经典与机器学习原子间相互作用势的发展及应用进展  

Development and application of classical and machine learning interatomic potential

作　　者：赵浩然 沈强 王鹏[2] ZHAO Haoran;SHEN Qiang;WANG Peng(Materials Genome Institute,Shanghai University,Shanghai 200444,China;School of Mechanics and Engineering Science,Shanghai University,Shanghai 200444,China)

机构地区：[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　　处：《上海大学学报（自然科学版）》2024年第5期802-812,共11页Journal of Shanghai University:Natural Science Edition

基　　金：国家自然科学基金资助项目(12472106)。

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.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.

关 键 词：分子动力学 原子间作用势 机器学习 变形机制 

分 类 号：O344[理学—固体力学]