Physics-Informed Spatiotemporal Deep Learning Framework for Turbulent Systems

A 2.8 MB PDF file authored by Luca Menicali, last updated on April 22, 2026. The dataset likely contains supplementary materials for a novel physics-informed surrogate model for Rayleigh-Bénard convection, a canonical turbulent fluid flow system. The model combines convolutional and recurrent neural networks with a conformal prediction framework for uncertainty quantification.

Use Cases

• Developing surrogate models for turbulent convective flows based on the described physics-informed deep learning framework.
• Quantifying uncertainty in fluid dynamics simulations based on the mentioned conformal prediction framework.
• Reducing computational cost for long-term fluid simulations based on the spatiotemporal surrogate modeling approach.
• Studying the application of large language model-inspired architectures to temporal dynamics in physical systems.

Strengths

• Dataset is 2.8 MB in size, making it a lightweight download.
• Released under a permissive CC-BY-4.0 license, allowing for broad reuse.
• Focuses on a canonical and well-studied physical system (Rayleigh-Bénard convection).

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The primary file format is PDF, which may not contain structured, machine-readable data.

Provenance

Source

figshare

Collection Method

Likely contains supplementary materials for a research article on a novel machine learning framework.

Time Range

null

Freshness

Last updated 2026-04-22 15:18:24; freshness should be verified.

Geography

null