Spatio-temporal, multi-field deep learning of shock propagation in meso-structured media

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fernández-Godino, M. Giselle, Shachar, Meir H., Korner, Kevin, Belof, Jonathan L., Kumar, Mukul, Lind, Jonathan, Schill, William J.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910252841041920
author Fernández-Godino, M. Giselle
Shachar, Meir H.
Korner, Kevin
Belof, Jonathan L.
Kumar, Mukul
Lind, Jonathan
Schill, William J.
author_facet Fernández-Godino, M. Giselle
Shachar, Meir H.
Korner, Kevin
Belof, Jonathan L.
Kumar, Mukul
Lind, Jonathan
Schill, William J.
contents Predicting the extreme hydrodynamic response of porous and architected lattice materials is a fundamental challenge in high energy density physics, where shock-induced pore collapse, baroclinic vorticity, and anomalous kinetic and thermodynamic states must be resolved across multiple scales. Traditional high-fidelity hydrocodes are computationally prohibitive for large-scale design exploration in applications like planetary defense and inertial confinement fusion. We present a multi-field spatio-temporal model (MSTM) designed to overcome the limitations of standard machine learning surrogates, which often fail to capture the sharp gradients and non-linear field couplings characteristic of shock propagation. By training on high-fidelity, multiscale multiphysics data, MSTM simultaneously evolves seven coupled thermodynamic and kinetic fields - including pressure, temperature, density, and velocity - across complex material architectures. Our framework demonstrates the ability to accurately predict anomalous responses, such as counterintuitive post-shock density reductions and localized hotspot formation, with mean root mean squared errors as low as 1.4%. Crucially, the model's multi-field formulation maintains physical consistency and interface stability over long autoregressive rollouts, outperforming single-field models by 94% in structural fidelity. This framework enables a 1000x reduction in time to solution, providing a practical pathway for the real-time analysis and optimization of energy dissipation and momentum transfer in meso-structured media.
format Preprint
id arxiv_https___arxiv_org_abs_2509_16139
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Spatio-temporal, multi-field deep learning of shock propagation in meso-structured media
Fernández-Godino, M. Giselle
Shachar, Meir H.
Korner, Kevin
Belof, Jonathan L.
Kumar, Mukul
Lind, Jonathan
Schill, William J.
Machine Learning
68T07 (Primary) 76L05, 65Z05 (Secondary)
I.2.6; I.6.3; J.2; I.6.8
Predicting the extreme hydrodynamic response of porous and architected lattice materials is a fundamental challenge in high energy density physics, where shock-induced pore collapse, baroclinic vorticity, and anomalous kinetic and thermodynamic states must be resolved across multiple scales. Traditional high-fidelity hydrocodes are computationally prohibitive for large-scale design exploration in applications like planetary defense and inertial confinement fusion. We present a multi-field spatio-temporal model (MSTM) designed to overcome the limitations of standard machine learning surrogates, which often fail to capture the sharp gradients and non-linear field couplings characteristic of shock propagation. By training on high-fidelity, multiscale multiphysics data, MSTM simultaneously evolves seven coupled thermodynamic and kinetic fields - including pressure, temperature, density, and velocity - across complex material architectures. Our framework demonstrates the ability to accurately predict anomalous responses, such as counterintuitive post-shock density reductions and localized hotspot formation, with mean root mean squared errors as low as 1.4%. Crucially, the model's multi-field formulation maintains physical consistency and interface stability over long autoregressive rollouts, outperforming single-field models by 94% in structural fidelity. This framework enables a 1000x reduction in time to solution, providing a practical pathway for the real-time analysis and optimization of energy dissipation and momentum transfer in meso-structured media.
title Spatio-temporal, multi-field deep learning of shock propagation in meso-structured media
topic Machine Learning
68T07 (Primary) 76L05, 65Z05 (Secondary)
I.2.6; I.6.3; J.2; I.6.8
url https://arxiv.org/abs/2509.16139