Large-Eddy and Direct Numerical Simulations of the Bachalo-Johnson Flow with Shock-Induced Separation

Spalart, Philippe R.; Belyaev, Kirill V.; Garbaruk, A. V.; Shur, M. L.; Strelets, M. Kh.; Travin, Andrey

doi:10.1007/s10494-017-9832-z

Cited by 48 publications

(35 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Existing forcing methods have typically addressed this issue by generating fluctuations only at a prescribed LES inlet [28,29]. In addition to requiring specification of the distinct LES region and entire spectrum of locally resolved synthetic turbulent scales and intensities, the LES inlet location must be sufficiently upstream of the actual feature of interest so that the artificial inflow condition can "heal" to a realistic state.…”

Section: B Active Energy Transfermentioning

confidence: 99%

Towards a Predictive Hybrid RANS/LES Framework

Haering

Oliver

Moser

2019

AIAA Scitech 2019 Forum

View full text Add to dashboard Cite

Predictive simulation of many complex flows requires moving beyond Reynolds-averaged Navier-Stokes (RANS) based models to representations resolving at least some scales of turbulence in at least some regions of the flow. To resolve turbulence where necessary while avoiding the cost of performing large eddy simulation (LES) everywhere, a broad range of hybrid RANS/LES methods have been developed. While successful in some situations, existing methods exhibit a number of deficiencies which limit their predictive capability in many cases of interest, for instance in many flows involving smooth wall separation. These deficiencies result from inappropriate blending approaches and resulting inconsistencies between the resolved and modeled turbulence as well as errors inherited from the underlying RANS and LES models. This work details these problems and their effects in hybrid simulations, and develops a modeling paradigm aimed at overcoming these challenges. The new approach discards typical blending approaches in favor of a hybridization strategy in which the RANS and LES model components act through separate models formulated using the mean and fluctuating velocity, respectively. Further, a forcing approach in which fluctuating content is actively transferred from the modeled to the resolved scales is introduced. Finally, the model makes use of an anisotropic LES model that is intended to represent the effects of grid anisotropy. The model is demonstrated on fully-developed, incompressible channel flow and shown to be very promising.

show abstract

Section: B Active Energy Transfermentioning

confidence: 99%

Towards a Predictive Hybrid RANS/LES Framework

Haering

Oliver

Moser

2019

AIAA Scitech 2019 Forum

View full text Add to dashboard Cite

show abstract

“…The bump leading-edge fillet radius could have an important effect on the results as it determines how fast the flow accelerates over the front portion of the bump. In the absence of absolute knowledge, and to facilitate comparison with the results of Spalart et al, 8,9 we use the value chosen by them, which is R 2 = 18.30 cm. This allows a more meaningful comparison with their results.…”

Section: Test Case: Transonic Shock-induced Flow Separationmentioning

confidence: 99%

“…The better performance of the SST model in this problem is not surprising since the data from the Bachalo-Johnson experiment was used in the calibration of the model coefficients. To our knowledge, Spalart et al 8,9 are the first group that used WMLES to study the Bachalo-Johnson flow. Their WMLES methodology is the same as that by Shur et al, 3 which reported good success in the NASA wall-mounted hump problem.…”

Section: Introductionmentioning

confidence: 99%

“…Despite this success, the same methodology faced difficulty in correctly computing the shock location in the Bachalo-Johnson flow. 8,9 This suggests that a particular wall model that works well under certain flow conditions may fail under different flow conditions. The incorrect prediction of the shock position by WMLES in the Bachalo-Johnson flow 8, 9 leads to incorrect flow separation and reattachment locations.…”

Section: Introductionmentioning

confidence: 99%

“…The incorrect prediction of the shock position by WMLES in the Bachalo-Johnson flow 8, 9 leads to incorrect flow separation and reattachment locations. Faced with the difficulty of accurately predicting the shock position in a WMLES performed using nearly 1.7 billion grid points, Spalart et al 8,9 had to resort to a hybrid simulation (with about 8.45 billion points) in which the attached flow region was computed using direct numerical simulation (DNS) while the separated region had to be simulated using delayed detached eddy simulation (DDES) due to computational resource limitations. This hybrid simulation, which covered only 15 degrees of azimuth, was able to improve the shock position prediction.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Wall-Resolved Large-Eddy Simulations of Transonic Shock-Induced Flow Separation

Uzun

Malik

2019

AIAA Journal

View full text Add to dashboard Cite

This paper reports the wall-resolved large eddy simulations of shock-induced boundary layer separation over an axisymmetric bump for a flow Mach number of 0.875 and a chord-based Reynolds number of 2.763 million. The incoming boundary layer has a momentum-thickness Reynolds number of 6600 at one and a half chord lengths upstream of the leading edge. The calculations simulate the experiment by Bachalo and Johnson (AIAA Journal, Vol. 24, No. 3, 1986), except that the tunnel walls are ignored and the simulations are performed assuming free air with as many as 24 billion grid points. The effects of domain span, grid resolution and time step on the predictions are examined. The results are found to show some sensitivity to the studied parameters. Owing to the outer boundary conditions, the predicted surface pressure distribution as well as the flow separation and reattachment locations tend to agree better with the experimental results from the larger (6 × 6 ft) tunnel than those from the smaller (2 × 2 ft) tunnel. The predicted Reynolds shear stress profiles in the separated region differ by as much as 31% from the experimental results that were only obtained in the smaller tunnel. The most accurate surface pressure distribution obtained in this study lies within the scatter of the measurements taken in the two facilities.

show abstract

Simulation of a turbulent flow subjected to favorable and adverse pressure gradients

Uzun

Malik

2021

Theor. Comput. Fluid Dyn.

View full text Add to dashboard Cite

Large-Eddy and Direct Numerical Simulations of the Bachalo-Johnson Flow with Shock-Induced Separation

Cited by 48 publications

References 11 publications

Towards a Predictive Hybrid RANS/LES Framework

Towards a Predictive Hybrid RANS/LES Framework

Wall-Resolved Large-Eddy Simulations of Transonic Shock-Induced Flow Separation

Simulation of a turbulent flow subjected to favorable and adverse pressure gradients

Contact Info

Product

Resources

About