Universal machine learning potential accelerates atomistic modeling of materials  

作　　者：Zhongheng Fu Dawei Zhang 

机构地区：[1]Beijing Advanced Innovation Center for Materials Genome Engineering,Institute for Advanced Materials and Technology,University of Science and Technology Beijing.Beijing 100083,China [2]National Materials Corrosion and Protection Data Center,University of Science and Technology Beijing,Beijing 100083,China

出　　处：《Journal of Energy Chemistry》2023年第8期1-2,I0002,共3页能源化学（英文版）

基　　金：supported by the National Natural Science Foundation of China (22209094)。

摘　　要：With the rapid development of computer techniques,atomistic modeling is playing an increasingly important role in understanding the structure-activity relationship of materials.Molecular dynamics (MD) is a computational simulation approach to predicting the structural evolution of an atomic system over time,widely used to understand physical and chemical phenomena including phase transition,diffusion,crystallization,and reaction [1].

关 键 词：MACHINELEARNING Atomisticmodeling Neural networkpotential Solid-statematerials 

分 类 号：TB30[一般工业技术—材料科学与工程] TP181[自动化与计算机技术—控制理论与控制工程]