RANS turbulence model development using CFD-driven machine learning

This paper presents a novel CFD-driven machine learning framework to develop Reynolds-averaged Navier-Stokes (RANS) models. The CFD-driven training is an extension of the gene expression programming method Weatheritt and Sandberg (2016) [8], but crucially the fitness of candidate models is now evalu...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Journal of computational physics 2020-06, Vol.411 (C), p.109413, Article 109413
Hauptverfasser:	Zhao, Yaomin, Akolekar, Harshal D., Weatheritt, Jack, Michelassi, Vittorio, Sandberg, Richard D.
Format:	Artikel
Sprache:	eng
Schlagworte:	Computational fluid dynamics Computational physics Cost function Gene expression Machine learning Nozzles Reynolds averaged Navier-Stokes method Training Turbines Turbomachinery Turbulence modelling Turbulence models Wake mixing
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!