A Physics-Informed Spatiotemporal Deep Learning Framework for Turbulent Systems

• 成果类型: Article; Early Access
• 署名作者: Menicali, Luca; Grace, Andrew P.; Richter, David H.; Castruccio, Stefano
• 署名单位: University of Notre Dame; University of Notre Dame
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459; 1537-274X
• DOI: 10.1080/01621459.2026.2632865
• 发表日期: 2026-04-22
• 关键词: Fluid thermodynamics; Long-range forecasting; Physics-informed inference; State-space modeling; convection; networks

摘要

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-B & eacute;nard 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 Supplementary materials for this article are available online.