Abstract#
This work studies the application of Dynamic Mode Decomposition (DMD) to analyze and reduce data from CFD simulations of military projectiles with base-bleed technology (a system that reduces aerodynamic drag). From 870 axisymmetric two-dimensional simulations, DMD identified dominant dynamic modes of the axial velocity field. The approach resulted in a dimensionality reduction of approximately 99.31% and highlighted relevant modes. The next steps involve using these modes in machine learning models for national defense applications.
Keywords#
Dynamic Mode Decomposition Aerodynamic Simulation Machine Learning Dimensionality Reduction Base Bleed Projectile
Citation#
RODRIGUES, Thiago N. E.; LUCENA, Rachel M.; LEMOS, Maurício F.; MANGIAVACCHI, Norberto; OLIVEIRA, Gustavo P.. Decomposição em Modos Dinâmicos de Simulações Aerodinâmicas em Larga Escala de Projéteis Base-Bleed. In: ESCOLA REGIONAL DE ALTO DESEMPENHO DA REGIÃO NORDESTE (ERAD-NE), 6. , 2025, Natal/RN. Porto Alegre: Sociedade Brasileira de Computação, 2025. p. 39-40.
@inproceedings{Rodrigues2025,
author = {Thiago Rodrigues and Rachel Lucena and Maurício Lemos and Norberto Mangiavacchi and Gustavo Oliveira},
title = { Decomposição em Modos Dinâmicos de Simulações Aerodinâmicas em Larga Escala de Projéteis Base-Bleed},
booktitle = {Anais da VI Escola Regional de Alto Desempenho da Região Nordeste},
location = {Natal/RN},
year = {2025},
pages = {39--40},
publisher = {SBC},
address = {Porto Alegre, RS, Brasil},
doi = {10.5753/erad-ne.2025.11745},
url = {https://sol.sbc.org.br/index.php/erad-ne/article/view/36399}
}
Related Material#
ERAD-NE 2025 aims to encourage the study and research in high-performance computing, computer architecture, distributed systems, parallel processing, and their applications in fields such as biology, engineering, physics, mathematics, computer science, medicine, finance, chemistry, among others.