DNS 1-5 Description: Difference between revisions

From KBwiki
Jump to navigation Jump to search
No edit summary
 
(73 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 
=HiFi-TURB-DLR rounded step=
{{DNSHeaderLib
{{DNSHeader
|area=1
|area=1
|number=5
|number=5
Line 6: Line 6:
__TOC__
__TOC__
= Introduction =
= Introduction =
Give a brief overview of the test case. Describe the main characteristics of the flow. In particular, what are the underlying flow physics which must be captured by the computations ? Give reasons for this choice (e.g. poorly understood flow physics, difficulty to predict the flow with standard turbulence models, ...). Detail any case-specific data that needs to be generated.
The HiFi-TURB-DLR rounded step test case is a case in which a turbulent boundary layer flows over a flat plate before reaching a planar rounded backward step.
The particular geometry generates an adverse pressure gradient which accelerates the boundary layer upstream of the step before inducing an incipient separation above the smooth step,
a flow physics that is well known to be not correctly predicted by state-of-the-art Reynolds-Averaged Navier-Stokes (RANS) models.
 
= Review of previous studies =
= Review of previous studies =
Provide a brief review of related past studies, either experimental or computational. Identify
The design of the test case was inspired by the experimental work on the flow around an axisymmetric test body of [[DNS_1-5_description#1|Disotell ''et al.'']] and  [[DNS_1-5_description#2|Disotell ''et al.'' (2017)]]. Disotell et al. focused their study of the possible separation and reattachment in the rear part of the body. The original test case is here revised as a 2D planar flow problem without any tunnel walls but with a far-field condition opposite to the solid wall. The reasons for using a planar geometry are thoroughly discussed in [[UFR_3-36|UFR 3-36 Test Case]].
the configuration chosen for the present study and position it with respect to previous studies.
 
If the test case is geared on a certain experiment, explain what simplifications ( e.g. concern-
ing geometry, boundary conditions) have been introduced with respect to the experiment in the
computational setup to make the computations feasible and avoid uncertainty or ambiguity.
= Description of the test case =
= Description of the test case =
A detailed self-contained description should be provided. It can be kept fairly short if a link can be made to an external data base where details are given. Then only the differences should be clearly indicated.
==Geometry and flow parameters==
==Geometry and flow parameters==
Describe the general set up of the test case and provide a sketch of the geometry, clearly identifying location and type of boundaries. Specify the non-dimensional flow parameters which define the flow regime (e.g. Reynolds number, Rayleigh number, angle of incidence etc), including the scales on which they are based. Provide a detailed geometrical description, by preference in form of a CAD, or alternatively as lists of points and a description of the interpolation.
The test case is designed as a numerical experiment with the aim of comparing RANS results to DNS data.
 
The flow problem was defined by DLR with the aid of RANS calculations to get a condition with incipient separation. Other cases with moderate and full separation were also designed by DLR, see [[DNS_1-5_description#3|Alaya ''et&nbsp;al. (2021)'']], but for the present uDNS only the incipient condition for the Reynolds number <math>{Re_H=78,490}</math> and the Mach number <math>Ma=0.13455</math> was considered, where <math>H</math> is the step height.
 
The geometry of the curved step is shown in [[DNS_1-5_description#figure2|Fig. 2]] and its definition is reported in [[UFR_3-36_Test_Case|UFR 3-36 Test Case]].
The test case is designed as a numerical experiment with the aim of comparing RANS results to DNS data. For the set-up of the DNS, the inflow boundary conditions are different, e.g. a recycling method can be used to generate the turbulent input or synthetic turbulence can be injected. It is also possible to numerically trip the boundary layer from laminar to turbulent to generate the desired turbulent boundary layer. Hence, to ensure a comparison to the results achieved with RANS turbulence models, a reference position upstream of the APG-area is defined where boundary layer properties need to match between RANS and DNS computations to permit the comparison downstream in the region of interest. The reference position is located at <math>{x/H = -3.51}</math>. Depending on the generation of turbulence at the inlet, the computational domain needs to be adapted to ensure the correct boundary layer properties at the reference position. If numerical tripping is performed, the laminar and turbulent distances need to be determined upstream of the reference position by precursor simulation as displayed in [[Lib:UFR_3-36_Test_Case#figure2|Fig. 2]].


