Accuracy, Performance, and Transferability of Interparticle Potentials for Al–Cu Alloys: Comparison of Embedded Atom and Deep Machine Learning Models

In several recent years, a significant progress has been made in atomistic simulation of materials, involving the application of machine learning methods to constructing classical interatomic interaction potentials. These potentials are many-body functions with a large number of variable parameters...

Full description

Saved in:

Bibliographic Details
Published in:	Journal of experimental and theoretical physics Vol. 137; no. 6; pp. 864 - 877
Main Authors:	Khazieva, E. O., Shchelkatchev, N. M., Tipeev, A. O., Ryltsev, R. E.
Format:	Journal Article
Language:	English
Published:	Moscow Pleiades Publishing 01-12-2023 Springer Nature B.V
Subjects:	Accuracy Aluminum base alloys Artificial neural networks Classical and Quantum Gravitation Computational efficiency Computing time Configurations Crystal defects Elementary Particles Embedded atom method Machine learning Melts (crystal growth) Neural networks Particle and Nuclear Physics Physics Physics and Astronomy Quantum Field Theory Relativity Theory Simulation Solid State Physics Solids and Liquids Transport properties
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	In several recent years, a significant progress has been made in atomistic simulation of materials, involving the application of machine learning methods to constructing classical interatomic interaction potentials. These potentials are many-body functions with a large number of variable parameters whose values are optimized with the use of energies and forces calculated for various atomic configurations by ab initio methods. In the present paper a machine learning potential is developed on the basis of deep neural networks (DP) for Al–Cu alloys, and the accuracy and performance of this potential is compared with the embedded atom potential. The analysis of the results obtained implies that the DP provides a sufficiently high accuracy of calculation of the structural, thermodynamic, and transport properties of Al–Cu alloys in both solid and liquid states over the entire range of compositions and a wide temperature interval. The accuracy of the embedded atom model (EAM) in calculating the same properties is noticeably lower on the whole. It is demonstrated that the application of the potentials based on neural networks to the simulation on modern graphic processors allows one to reach a computational efficiency on the same order of magnitude as those of the embedded atom calculations, which at least four orders of magnitude higher than the computational efficiency of ab initio calculations. The most important result is that about the possibility of application of DP parameterized with the use of configurations corresponding to melts and perfect crystals to the simulation of structural defects in crystals and interphase surfaces.
ISSN:	1063-7761 1090-6509
DOI:	10.1134/S1063776123120208