Numerical investigation of three turbulence simulation models for S809 wind turbine airfoil

The contributions in terms of energy production and reliability benefits of horizontal axis wind turbines are growing on a yearly basis throughout the world, which has made them the predominant technology for wind energy harvesting. When numerically simulating wind turbines, the selection of turbule...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings of the Institution of Mechanical Engineers. Part A, Journal of power and energy Journal of power and energy, 2018-12, Vol.232 (8), p.1037-1048
Hauptverfasser: Sanei, Maryam, Razaghi, Reza
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The contributions in terms of energy production and reliability benefits of horizontal axis wind turbines are growing on a yearly basis throughout the world, which has made them the predominant technology for wind energy harvesting. When numerically simulating wind turbines, the selection of turbulence modelling approach can be of pivotal importance since it defines the numerical costs and gains of the simulation. In this paper, 2D computational fluid analysis of S809 Don Somners airfoil is targeted to evaluate the merits, drawbacks, and capabilities of three different common turbulence-modeling approaches, namely Spalart-Allmaras, k-w SST, and Transition SST models. The results are validated through comparison against experimental data on pressure coefficient distribution; a significant agreement between experimental data and Transition SST model is observed, which makes this model the reference model in this study. The other two approaches are compared with the results of the aforementioned model in terms of pressure distribution and flow separation predictions. The results indicate that k-w and Spalart-Allmaras models are as reliable as Transition SST when the Reynolds number and angle of attack are high enough (i.e. higher than 10 to 12°). At low Reynolds numbers, which means under 5 to 6°, the reference model is observed to be the most reliable among the three. The results of this study are of great importance when full rotor simulation is the task in hand, since they define the most appropriate approach to use with each flow condition to achieve an optimized numerical simulation procedure.
ISSN:0957-6509
2041-2967
DOI:10.1177/0957650918767301