<div id="figure2"></div>
<div id="figure2"></div>
{|align="center" border="0" width="500"
{|align="center" border="0" width="500"
|[[Image:UFR3-36_fig3.png|500px]]
|[[Image:DNS1-5 rounded step DNS setup.png|800px]]
|-
|-
!align="center"|Figure 2: Set-up for DNS Simulation
!align="center"|Figure 2: Set-up for the DNS simulation and pressure distribution
|}
|}


At the reference position <math>{x_{ref}}</math>, the properties of the turbulent boundary layer are determined by the Reynolds number based on the momentum thickness <math>{Re_{\theta} = 1496}</math> and the Reynolds number based on the friction velocity <math>{Re_{\tau} = 609}</math> computed with SA-neg-RC-QCR-LRe model. To estimate the computational effort for DNS, the largest values of <math>{Re_{\theta}}</math> and <math>{Re_{\tau}}</math> are also given at the position <math>{x/H = 10.5}</math> by <math>4,624</math> and <math>1,399</math> respectively.  
==Computation domain and boundary conditions==
 
Although for DNS the inflow boundary conditions are different than RANS, to allow a valid comparison they must guarantee the same boundary layer properties at a given point, hereinafter referred to as checkpoint, upstream of the rounded step, i.e., <math>{x_{ckp}/H = -3.5}</math>.
Amongst the different flow condition proposed, UNIBG focused the effort on the incipient separation configuration case with <math>{Re_H = 78,490}</math>. A precursory computational campaign for the turbulent flow over a flat plate has been performed with the purpose of (i) assessing the effectiveness of the Synthetic Inlet Turbulence (SIT) injection strategy inspired by the work of Housseini et al. [&#8204;[[Lib:UFR_3-36_References#18|18]]] and Schlatter and Örlü [&#8204;[[Lib:UFR_3-36_References#19|19]]]; (ii) investigating the influence of the mesh density on the solution; (iii) defining the inlet boundary position and flow condition that ensure the target boundary layer integral parameters at the reference location <math>{x_{ref}}</math> (see [[Lib:UFR_3-36_Test_Case#figure3|Fig. 3]]). According to the outcomes of this campaign, the inlet boundary is set at <math>{x/H = -12.71}</math>, where the Blasius laminar velocity profile computed at <math>{Re_x=650,000}</math>, the uniform static pressure and the uniform total temperature (see [[Lib:UFR_3-36_Test_Case#table2|Table 2]]) are imposed. At location <math>{x/H = -12.1}</math>, within the laminar boundary layer region, the tripping source term is activated to promote the transition to turbulence [&#8204;[[Lib:UFR_3-36_References#18|18]]][&#8204;[[Lib:UFR_3-36_References#19|19]]]. The outlet boundary is positioned at <math>{x/H = 23.95}</math>, with a pressure outflow condition having the same value of the inlet static pressure and at the end of a streamwise coarsened mesh region starting at <math>{x/H = 13.82}</math>. Following the setup of the RANS computations, the no-slip adiabatic boundary condition is set while at the upper boundary, situated at <math>{179 H}</math> from the wall upstream the smooth bump, the far-field boundary condition is imposed. Side planes, instead, are considered as periodic with a distance from each other of <math>{\Delta z = 3 H}</math>.  
At this position <math>{x_{ckp}}</math>, the properties to be matched are the Reynolds number based on the momentum thickness <math>{Re_{\theta} = 1,780}</math>, the Reynolds number based on the friction velocity <math>{Re_{\tau} = 700}</math> and the boundary layer thickness <math>\delta_{99}/H=0.241</math>.  
As a technique to promote the laminar-turbulent transition of the boundary layer and reduce the upstream length of the domain, a local tripping term inspired by the work of [[DNS_1-5_description#5|Housseini ''et&nbsp;al.'' (2016)]] and [[DNS_1-5_description#6|Schlatter and Örlü (2012)]] is used.
To define the mesh density and the computational domain size that ensure the target boundary layer integral parameters at <math>{x_{chp}}</math>, a precursory computational campaign for the turbulent flow over a flat plate was performed.
According to the outcomes of this campaign, the inlet boundary is located at <math>{x/H = -12.7}</math>, where the Blasius laminar velocity profile computed at <math>{Re_x=650,000}</math>, the uniform static pressure <math>{P_{s,ref}}</math> and the uniform total temperature <math>{T_{t,inflow}}</math> are imposed, see [[UFR_3-36_Test_Case#table2|UFR 3-36: Table 2]] for dimensional values of each quantity. At location <math>{x/H = -12.1}</math>, within the laminar boundary layer region, the tripping source term is activated to promote the transition to turbulence, see [[DNS_1-5_description#figure2|Fig. 2]].  
At the outlet boundary, placed at <math> x/H=24.0 </math>, the static pressure <math>{P_{s,ref}}</math> is imposed with an exit-pressure outflow boundary condition. To mitigate spurious perturbations possibly originating at the outlet boundary, the mesh is coarsened in the streamwise direction for <math>{x/H > 13.82}</math>. The upper boundary is a permeable far-field Riemann boundary condition located <math>180.0 H</math> from the no-slip adiabatic wall downstream the smooth step and computed via the exact Riemann solver. Finally, side planes are considered as periodic with a distance from each other of <math>{\Delta z = 3 H}</math>.


The DNS have been performed by using the UNIBG in-house software MIGALE [&#8204;[[Lib:UFR_3-36_References#20|20]]]. MIGALE is an implicit high-order discontinuous Galerkin solver for the compressible and incompressible Navier-Stokes equations. Godunov fluxes are treated with the exact solution of the local Riemann problems while viscous fluxes are handled by means of the BR2 scheme [&#8204;[[Lib:UFR_3-36_References#21|21]]]. The time integration is performed with linearly implicit Rosenbrock type Runge-Kutta schemes with optimal stability properties up to order five.
==References==
The computational mesh is made of <math>15,016,384</math> hexahedral elements with quadratic edges concentrated near the wall region. The wall-normal growth ratio is approximatively <math>1.2</math> with a first cell height of <math>{y^+ \approx 1}</math>. Time integration is performed with the fifth order – eight stages Rosenbrock scheme ROD5_1 [&#8204;[[Lib:UFR_3-36_References#22|22]]] using a global time step adaptation strategy [&#8204;[[Lib:UFR_3-36_References#23|23]]]. The corresponding average step size is <math>{\Delta t = 19/14000}</math> CTU, where the convective time unit (CTU) is defined as the ratio between <math>{H}</math> and the freestream velocity <math>{U_\infty}</math>. Turbulence statistics have been collected for <math>26</math> CTU.


==Boundary conditions==
#<div id="1">'''Disotell, K. J. and Rumsey, C. L.''': Modern CFD validation of turbulent flow separation on axisymmetric afterbodies.</div>
Specify the prescribed boundary conditions, as well as the means to verify the initial flow development. In particular describe the procedure for determining the in flow conditions comprising the instantaneous (mean and fluctuating) velocity components and other quantities. Provide reference profiles for the mean flow and fluctuations at in flow - these quantities must be supplied separately as part of the statistical data as they are essential as input for turbulence-model calculations. For checking purposes, these profiles should ideally also be given downstream where transients have disappeared; the location and nature of these cuts should be specified, as well as the reference result.
#<div id="2">'''Disotell, K. J. and Rumsey, C. L. (2017)''': Development of an axisymmetric afterbody test case for turbulent flow separation validation. ''NASA/TM-2017219680'', Langley Research Center, Hampton, Virginia</div>
<br/>
#<div id="3">'''Alaya, E., Grabe, C. and Knopp, T. (2021)''': Design of a parametrized numerical experiment for a 2D turbulent boundary layer flow with varying adverse pressure gradient and separation behaviour. ''DLR-IB-AS-GO-2020-109'', DLR-Interner Bericht, DLR Institute of Aerodynamics and Flow Technology</div>
#<div id="4">'''Bassi, F., Botti, L., Colombo, A. C, Ghidoni, A. and Massa, F. (2016)''': On the development of an implicit high-order Discontinuous Galerkin method for DNS and implicit LES of turbulent flows. ''European Journal of Mechanics, B/Fluids'', Vol. 55(2), pp. 367-379</div>
# <div id="5">'''S. Hosseini, R. Vinuesa, P. Schlatter, A. Hanifia and D. Henningson (2016):''' Direct numerical simulation of the flow around a wing section at moderate Reynolds number, ''International Journal of Heat and Fluid Flow,''&nbsp;61:117&ndash;128</div>
# <div id="6">'''Schlatter, P. and &Ouml;rl&uuml;, R. (2012):''' Turbulent boundary layers at moderate Reynolds numbers: inflow length and tripping effects,''Journal of Fluid Mechanics,''&nbsp;710:5&ndash;34</div>
----
----
{{ACContribs
{{ACContribs
Line 43: Line 47:
| organisation=Università degli studi di Bergamo (UniBG)
| organisation=Università degli studi di Bergamo (UniBG)
}}
}}
{{DNSHeaderLib
{{DNSHeader
|area=1
|area=1
|number=5
|number=5

Latest revision as of 16:10, 17 February 2023

HiFi-TURB-DLR rounded step

Front Page

Description

Computational Details

Quantification of Resolution

Statistical Data

Instantaneous Data

Storage Format

Introduction

The HiFi-TURB-DLR rounded step test case is a case in which a turbulent boundary layer flows over a flat plate before reaching a planar rounded backward step. The particular geometry generates an adverse pressure gradient which accelerates the boundary layer upstream of the step before inducing an incipient separation above the smooth step, a flow physics that is well known to be not correctly predicted by state-of-the-art Reynolds-Averaged Navier-Stokes (RANS) models.

