Accurate machine learning models based on small dataset of energetic materials through spatial matrix featurization methods

• 作者: Chao Chen Danyang Liu Siyan Deng Lixiang Zhong Serene Hay Yee Chan Shuzhou Li Huey Hoon Hng
• 来源期刊: 能源化学
• 年，卷（期）: 2021,（12）
• 页码范围: 364-375
• 页数: 12页
• 语种: 英文

摘要

A large database is desired for machine learning (ML) technology to make accurate predictions of mate-rials physicochemical properties based on their molecular structure.When a large database is not avail-able,the development of proper featurization method based on physicochemical nature of target proprieties can improve the predictive power of ML models with a smaller database.In this work,we show that two new featurization methods,volume occupation spatial matrix and heat contribution spa-tial matrix,can improve the accuracy in predicting energetic materials' crystal density (ρcrystal) and solid phase enthalpy of formation (Hf,solid) using a database containing 451 energetic molecules.Their mean absolute errors are reduced from 0.048 g/cm3 and 24.67 kcal/mol to 0.035 g/cm3 and 9.66 kcal/mol,respectively.By leave-one-out-cross-validation,the newly developed ML models can be used to deter-mine the performance of most kinds of energetic materials except cubanes.Our ML models are applied to predict ρcrystal and Hf,solid of CHON-based molecules of the 150 million sized PubChem database,and screened out 56 candidates with competitive detonation performance and reasonable chemical struc-tures.With further improvement in future,spatial matrices have the potential of becoming multifunc-tional ML simulation tools that could provide even better predictions in wider fields of materials science.