employed to predict thermal conductivity of metals with high accuracy using deficient results from M D simulations. A regression analysis with the least absolute shrinkage and selection operator (Lasso) including electrical conductivity as predictor variables successfully predict the thermal conductivity of metals with negative temperature dependence, which indicates a significant contribution of electrons to thermal conduction in metals. It should be stressed that our prediction is better than the conventional estimation from the Wie dema nn - Fr a nz law. This study shows us new possibilities of new ML approach for improving the accuracy of physical properties obtained from M D simulations.