Compressor cascade total pressure loss correlation modelling at design points using artificial neural networks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21220%2F22%3A00361580" target="_blank" >RIV/68407700:21220/22:00361580 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.kme.zcu.cz/compmech/download/proceedings/CM2022_Conference_Proceedings.pdf" target="_blank" >https://www.kme.zcu.cz/compmech/download/proceedings/CM2022_Conference_Proceedings.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Compressor cascade total pressure loss correlation modelling at design points using artificial neural networks

Popis výsledku v původním jazyce

The analysis of the flows by computational fluid dynamics becomes useful design and optimization method during recent years. Despite the advances in the computational power but it could be still very demanding. Therefore empirical models are commonly used as a main tool for design and prediction of basic performance of axial compressor cascades. The empirical correlations are derived from experimental data obtained from two-dimensional measurements. Unfortunately, sufficient amount of data is available only in cases of well-known airfoils as e.g. NACA 65-series or C.4 profiles. Thus, there is en effort to find a similar relation which will serve in the same manner for another family of the airfoils. The construction of such correlations using artificial neural networks is proposed in this work. In contrast to standard deep neural network, the proposed neural network is built using higher order neural units.

Název v anglickém jazyce

Compressor cascade total pressure loss correlation modelling at design points using artificial neural networks

Popis výsledku anglicky

The analysis of the flows by computational fluid dynamics becomes useful design and optimization method during recent years. Despite the advances in the computational power but it could be still very demanding. Therefore empirical models are commonly used as a main tool for design and prediction of basic performance of axial compressor cascades. The empirical correlations are derived from experimental data obtained from two-dimensional measurements. Unfortunately, sufficient amount of data is available only in cases of well-known airfoils as e.g. NACA 65-series or C.4 profiles. Thus, there is en effort to find a similar relation which will serve in the same manner for another family of the airfoils. The construction of such correlations using artificial neural networks is proposed in this work. In contrast to standard deep neural network, the proposed neural network is built using higher order neural units.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20301 - Mechanical engineering

Projekt

<a href="/cs/project/TK03030121" target="_blank" >TK03030121: Koncepční Návrh Bezpečnostního Systému pro Plynem Chlazené Jaderné Reaktory</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2022

Kód důvěrnosti údajů

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

Název statě ve sborníku

PROCEEDINGS OF COMPUTATIONAL MECHANICS 2022

ISBN

978-80-261-1116-0

ISSN

—

e-ISSN

—

Počet stran výsledku

4

Strana od-do

50-53

Název nakladatele

Západočeská univerzita v Plzni

Místo vydání

Plzeň

Místo konání akce

Srní

Datum konání akce

7. 11. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—