Numerical investigation of flow structures around a cylindrical afterbody under supersonic condition

Abstract

Large Eddy Simulation (LES) with dynamic Smagorinsky model has been applied to numerically investigate the complicated flow structures that evolve in the near wake of a cylindrical after body aligned with a uniform Mach 2.46 flow. Mean flow field properties obtained from numerical simulations, such as axial velocity, pressure on base surface, have been compared with the experimental measurements as well as with other published results. It has been found that standard k-epsilon model fails to predict the flow properties in the recirculation region where better agreement has been observed between the data obtained from LES and experimental measurements. Flow Statistics like turbulent kinetic energy and primary Reynolds' stress have also been calculated and compared with the results obtained from experiments in order to quantitatively assess the ability of LES technique to predict the turbulence properties of flow field in the highly compressible shear layer region. The data obtained from LES has been further analyzed to understand the evolution of coherent structures in the flow field. Proper Orthogonal Decomposition (POD) of the data obtained from central plane in the wake region has been performed in order to reveal the most energetic structures present in the flow field.

Numerical study of flow physics in supersonic base-flow with mass bleed

“…Therefore, the growth of the shear layers is affected and the core has been shifted downstream. (b) Secondly, the eddy viscosity predictions may affect the growth of the shear layers as reported in the literature [27]. Hence, the discrepancies may come due to either of these possible reasons or due to both of them; it is a bit difficult to identify one over other and that can be considered as a part of future work which requires detailed parametric investigation.…”

“…In this study, dimensions of geometry and the boundary conditions are selected according [27], all the grids produce similar results and the error lies within 2%. Hence, the grid with a radial span of 10Ro & axial span of 20Ro is considered for further study along the similar line with the published literature, while the approach length has been considered to be 4Ro (as previously used [27]) for both the bleed ratios (I=0.0148 & I=0.0226).…”

“…Since we are using a structured solver, O-H type blocking strategy is employed in the region behind the based to reduce the skewness and to achieve a better quality of mesh so that the convergence issues can be dealt with at a higher CFL number. However, these components are compared in the base-flow case without control and reported in previously published work [27], and this grid is the extension/modification to that framework only. Therefore, for further study the mesh with 1.85 million grid points is used for the geometry with approach length 4R0 as used in the previously published work [27].…”

“…However, these components are compared in the base-flow case without control and reported in previously published work [27], and this grid is the extension/modification to that framework only. Therefore, for further study the mesh with 1.85 million grid points is used for the geometry with approach length 4R0 as used in the previously published work [27].…”

“…We have also compared our LES results with the experimental data obtained by Herrin & Dutton [1] for supersonic base flow without any flow control. Detailed analysis of our numerical study of base flow without flow control has already been reported and can be found in the published literature [27].…”