上海大学学报(自然科学版) ›› 2024, Vol. 30 ›› Issue (5): 802-812.doi: 10.12066/j.issn.1007-2861.2603

• • 上一篇    下一篇

经典与机器学习原子间相互作用势的发展及应用进展

赵浩然1 , 沈 强2 , 王 鹏2   

  1. 1. 上海大学 材料基因组工程研究院, 上海 200444; 2. 上海大学 力学与工程科学学院, 上海 200444
  • 出版日期:2024-10-30 发布日期:2024-11-07
  • 通讯作者: 王 鹏 (1988—), 男, 副研究员, 博士, 研究方向为微纳米力学. E-mail:wangp@shu.edu.cn
  • 基金资助:
    国家自然科学基金资助项目 (12472106)

Development and application of classical and machine learning interatomic potential

ZHAO Haoran1 , SHEN Qiang2 , WANG Peng2   

  1. 1. Materials Genome Institute, Shanghai University, Shanghai 200444, China; 2. School of Mechanics and Engineering Science, Shanghai University, Shanghai 200444, China
  • Online:2024-10-30 Published:2024-11-07

摘要: 通过对几种经典和机器学习 (maching learning, ML) 势函数发展历史的回顾, 重点介绍了各类势函数在金属及共价键材料中的发展与应用等, 统筹分析了 ML 势函数与经典势函数的优缺点, 并对未来发展出更有效的原子间作用势的思路提出了展望.

关键词: 分子动力学, 原子间作用势, 机器学习, 变形机制

Abstract: This study reviewed the history of several classical and machine learning (ML) potential functions and focused on the recent developments and applications of these potential functions in metal and covalent-bond materials. A comprehensive analysis of the advantages and disadvantages of ML and traditional potential functions was provided and a perspective on the development of more effective interatomic potentials was offered.

Key words: molecular dynamics (MD), interatomic potential, machine learning (ML); deformation mechanism

中图分类号: