Nima Tajallipour

Self-Adaptive Upwinding for Large Eddy Simulation of Turbulent Flows on Unstructured Elements

A self-adaptive upwinding method for large eddy simulation is proposed to reduce the numerical dissipation of a low-order numerical scheme on unstructured elements. This method is used to extend an existing Reynolds-averaged Navier-Stokes code to a large eddy simulation code by adjusting the contribution of the upwinding term to the convective flux. This adjustment is essentially controlled by the intensity of the local wiggle and reduces the upwind contribution in the Roe-MUSCL scheme. First, the stability characteristic of the new scheme is studied using a channel flow stability test. It is essential to ensure that the proposed scheme is able to adjust upwinding in the presence of very high gradients and that it prohibits the divergence of the simulation. Second, the decaying isotropic turbulence is simulated to study the capability of the new scheme to generate the suitable decaying rate for the total kinetic energy and its influence over the slope of the energy spectrum at different computational times. Finally, the flow separation phenomenon over a NACA0025 profile is numerically investigated and results are compared with experimental data.

Large‐eddy simulation of a compressible free jet flow on unstructured elements

Purpose – The purpose of this paper is to investigate a large‐eddy simulation, using low order numerical discretization and upwinding schemes on unstructured grids, for a turbulent free jet at Mach number 0.95. The accuracy and stability performance is discussed for the finite element/volume upwinding numerical code used.Design/methodology/approach – This code is equipped with a self‐adaptive upwinding method which has been previously developed to reduce the numerical dissipation of applied low order flux calculation on unstructured elements using Roes scheme. Herein, this method is used to numerically investigate a high Reynolds, compressible turbulent free jet and compare the results with a recently published set of experimental data. The effect of grid size is also investigated. A reasonable good agreement with the experimental measurements is obtained.Findings – Based on the results, it is concluded that the developed self‐adaptive upwinding scheme provides a considerably better emulation of the flow...

Effects of Upwinding in Large Eddy Simulation of Turbulent Flows

In this paper a new adaptive upwinding method, compatible with the available numerical scheme, is developed and implemented. This method improves the results of large eddy simulation (LES) by adjusting the contribution of upwinding term to the convective flux. This adjustment is essentially controlled by the intensity of the local wiggle. This work is an attempt to study and evaluate the numerical dissipation of the available low order numerical tool and to prepare and improve this tool for the purpose of LES. At first, the available finite element/volume numerical code, previously used for the Reynolds-averaged Navier–Stokes (RANS) simulations of compressible flows, is extensively studied, using channel flow stability test and decaying isotropic turbulence. The goal is to use these numerical tests in order to investigate the ability of the numerical tool in order to emulate necessary turbulent characteristic. The new adaptive upwinding method is then introduced in order to improve the results. In addition, a review of the main aspects of LES of turbulent flows such as cascade of energy from high to low scale eddies, effects of subgrid modeling, numerical dissipation, accuracy and stability, is also presented. It has been a main concern in our work to choose those numerical tests which are relatively simple and don’t require very high computational efforts, but are also viable enough to show main features necessary to assess the performance of the numerical scheme. I. Introduction The Navier–Stokes equations (NSE), supplemented by empirical laws for the dependence of viscosity and thermal conductivity to other flow variables and by a constitutive law defining how the pressure depends on the other flow variables, can be used to describe all flow phenomena in a linear viscous fluid. In addition, appropriate initial and boundary conditions must be supplied to ensure the well-posedness of the NSE and to select the specific physical flow realization which is going to be emulated. From a computational point of view, the NSE can be solved directly (without any need for filtration or averaging) for laminar flows, while for turbulent flows the wide range of eddy scales, required to be captured, prohibits direct numerical simulation (DNS). That’s specially the case for high Reynolds numbers. 18 , 17 Therefore direct numerical simulation of turbulent flows is still far out of range for flows of practical industrial interest and most of the DNS simulations reported in the literature are limited to simple geometries and moderate Reynolds numbers. Moreover, some of the recommendations given in the literature calling for required highly resolved grids and high-order numerical schemes are clearly difficult to respect in an industrial context. 12