scispace - formally typeset
Search or ask a question

Showing papers by "Parviz Moin published in 2012"


Journal ArticleDOI
TL;DR: In this paper, the authors modified the resolution requirements for large eddy simulation (LES) using accurate formulae for high Reynolds number boundary layer flow and showed that the number of grid points required for wall-modeled LES is proportional to ReLx, where Lx is the flat-plate length in the streamwise direction.
Abstract: Resolution requirements for large eddy simulation (LES), estimated by Chapman [AIAA J. 17, 1293 (1979)], are modified using accurate formulae for high Reynolds number boundary layer flow. The new estimates indicate that the number of grid points (N) required for wall-modeled LES is proportional to ReLx, but a wall-resolving LES requires NReLx13/7, where Lx is the flat-plate length in the streamwise direction. On the other hand, direct numerical simulation, resolving the Kolmogorov length scale, requires NReLx37/14.

583 citations


Journal ArticleDOI
TL;DR: In this paper, a large eddy simulation of H-and K-type transitions in a spatially developing zero-pressure-gradient boundary layer at Ma∞ = 0.2 is investigated using several subgrid scale (SGS) models including constant coefficient Smagorinsky and Vreman models and their dynamic extensions, dynamic mixed scale-similarity, dynamic one-equation kinetic energy model, and global coefficient V reman models.
Abstract: Large eddy simulation of H- and K-type transitions in a spatially developing zero-pressure-gradient boundary layer at Ma∞ = 0.2 is investigated using several subgrid scale (SGS) models including constant coefficient Smagorinsky and Vreman models and their dynamic extensions, dynamic mixed scale-similarity, dynamic one-equation kinetic energy model, and global coefficient Vreman models. A key objective of this study is to assess the capability of SGS models to predict the location of transition and the skin friction throughout the transition process. The constant coefficient models fail to detect transition, but the dynamic procedure allows for a negligible turbulent viscosity in the early transition region. As a result, the “point” of transition is estimated correctly. However, after secondary instabilities set in and result in the overshoot in the skin friction profile, all models fail to produce sufficient subgrid scale shear stress required for the correct prediction of skin friction and the mean velocity profile. The same underprediction of skin friction persists into the turbulent region. Spatially filtered direct numerical simulation data in the same boundary layers are used to provide guidelines for SGS model development and validation.

70 citations


Journal ArticleDOI
TL;DR: Simulation results show that linear interpolation of the equation-of-state causes numerical fluctuations that impede convergence and reduce accuracy in the low-Mach number, variable-density flow code, highlighting the importance of careful code and solution verification in the simulation of variable- density flows.

61 citations


Journal ArticleDOI
TL;DR: In this paper, large-eddy simulations of supersonic jets are performed to validate the development of a second-order finite volume unstructured solver for aeroacoustic applications.
Abstract: Large-eddy simulations of supersonic jets are performed to validate the development of a second-order finite volume unstructured solver for aeroacoustic applications. Two supersonic jets issuing from an axisymmetric nozzle at Mach number 1.4 are computed: one unheated jet with a Reynolds number of 150,000 and one heated jet with a Reynolds number of 76,000 and a temperature ratio of 1.75. Flow and noise results are compared with the experimental database fromNASAGlennResearchCenter. The nozzle is included in the computational domain.The present study shows that the results from the unstructured solver are in good agreement with the experimental data for time-averaged and fluctuating quantities, velocity spectra in the jet, and the sound obtained in the near field and the far field using the integration of the Ffowcs–Williams and Hawkings equation.

51 citations


Journal ArticleDOI
TL;DR: In this article, Wu and Moin compared the statistical properties of the turbulence in transition at Reθ ≈ 300, from individual turbulent spots, and where the spots merge (distributions of the mean velocity, Reynolds stresses, kinetic energy production, and dissipation rates, enstrophy and its components).
Abstract: Using the recent direct numerical simulations by Wu and Moin [“Transitional and turbulent boundary layer with heat transfer,” Phys. Fluids 22, 85 (2010)] of a flat-plate boundary layer with a passively heated wall, statistical properties of the turbulence in transition at Reθ ≈ 300, from individual turbulent spots, and at Reθ ≈ 500, where the spots merge (distributions of the mean velocity, Reynolds stresses, kinetic energy production, and dissipation rates, enstrophy and its components) have been compared to these statistical properties for the developed boundary layer turbulence at Reθ = 1840. When the distributions in the transitional regions are conditionally averaged so as to exclude locations and times when the flow is not turbulent, they closely resemble the distributions in the developed turbulent state at the higher Reynolds number, especially in the buffer layer. Skin friction coefficients, determined in this conditional manner at the two Reynolds numbers in the transitional flow are, of course,...

39 citations


Proceedings ArticleDOI
04 Jun 2012
TL;DR: In this paper, the effect of chevrons on supersonic jet noise was investigated by comparing simulations with and without the addition of Chevrons around the nozzle lip, and the results showed that the chevron effect significantly reduced the amount of noise produced.
Abstract: Fully unstructured large-eddy simulation coupled with a Ffowcs Williams–Hawkings solver was used to predict the noise from underexpanded supersonic jets issuing from rectangular nozzles. The effect of the chevrons was considered by comparing simulations with and without the addition of chevrons around the nozzle lip. The entire flow in and around the nozzle is simulated using the CharLES solver, on unstructured meshes containing up to 528 million control volumes. The massively parallel code was scaled up to as many as 163,840 processors. A mesh sensitivity study was performed, and the far-field noise predictions were compared with measurements conducted at the NASA Glenn Research Center. For the underexpanded isothermal operating conditions considered, chevrons are observed to produce a substantial noise reduction.

32 citations



Journal ArticleDOI
TL;DR: In this article, recent simulations of transition to turbulence in compressible (M 1 = 0.2), zero-pressure-gradient flatplate boundary layers triggered by fundamental (Klebanoff K-type) and subharmonic (Herbert H-type), secondary instabilities of TollmienSchlichting waves are highlighted.
Abstract: In this fluid dynamics video, recent simulations of transition to turbulence in compressible (M1 = 0.2), zero-pressure-gradient flatplate boundary layers triggered by fundamental (Klebanoff K-type) and subharmonic (Herbert H-type) secondary instabilities of TollmienSchlichting waves are highlighted.

4 citations