铝铜合金粒子间电位的准确性、性能和可转移性:嵌入式原子模型与深度机器学习模型的比较

IF 1 4区 物理与天体物理 Q3 PHYSICS, MULTIDISCIPLINARY
E. O. Khazieva, N. M. Shchelkatchev, A. O. Tipeev, R. E. Ryltsev
{"title":"铝铜合金粒子间电位的准确性、性能和可转移性:嵌入式原子模型与深度机器学习模型的比较","authors":"E. O. Khazieva,&nbsp;N. M. Shchelkatchev,&nbsp;A. O. Tipeev,&nbsp;R. E. Ryltsev","doi":"10.1134/S1063776123120208","DOIUrl":null,"url":null,"abstract":"<p>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 <i>ab initio</i> 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.</p>","PeriodicalId":629,"journal":{"name":"Journal of Experimental and Theoretical Physics","volume":null,"pages":null},"PeriodicalIF":1.0000,"publicationDate":"2024-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Accuracy, Performance, and Transferability of Interparticle Potentials for Al–Cu Alloys: Comparison of Embedded Atom and Deep Machine Learning Models\",\"authors\":\"E. O. Khazieva,&nbsp;N. M. Shchelkatchev,&nbsp;A. O. Tipeev,&nbsp;R. E. Ryltsev\",\"doi\":\"10.1134/S1063776123120208\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>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 <i>ab initio</i> 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.</p>\",\"PeriodicalId\":629,\"journal\":{\"name\":\"Journal of Experimental and Theoretical Physics\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.0000,\"publicationDate\":\"2024-01-26\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Experimental and Theoretical Physics\",\"FirstCategoryId\":\"101\",\"ListUrlMain\":\"https://link.springer.com/article/10.1134/S1063776123120208\",\"RegionNum\":4,\"RegionCategory\":\"物理与天体物理\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"PHYSICS, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Experimental and Theoretical Physics","FirstCategoryId":"101","ListUrlMain":"https://link.springer.com/article/10.1134/S1063776123120208","RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"PHYSICS, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

摘要

摘要 近年来,原子模拟材料取得了重大进展,其中包括应用机器学习方法构建经典原子间相互作用势。这些位势是具有大量可变参数的多体函数,其数值是通过使用ab initio方法计算出的各种原子构型的能量和力进行优化的。本文在深度神经网络(DP)的基础上为铝铜合金开发了一种机器学习势,并将这种势的精度和性能与嵌入式原子势进行了比较。对所获结果的分析表明,在整个成分范围和较宽的温度区间内,深度神经网络对铝铜合金在固态和液态下的结构、热力学和传输特性的计算提供了足够高的精度。而嵌入式原子模型(EAM)计算相同性质的精度总体上明显较低。研究表明,在现代图形处理器上应用基于神经网络的电位进行模拟,可以达到与嵌入式原子计算相同数量级的计算效率,比原子模型的计算效率至少高出四个数量级。最重要的结果是,利用与熔体和完美晶体相对应的构型参数化 DP 可以模拟晶体和相间表面的结构缺陷。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

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

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

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 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.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
1.90
自引率
9.10%
发文量
130
审稿时长
3-6 weeks
期刊介绍: Journal of Experimental and Theoretical Physics is one of the most influential physics research journals. Originally based on Russia, this international journal now welcomes manuscripts from all countries in the English or Russian language. It publishes original papers on fundamental theoretical and experimental research in all fields of physics: from solids and liquids to elementary particles and astrophysics.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信