Work

Supplementary Visualization Results for: Data Assimilation Module by a Neural Physics Framework

Year: 2026

Type: supplementary-materials

Abstract: This repository contains the supplementary visualization results associated with the manuscript entitled: "Data Assimilation Module by a Neural Physics Framework" The material includes qualitative analyses of Artificial Neural Network-based Data Assimilation (ANN-DA) applied to the Fluidity NN4PDEs framework for simulations of flow past a sphere at Reynolds number (Re = 1,000). The repository provides visualization reports comparing simulations with and without data assimilation under different source-term perturbation levels (3% and 10%) and assimilation frequencies (10, 20, 40, and 100 time steps). Only a representative subset of these results is included in the manuscript; the complete collection is provided to support reproducibility and detailed analysis.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.20733401

Authors: Ramos, Marcelo Paiva, Dias, Luiz Alberto Vieira, Chen, Boyang, Pain, Christopher C., Li, Linfeng, Fang, Fangxin, Souto, Roberto Pinto, Campos Velho, Haroldo Fraga de

Institutions: Instituto Tecnológico de Aeronáutica, Instituto Nacional de Pesquisas Espaciais, Imperial College London, Laboratório Nacional de Computação Científica

Language: English

Topic: Model Reduction and Neural Networks

Subfield: Statistical and Nonlinear Physics

Field: Physics and Astronomy

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7165048112

Cites:

Cited by:

Related to: