Saved in:
Bibliographic Details
Main Authors: Menicali, Luca, Grace, Andrew, Richter, David H., Castruccio, Stefano
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.10919
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917261721206784
author Menicali, Luca
Grace, Andrew
Richter, David H.
Castruccio, Stefano
author_facet Menicali, Luca
Grace, Andrew
Richter, David H.
Castruccio, Stefano
contents Fluid thermodynamics underpins atmospheric dynamics, climate science, industrial applications, and energy systems. However, direct numerical simulations (DNS) of such systems can be computationally prohibitive. To address this, we present a novel physics-informed spatiotemporal surrogate model for Rayleigh-Benard convection (RBC), a canonical example of convective fluid flow. Our approach combines convolutional neural networks, for spatial dimension reduction, with an innovative recurrent architecture, inspired by large language models, to model long-range temporal dynamics. Inference is penalized with respect to the governing partial differential equations to ensure physical interpretability. Since RBC exhibits turbulent behavior, we quantify uncertainty using a conformal prediction framework. This model replicates key physical features of RBC dynamics while significantly reducing computational cost, offering a scalable alternative to DNS for long-term simulations.
format Preprint
id arxiv_https___arxiv_org_abs_2505_10919
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Physics-Informed Spatiotemporal Deep Learning Framework for Turbulent Systems
Menicali, Luca
Grace, Andrew
Richter, David H.
Castruccio, Stefano
Fluid Dynamics
Machine Learning
Fluid thermodynamics underpins atmospheric dynamics, climate science, industrial applications, and energy systems. However, direct numerical simulations (DNS) of such systems can be computationally prohibitive. To address this, we present a novel physics-informed spatiotemporal surrogate model for Rayleigh-Benard convection (RBC), a canonical example of convective fluid flow. Our approach combines convolutional neural networks, for spatial dimension reduction, with an innovative recurrent architecture, inspired by large language models, to model long-range temporal dynamics. Inference is penalized with respect to the governing partial differential equations to ensure physical interpretability. Since RBC exhibits turbulent behavior, we quantify uncertainty using a conformal prediction framework. This model replicates key physical features of RBC dynamics while significantly reducing computational cost, offering a scalable alternative to DNS for long-term simulations.
title A Physics-Informed Spatiotemporal Deep Learning Framework for Turbulent Systems
topic Fluid Dynamics
Machine Learning
url https://arxiv.org/abs/2505.10919