Energy-free machine learning force field for aluminum

Abstract We used the machine learning technique of Li et al. (PRL 114, 2015) for molecular dynamics simulations. Atomic configurations were described by feature matrix based on internal vectors, and linear regression was used as a learning technique. We implemented this approach in the LAMMPS code....

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Ivan Kruglov, Oleg Sergeev, Alexey Yanilkin, Artem R. Oganov
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2017
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/13e9687800914490b751cb5d5f07ee43
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:13e9687800914490b751cb5d5f07ee43
record_format	dspace
spelling	oai:doaj.org-article:13e9687800914490b751cb5d5f07ee432021-12-02T12:31:48ZEnergy-free machine learning force field for aluminum10.1038/s41598-017-08455-32045-2322https://doaj.org/article/13e9687800914490b751cb5d5f07ee432017-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-017-08455-3https://doaj.org/toc/2045-2322Abstract We used the machine learning technique of Li et al. (PRL 114, 2015) for molecular dynamics simulations. Atomic configurations were described by feature matrix based on internal vectors, and linear regression was used as a learning technique. We implemented this approach in the LAMMPS code. The method was applied to crystalline and liquid aluminum and uranium at different temperatures and densities, and showed the highest accuracy among different published potentials. Phonon density of states, entropy and melting temperature of aluminum were calculated using this machine learning potential. The results are in excellent agreement with experimental data and results of full ab initio calculations.Ivan KruglovOleg SergeevAlexey YanilkinArtem R. OganovNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 7, Iss 1, Pp 1-7 (2017)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Ivan Kruglov Oleg Sergeev Alexey Yanilkin Artem R. Oganov Energy-free machine learning force field for aluminum
description	Abstract We used the machine learning technique of Li et al. (PRL 114, 2015) for molecular dynamics simulations. Atomic configurations were described by feature matrix based on internal vectors, and linear regression was used as a learning technique. We implemented this approach in the LAMMPS code. The method was applied to crystalline and liquid aluminum and uranium at different temperatures and densities, and showed the highest accuracy among different published potentials. Phonon density of states, entropy and melting temperature of aluminum were calculated using this machine learning potential. The results are in excellent agreement with experimental data and results of full ab initio calculations.
format	article
author	Ivan Kruglov Oleg Sergeev Alexey Yanilkin Artem R. Oganov
author_facet	Ivan Kruglov Oleg Sergeev Alexey Yanilkin Artem R. Oganov
author_sort	Ivan Kruglov
title	Energy-free machine learning force field for aluminum
title_short	Energy-free machine learning force field for aluminum
title_full	Energy-free machine learning force field for aluminum
title_fullStr	Energy-free machine learning force field for aluminum
title_full_unstemmed	Energy-free machine learning force field for aluminum
title_sort	energy-free machine learning force field for aluminum
publisher	Nature Portfolio
publishDate	2017
url	https://doaj.org/article/13e9687800914490b751cb5d5f07ee43
work_keys_str_mv	AT ivankruglov energyfreemachinelearningforcefieldforaluminum AT olegsergeev energyfreemachinelearningforcefieldforaluminum AT alexeyyanilkin energyfreemachinelearningforcefieldforaluminum AT artemroganov energyfreemachinelearningforcefieldforaluminum
_version_	1718394259971768320

Energy-free machine learning force field for aluminum

Ejemplares similares