Conference papers

Permanent URI for this collection

https://testres.nu.edu.kz/handle/123456789/826

Browse

Browsing Conference papers by Author "Azevedo, R."

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Simulation of the Velocity field in compound channel flow using different closure models
    (2010) Filonovich, M.; Azevedo, R.; Rojas-Solorzano, L.; Leal, J. B.
    In this study a comparison of three turbulence closure models (two isotropic and one anisotropic) with experimental data is performed. The interaction between the main channel (MC) flow and the floodplain (FP) generates a complex flow structure. A shallow mixing layer develops between the MC flow and the slower FP flow generating a high horizontal shear layer, streamwise and vertical vortices, momentum transfer and other phenomena, related to velocity retardation and acceleration. This phenomenon dissipates part of the kinetic energy and contributes to the reduction of the velocity differences between the MC and the FP. The large scale vortices that are generated in the shear layer are anisotropic, provoking the formation of secondary flow cells that influence the primary velocity distribution. These threedimensional turbulent structures can be reasonable well reproduced by a simple anisotropic model (Algebraic Stress Model). The isotropic models are capable of simulating the boundary layer, especially the model base in k-ω equations, but cannot simulate the shear layer that develops at the interface
  • Thumbnail Image
    ItemOpen Access
    Verification and validation of computational fluid dynamics simulations of compound channel
    (2010) Filonovich, M. S.; Azevedo, R.; Rojas-Solorzano, L.; Leal, J. B.
    In this study the verification and validation of a 2nd order turbulence closure model is performed for an experimental compound channel flow, where the velocity field was measured by a Laser Doppler Velocimeter. Detailed Explicit Algebraic Reynolds Stress Model (EARSM) simulation is reported. The Grid Convergence Index (GCI) approach proposed by Roache (1998) was adopted to evaluate the uncertainty associated to grid resolution. The velocity components, the turbulent kinetic energy (TKE) and the dissipation rate were used as variables of interest. The GCI results present low values for the streamwise velocity, TKE and dissipation rate, but higher values in what concerns vertical and spanwise velocities. This indicates that the mean primary flow has converged but the secondary flow field still depends on grid resolution. Based on GCI values distribution, the mesh was locally refined. Comparison of numerical and experimental results shows good agreement