An approach for full space inverse materials design by combining universal machine learning potential,universal property model,and optimization algorithm

作     者：Guanjian Cheng Xin-Gao Gong Wan-Jian Yin 程观剑;龚新高;尹万健

作者机构：College of EnergySoochow Institute for Energy and Materials InnovationS(SIEMIS)and Jiangsu Provincial Key Laboratory for Advanced Carbon Materials and Wearable Energy TechnologiesSoochow UniversitySuzhou 215006China Shanghai Qi Zhi InstituteShanghai 200232China Key Laboratory for Computational Physical Sciences(MOE)Institute of Computational Physical SciencesFudan UniversityShanghai 200438China 

出 版 物：《Science Bulletin》 (科学通报（英文版）)

年 卷 期：2024年第69卷第19期

页      面：3066-3074页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0805[工学-材料科学与工程（可授工学、理学学位）] 080502[工学-材料学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：funding support by the National Key Research and Development Program of China(2020YFB1506400) the National Natural Science Foundation of China(11974257 and 12188101) Jiangsu Distinguished Young Talent Funding(BK20200003) Soochow Municipal Laboratory for low carbon technologies and industries 

主　　题：Inverse materials design Universal machine learning potential Graph neural networks Bayesian optimization 

摘      要：We present a full space inverse materials design(FSIMD)approach that fully automates the materials design for target physical properties without the need to provide the atomic composition,chemical stoichiometry,and crystal structure in ***,we used density functional theory reference data to train a universal machine learning potential(UPot)and transfer learning to train a universal bulk modulus model(UBmod).Both UPot and UBmod were able to cover materials systems composed of any element among 42 *** with optimization algorithm and enhanced sampling,the FSIMD approach is applied to find the materials with the largest cohesive energy and the largest bulk modulus,***-type ZrC was found to be the material with the largest cohesive *** bulk modulus,diamond was identified to have the largest *** FSIMD approach is also applied to design materials with other multi-objective properties with accuracy limited principally by the amount,reliability,and diversity of the training *** FSIMD approach provides a new way for inverse materials design with other functional properties for practical applications.