RANS and Large Eddy Simulation (LES) for Prediction of Separated Flow Characteristics for Flow Over a Wall-Mounted Hump

Abstract

One of the critical challenges in fluid flow research is to understand and predict separated flows. Separated flows are usually unsteady and turbulent. This work focuses on the calculations for the flow over a wall-mounted hump, an example of a turbulent separated flow. To validate the results of numerical simulation, the experimental data chosen was case 3 of the 2004 CFD Validation on Synthetic Jets and Turbulent Separation Control Workshop (http://cfdval2004.larc.nasa.gov/case3.html) conducted by NASA. This particular hump is the upper surface of a Glauert-Goldschmied type airfoil and has a chord length of C = 0.42 m, a maximum height of 0.0537 m, and a span of 0.5842 m. For this configuration, as the flow approaches the hump, the boundary layer experiences an adverse pressure gradient. Over the front convex portion of the hump, the flow experiences a favorable pressure gradient and accelerates, and then it separates over a relatively short concave section in the presence of a strong adverse pressure gradient. The hump has a simple geometry, but, nevertheless, is rich in many complex flow phenomena such as shear layer, separation, reattachment, and vortex interactions. Advances made in Computational Fluid Dynamics (CFD) have made available wide variety of turbulence models. A variety of turbulence model and simulation approaches are being used for this work. The flow is simulated using steady and unsteady-state three-dimensional Reynolds-Averaged Navier-Stokes (RANS) equations based turbulence models and three-dimensional time-dependent Detached Eddy Simulation (DES) and Large-Eddy Simulation (LES) methods. Case 3 of the CFD Validation workshop was done for both the baseline case as well as with flow control. In this work, the baseline case with Reynolds number of 371,600 based on the hump chord length, C, and Mach number, M, of 0.04 is simulated and analyzed. The experimental data reported by Greenblatt et. al. is used to validate the results for this work. K-ω, the Spalart-Allmaras and the SST turbulence models are implemented for the RANS simulation. Mean-velocity profiles and turbulent kinetic energy profiles, is reported and analyzed at several streamwise (x/C) locations, where C is the chord length of the hump. Detailed comparisons have been made of mean and turbulence statistics such as the pressure coefficient, skin-friction coefficient, and Reynolds stress profiles, with experimental results. The location of the reattachment behind the hump has been compared with experimental results.