Uncertainty-aware and parametrized dynamic reduced-order model: Application to unsteady flows

This uncertainty-aware Reduced Order Model (ROM) demonstrates enhanced robustness and generalization across varying dynamical regimes of unsteady flow. It provides systematic and reliable predictions by leveraging a Variational Autoencoder (VAE) to construct a suitable latent manifold, and attention mechanisms in the latent space to capture temporal and parametric dependencies.

Authors

Taraneh Sayadi (ORCID: https://orcid.org/0000-0001-9689-4528)
Nicolas Thome (ORCID: https://orcid.org/0000-0003-4871-3045)
Patrick Gallinari (ORCID: https://orcid.org/0000-0003-3360-8269)
Ismaël Zighed

Institutions

Conservatoire National des Arts et Métiers (FR)
Institut Universitaire de France (FR)
Sorbonne Université (FR)
Institut Jean Le Rond d'Alembert (FR)
Laboratoire Médiations (FR)

Publication Details

Journal: Physical Review Fluids
Published: 2025-11-12
DOI: https://doi.org/10.1103/f6ty-t6gl
Primary Topic: Model Reduction and Neural Networks
Type: article
Field-Weighted Citation Impact: 0.00

Controls

|||

ALL TIME

JAN

FEB

MAR

APR

MAY

JUN

JUL

AUG

SEP

OCT

article

Uncertainty-aware and parametrized dynamic reduced-order model: Application to unsteady flows

Taraneh Sayadi, Nicolas Thome, Patrick Gallinari, Ismaël Zighed

Nov 12, 2025

Physical Review Fluids

Model Reduction and Neural Networks

article

Uncertainty-aware and parametrized dynamic reduced-order model: Application to unsteady flows

Taraneh Sayadi, Nicolas Thome, Patrick Gallinari, Ismaël Zighed

November 12, 2025article en

Abstract

This uncertainty-aware Reduced Order Model (ROM) demonstrates enhanced robustness and generalization across varying dynamical regimes of unsteady flow. It provides systematic and reliable predictions by leveraging a Variational Autoencoder (VAE) to construct a suitable latent manifold, and attention mechanisms in the latent space to capture temporal and parametric dependencies.

Physical Review FluidsVol. 10(11)

Conservatoire National des Arts et Métiers (FR), Institut Universitaire de France (FR), Sorbonne Université (FR), Institut Jean Le Rond d'Alembert (FR), Laboratoire Médiations (FR)

Openalex Percentile: Top 73%

Model Reduction and Neural Networks

DOI Open Access PDF OpenAlex

AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.