Technical gazette, Vol. 22 No. 5, 2015.
Original scientific paper
https://doi.org/10.17559/TV-20140404130058
A hybrid rans-les method with compressible k-omegaSSTSAS turbulence model for high Reynolds number flow applications
Bojan Šekutkovski
; University of Belgrade, Faculty of Mechanical Engineering in Belgrade, Kraljice Marije 16, Belgrade, Serbia
Ivan Kostić
; University of Belgrade, Faculty of Mechanical Engineering in Belgrade, Kraljice Marije 16, Belgrade, Serbia
Zoran Stefanović
; University of Belgrade, Faculty of Mechanical Engineering in Belgrade, Kraljice Marije 16, Belgrade, Serbia
Aleksandar Simonović
; University of Belgrade, Faculty of Mechanical Engineering in Belgrade, Kraljice Marije 16, Belgrade, Serbia
Olivera Kostić
; University of Belgrade, Faculty of Mechanical Engineering in Belgrade, Kraljice Marije 16, Belgrade, Serbia
Abstract
Three-dimensional, compressible, viscous and transient transonic turbulent flow over the wing was simulated by a hybrid RANS-LES modelling method, combined with the compressible k-omegaSSTSAS turbulence model. This approach is based on dividing the contribution of the fluctuating and the averaged velocity fields in the subgrid tensor, and modelling each of them with its corresponding turbulent viscosity. The "RANS mode" is used in flow field domains which can be treated with acceptable accuracy as relatively steady, such as in the boundary layer, while the "LES mode" is applied in the dominantly unsteady regions, far from the wing. Discretization of the governing equations is performed by Finite Volume Method on unstructured mesh. The method has been verified on the Onera M6 wing. The parallelization is achieved by decomposing the mesh into sub-domains and using the Open MPI technology. The implementation of turbulence model has been done using OpenFOAM. The flow simulation was also performed using ANSYS Fluent, and the results of the two methods were compared mutually, and with the Onera M6 experiment.
Keywords
finite volume method; high Reynolds numbers; k-omegaSSTSAS; OpenFOAM; parallel computing; RANS-LES; transonic turbulent flow; unstructured meshes
Hrčak ID:
147293
URI
Publication date:
22.10.2015.
Visits: 3.112 *