Review of previous studies

The design of the test case was inspired by the experimental work on the flow around an axisymmetric test body of Disotell et al. and Disotell et al. (2017). Disotell et al. focused their study of the possible separation and reattachment in the rear part of the body. The original test case is here revised as a 2D planar flow problem without any tunnel walls but with a far-field condition opposite to the solid wall. The reasons for using a planar geometry are thoroughly discussed in UFR 3-36 Test Case.

Description of the test case

Geometry and flow parameters

The test case is designed as a numerical experiment with the aim of comparing RANS results to DNS data. The flow problem was defined by DLR with the aid of RANS calculations to get a condition with incipient separation. Other cases with moderate and full separation were also designed by DLR, see Alaya et al. (2021), but for the present uDNS only the incipient condition for the Reynolds number and the Mach number was considered, where is the step height. The geometry of the curved step is shown in Fig. 2 and its definition is reported in UFR 3-36 Test Case.

DNS1-5 rounded step DNS setup.png
Figure 2: Set-up for the DNS simulation and pressure distribution

Computation domain and boundary conditions

Although for DNS the inflow boundary conditions are different than RANS, to allow a valid comparison they must guarantee the same boundary layer properties at a given point, hereinafter referred to as checkpoint, upstream of the rounded step, i.e., . At this position , the properties to be matched are the Reynolds number based on the momentum thickness , the Reynolds number based on the friction velocity and the boundary layer thickness . As a technique to promote the laminar-turbulent transition of the boundary layer and reduce the upstream length of the domain, a local tripping term inspired by the work of Housseini et al. (2016) and Schlatter and Örlü (2012) is used. To define the mesh density and the computational domain size that ensure the target boundary layer integral parameters at , a precursory computational campaign for the turbulent flow over a flat plate was performed. According to the outcomes of this campaign, the inlet boundary is located at , where the Blasius laminar velocity profile computed at , the uniform static pressure and the uniform total temperature are imposed, see UFR 3-36: Table 2 for dimensional values of each quantity. At location , within the laminar boundary layer region, the tripping source term is activated to promote the transition to turbulence, see Fig. 2. At the outlet boundary, placed at , the static pressure is imposed with an exit-pressure outflow boundary condition. To mitigate spurious perturbations possibly originating at the outlet boundary, the mesh is coarsened in the streamwise direction for . The upper boundary is a permeable far-field Riemann boundary condition located from the no-slip adiabatic wall downstream the smooth step and computed via the exact Riemann solver. Finally, side planes are considered as periodic with a distance from each other of .

References

  1. Disotell, K. J. and Rumsey, C. L.: Modern CFD validation of turbulent flow separation on axisymmetric afterbodies.
  2. Disotell, K. J. and Rumsey, C. L. (2017): Development of an axisymmetric afterbody test case for turbulent flow separation validation. NASA/TM-2017219680, Langley Research Center, Hampton, Virginia
  3. Alaya, E., Grabe, C. and Knopp, T. (2021): Design of a parametrized numerical experiment for a 2D turbulent boundary layer flow with varying adverse pressure gradient and separation behaviour. DLR-IB-AS-GO-2020-109, DLR-Interner Bericht, DLR Institute of Aerodynamics and Flow Technology
  4. Bassi, F., Botti, L., Colombo, A. C, Ghidoni, A. and Massa, F. (2016): On the development of an implicit high-order Discontinuous Galerkin method for DNS and implicit LES of turbulent flows. European Journal of Mechanics, B/Fluids, Vol. 55(2), pp. 367-379
  5. S. Hosseini, R. Vinuesa, P. Schlatter, A. Hanifia and D. Henningson (2016): Direct numerical simulation of the flow around a wing section at moderate Reynolds number, International Journal of Heat and Fluid Flow, 61:117–128
  6. Schlatter, P. and Örlü, R. (2012): Turbulent boundary layers at moderate Reynolds numbers: inflow length and tripping effects,Journal of Fluid Mechanics, 710:5–34


Contributed by: Francesco Bassi, Alessandro Colombo, Francesco Carlo Massa — Università degli studi di Bergamo (UniBG)

Front Page

Description

Computational Details

Quantification of Resolution

Statistical Data

Instantaneous Data

Storage Format


© copyright ERCOFTAC 2024