The Efficacy of RANS Models for the Modelling of Turbulence Generating Grids
Loading...
Date
2021-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
The Ohio State University
Abstract
Turbulence generating grids are a widely used method for intentionally producing various
amounts of turbulence in wind tunnels. The analysis of turbulent flow generated by these grids
through experimental methods is well documented but has only been computationally
investigated primarily through computational fluid dynamics (CFD) using scale-resolving
models such as direct numerical simulation (DNS). Because these models are time intensive, the
incorporation of lower fidelity Reynolds' Averaged Navier Stokes (RANS) models in CFD can
allow for quicker convergence times and greater relaxation of mesh requirements during
evaluation of turbulence generating grids. In this work, an unstructured tetrahedral mesh was
incorporated with the RANS models Realizable k-ε and SST k-ω to predict the amount of
turbulence intensity downstream of the grid and investigate differences in turbulence production
at the plate hole walls. This analysis was verified experimentally with the collection of
turbulence data retrieved using hotwire anemometry in a blowdown wind tunnel. It was found
that the utilization of RANS models presented significant error (>50%) and were not deemed
viable for the prediction of downstream turbulence intensity. This study adds to existing
literature a more detailed understanding of the flow field of RANS models for analysis of
turbulence generating grids, and the results provide a framework of where fast analysis of
turbulence generating grids using RANS models may be used in future study and industry
application.
Description
Keywords
Turbulence, CFD, RANS, Turbulence Grids