Applying machine learning techniques to predict the properties of energetic materials

Abstract

We present a proof of concept that machine learning techniques can be used to predict the properties of CNOHF energetic molecules from their molecular structures. We focus on a small but diverse dataset consisting of 109 molecular structures spread across ten compound classes. Up until now, candidate molecules for energetic materials have been screened using predictions from expensive quantum simulations and thermochemical codes. We present a comprehensive comparison of machine learning models and several molecular featurization methods - sum over bonds, custom descriptors, Coulomb matrices, Bag of Bonds, and fingerprints. The best featurization was sum over bonds (bond counting), and the best model was kernel ridge regression. Despite having a small data set, we obtain acceptable errors and Pearson correlations for the prediction of detonation pressure, detonation velocity, explosive energy, heat of formation, density, and other properties out of sample. By including another dataset with ≈300 additional molecules in our training we show how the error can be pushed lower, although the convergence with number of molecules is slow. Our work paves the way for future applications of machine learning in this domain, including automated lead generation and interpreting machine learning models to obtain novel chemical insights.

Prediction of Detonation Velocity and N−O Composition of High Energy C−H−N−O Explosives by Means of Artificial Neural Networks

“…In the recent years ANN technique has been used for prediction of impact sensitivity of high energetic materials [18][19][20][21][22]. Besides, other machine learning techniques also have been used for prediction of properties of high energetic materials from their molecular structure [23] An approach to use ANN technique to predict the detonation velocity had been previously attempted by Chen et al [24]. But in their model they have considered only chemical composition of CÀHÀNÀO for predicting detonation velocity.…”

Machine learning in materials science

“…High‐throughput density functional calculations for molecular property prediction are highly time‐consuming. As an alternative, machine learning is a feasible approach for the fast prediction of structures or properties of molecules, compounds and materials; in addition, it can realize high accuracy . ElemNet is a model that is based on a DNN that takes elements as input for predicting material properties .…”

“…As an alternative, machine learning is a feasible approach for the fast prediction of structures or properties of molecules, compounds and materials; in addition, it can realize high accuracy. 13,14,42,43,62,74,75,[119][120][121][122][123][124][125][126][127][128] ElemNet is a model that is based on a DNN that takes elements as input for predicting material properties. 42 It extracts the physical and chemical interactions and similarities between elements automatically and makes fast and precise predictions.…”

Machine learning: Accelerating materials development for energy storage and conversion

“…By using the model trained with the least absolute shrinkage and selection operator method, boron‐containing perovskites were selected due to their excellent performances in high electric environment, and the breakdown fields of two of these materials are ~ 2 GV/m, which means good feasibility for experiments. Considering the blank of the utilization of ML to energetic property prediction, Elton et al further broadened the application scope of ML methods in the prediction of detonation pressure, explosive energy and other energetic properties of molecular structures. Even in the case of a small dataset, errors of the results calculated through KRR model are within an acceptable range.…”