Validation of correlations-based transition modeling strategies applied to the Spalart-Allmaras turbulence model for the computation of separation-induced transition

• Published in: Aerospace science and technology Vol. 119; p. 107045
• Main Authors: Bouchard, M., Marty, J., Deck, S., Costes, M.
• Format: Journal Article
• Language: English
• Published: Elsevier Masson SAS 01-12-2021; Elsevier
• Subjects: Boundary layer; Engineering Sciences; Langtry-Menter model; Physics; Separation-induced transition; Separation-induced transition; Langtry-Menter model; Boundary layer; BOUNDARY LAYER; SEPARATION-INDUCED TRANSITION; LANGTRY-MENTER MODEL
• ISSN: 1270-9638; 1626-3219
• DOI: 10.1016/j.ast.2021.107045

Predicting the laminar-turbulent transition is becoming a necessity in order to reduce margin costs in the design of turbines and helicopter rotors. Most RANS models lack the capacity to create turbulence sufficiently abruptly to accurately predict separation-induced transition. This article aims at evaluating the performances of the γ−Reθ,t strategy applied to the Spalart-Allmaras and k−ω−SST models (ν˜−γ−Reθ,t and k−ω−γ−Reθ,t). The chosen test case features a laminar separation bubble over a flat plate. The results are compared to the flows obtained using classical algebraic criteria and validated against DNS data. The effects of the correction by Dacles-Mariani et al. are detailed as well as those of two sets of correlations. The turbulence production mechanism of the present models is too weak downstream of the first increase of intermittency, and the laminar separation bubbles are longer than in the reference flow. More specifically, the ν˜−γ−Reθ,t model with Medida and Baeder's correlations fails to reach the correct levels of turbulence production in the fully turbulent boundary layer. Switching the correlations corrects this last behavior. Finally, using maps of the activation zones of the Dacles-Mariani correction, a sensor is exhibited that could be used in future separation-induced transition modeling efforts.