Application and Validation of a High Order Neural Networks Based Riemann Solver for 1D Euler Equations

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21220%2F24%3A00376023" target="_blank" >RIV/68407700:21220/24:00376023 - isvavai.cz</a>

Výsledek na webu

<a href="https://am2024.uniza.sk/scientific-program/" target="_blank" >https://am2024.uniza.sk/scientific-program/</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Application and Validation of a High Order Neural Networks Based Riemann Solver for 1D Euler Equations

Popis výsledku v původním jazyce

The Riemann solver holds a key position in most computational fluid dynamics (CFD) codes, crucial for simulating compressible flows. The time to obtain the exact solution of the Riemann problem for ideal fluids is high because of the complexity of the model. Consequently, approximate Riemann solvers are preferred over exact ones to mitigate this complexity. This contribution describes application and validation of state-of-the-art Riemann solver, which very accurately approximates the behaviour of an exact Riemann solver needing only a fraction of the exact solver’s computational time. This is achieved by employing a specific type of neural networks. This new approximate Riemann solver is implemented in classical Godunov’s scheme and validated on the well-known Sod’s shock tube problem with promising results in terms of accuracy and time efficiency.

Název v anglickém jazyce

Application and Validation of a High Order Neural Networks Based Riemann Solver for 1D Euler Equations

Popis výsledku anglicky

The Riemann solver holds a key position in most computational fluid dynamics (CFD) codes, crucial for simulating compressible flows. The time to obtain the exact solution of the Riemann problem for ideal fluids is high because of the complexity of the model. Consequently, approximate Riemann solvers are preferred over exact ones to mitigate this complexity. This contribution describes application and validation of state-of-the-art Riemann solver, which very accurately approximates the behaviour of an exact Riemann solver needing only a fraction of the exact solver’s computational time. This is achieved by employing a specific type of neural networks. This new approximate Riemann solver is implemented in classical Godunov’s scheme and validated on the well-known Sod’s shock tube problem with promising results in terms of accuracy and time efficiency.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10102 - Applied mathematics

Projekt

<a href="/cs/project/EF16_019%2F0000826" target="_blank" >EF16_019/0000826: Centrum pokročilých leteckých technologií</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

25th International Scientific Conference APPLIED MECHANICS 2024 AM 2024 BOOK OF ARTICLES

ISBN

978-80-554-2090-5

ISSN

—

e-ISSN

—

Počet stran výsledku

4

Strana od-do

166-169

Název nakladatele

University of Žilina

Místo vydání

Žilina

Místo konání akce

Vysoké Tatry

Datum konání akce

17. 4. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—