A physics-aware deep learning model for energy localization in multiscale shock-to-detonation simulations of heterogeneous energetic materials

Phong C. H Nguyen; Yen‐Thi Nguyen; Pradeep K. Seshadri; Joseph B. Choi; H. S Udaykumar; Stephen Baek

doi:10.1002/prep.202200268

Back

A physics-aware deep learning model for energy localization in multiscale shock-to-detonation simulations of heterogeneous energetic materials

Journal article

Open access

Peer reviewed

A physics-aware deep learning model for energy localization in multiscale shock-to-detonation simulations of heterogeneous energetic materials

Phong C. H Nguyen, Yen‐Thi Nguyen, Pradeep K. Seshadri, Joseph B. Choi, H. S Udaykumar and Stephen Baek

Propellants, explosives, pyrotechnics, Vol.48(4), e202200268

04/2023

DOI: 10.1002/prep.202200268

Files and links (1)

url

https://doi.org/10.1002/prep.202200268View

Published (Version of record) Open Access

Abstract

Predictive simulations of the shock-to-detonation transition (SDT) in heterogeneous energetic materials (EM) are vital to the design and control of their energy release and sensitivity. Due to the complexity of the thermo-mechanics of EM during the SDT, both macro-scale response and sub-grid mesoscale energy localization must be captured accurately. This work proposes an efficient and accurate multiscale framework for SDT simulations of EM. We introduce a new approach for SDT simulation by using deep learning to model the mesoscale energy localization of shock-initiated EM microstructures. The proposed multiscale modeling framework is divided into two stages. First, a physics-aware recurrent convolutional neural network (PARC) is used to model the mesoscale energy localization of shock-initiated heterogeneous EM microstructures. PARC is trained using direct numerical simulations (DNS) of hotspot ignition and growth within microstructures of pressed HMX material subjected to different input shock strengths. After training, PARC is employed to supply hotspot ignition and growth rates for macroscale SDT simulations. We show that PARC can play the role of a surrogate model in a multiscale simulation framework, while drastically reducing the computation cost and providing improved representations of the sub-grid physics. The proposed multiscale modeling approach will provide a new tool for material scientists in designing high-performance and safer energetic materials.

Condensed Matter - Materials Science

Computer Science - Machine Learning

FOS: Computer and information sciences

General Chemistry

General Chemical Engineering

Materials Science (cond-mat.mtrl-sci)

FOS: Physical sciences

Machine Learning (cs.LG)

Details

Title: Subtitle: A physics-aware deep learning model for energy localization in multiscale shock-to-detonation simulations of heterogeneous energetic materials
Creators: Phong C. H Nguyen
Yen‐Thi Nguyen
Pradeep K. Seshadri - University of Iowa
Joseph B. Choi - University of Virginia
H. S Udaykumar - University of Iowa
Stephen Baek - University of Virginia
Resource Type: Journal article
Publication Details: Propellants, explosives, pyrotechnics, Vol.48(4), e202200268
DOI: 10.1002/prep.202200268
ISSN: 0721-3115
eISSN: 1521-4087
Publisher: Wiley
Grant note: DOI: 10.13039/100000001, name: National Science Foundation, award: 2203580
Language: English
Electronic publication date: 11/08/2022
Date published: 04/2023
Academic Unit: IIHR--Hydroscience and Engineering; Injury Prevention Research Center; Mechanical Engineering
Record Identifier: 9984530393802771

Metrics

29 Record Views

22 Times Cited - Web of Science

See more details