Simulation of the flow around a circular cylinder at  Re =3900 with Partially-Averaged Navier-Stokes equations

Pereira, Filipe; Vaz, Guilherme; Eça, Luís; Girimaji, Sharath S.

doi:10.1016/j.ijheatfluidflow.2017.11.001

Cited by 72 publications

(49 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Very low values of the dynamic parameter -f k are observed for both these cases, resulting in insufficient turbulence modelling as the unresolved kinetic energy is very low (Krajnovi c et al (2016)). In such cases, the influence of the numerical scheme plays a larger role (see Pereira et al (2018b)), and CDS accurately predicts the flow asymmetry in the vertical midplane.…”

Section: Discussionmentioning

confidence: 99%

“…In such cases, the modelling of the small unresolved scales will not significantly impact the overall solution, and the numerical errors would mainly contribute to the overall accuracy (Basara et al (2018), Krajnovi c et al (2016), Park et al (2004)). More recently, Pereira et al (2018b) observed that the dependence on the spatial and temporal resolution is larger when f k → 0. For the GTS at G=H ¼ 1:1, the spatial and temporal requirements are satisfied by the use of LES-capable grids and very low time-step (the non-dimensionalised time-step for the current simulations is ' 0:002), respectively.…”

Section: β ¼ 0 ∘mentioning

confidence: 98%

“…At a gap height of G=H ¼ 1:1, both the input and output f k contours are lower, primarily on account of the unresolved kinetic energy being two orders of magnitude lower as compared to the total kinetic energy, thereby resulting in very low values of the output f k . In such cases, there is insufficient turbulence modelling, and the numerical scheme plays a larger role (Krajnovi c et al (2016b); Basara et al (2018); Pereira et al (2018b)). Furthermore, the values of the unresolved kinetic energy are much lower when PANS -CDS is used as compared to the PANS -ASB scheme.…”

Section: Appendix Cmentioning

confidence: 99%

See 2 more Smart Citations

Investigation of the near-wake flow topology of a simplified heavy vehicle using PANS simulations

Rao

Minelli

Zhang

et al. 2018

Journal of Wind Engineering and Industrial Aerodynamics

View full text Add to dashboard Cite

The near-wake flow topology of a ground transportation system (GTS) is investigated using partially-averaged Navier-Stokes (PANS) simulations at Re ¼ 2:7 Â 10 4. Recent numerical investigations for the GTS model using large eddy simulations (LES) showed an anti-symmetric flow topology (flow state II) in the vertical midplane compared to that observed in previous experimental studies (flow state I). The geometrical configuration of the GTS permits bi-stable behaviour, and the realisation of each of the two flow states, which are characterised by an asymmetrical flow topology, is achieved by varying the differencing scheme for the convective flux in the PANS simulations; AVL SMART schemes predict flow state I, while central differencing scheme (CDS) predicts flow state II. When the GTS model was placed away from the ground plane, the AVL SMART scheme fails to predict the flow asymmetry resulting in a pair of symmetrical vortices in the vertical midplane, while flow state II topology is observed when CDS is used. The switch from flow state I (II) to flow state II (I) is achieved by changing the numerical scheme from AVL SMART (CDS) to CDS (AVL SMART), with an intermediate transient-symmetric (TS) state being observed during the switching process. The numerical scheme in the PANS simulations thus plays a critical role in determining the initial flow topology in the near wake of the GTS.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: β ¼ 0 ∘mentioning

confidence: 98%

Section: Appendix Cmentioning

confidence: 99%

See 1 more Smart Citation

Investigation of the near-wake flow topology of a simplified heavy vehicle using PANS simulations

Rao

Minelli

Zhang

et al. 2018

Journal of Wind Engineering and Industrial Aerodynamics

View full text Add to dashboard Cite

show abstract

“…This allows for flexibility in the method as RANS modelling can provide sufficient levels of Reynolds stresses when the grid is not adequate to resolve the turbulence. PANS has previously been used for several different bluff body flows, such as flows around cubes (Krajnović et al, 2012a(Krajnović et al, , 2016, cylinders (Lakshmipathy and Girimaji, 2010;Jeong and Girimaji, 2010;Luo et al, 2014;Pereira et al, 2018) and simplified vehicles (Mirzaei et al, 2015;Han et al, 2013;Rao et al, 2018;Krajnović et al, 2012b). Most of these studies show that flow predictions of PANS are in good agreement with the experimental data, and often better than LES, when the mesh is relatively coarse.…”

Section: Introductionmentioning

confidence: 99%

Comparison of PANS and LES of the flow past a generic ship

et al. 2018

View full text Add to dashboard Cite

The prediction of flow past a generic ship has been performed using partially-averaged Navier-Stokes (PANS) turbulence model. The Reynolds number based on the width of the ship is × 8 10 4. A detailed comparison with resolved large-eddy simulation (LES) solution and available experimental data is made for a better understanding of the capability of PANS in predicting the turbulent flow in the wake. The results show that PANS produces similar trends to those predicted by LES. However, a coarse grid resolution will lead PANS to act as unsteady Reynolds-averaged Navier-Stokes (URANS). Although the geometry is symmetric for the incoming flow at a zero yaw angle, bi-stable behaviour is observed in the wake, contributing to an asymmetric flow distribution on the heli-deck.

show abstract

“…An estimation of physical resolution of the SRS simulations is crucial for an accurate prediction of flow dynamics. The influence of physical resolution on the accurate prediction was intensively investigated by Pereira et al [3,4]. Results of the studies were that PANS simulations with a low physical resolution (f k > 0.5) were unable to correctly capture flow dynamics for the transition of shear layers in case of the flow past a circular cylinder.…”

Section: Introductionmentioning

confidence: 99%

Partially-Averaged Navier-Stokes Simulation of the Flow Around Circular Cylinder at Re=3900: Comparison of Low and High Reynolds Number Approaches

Korinek¹,

Tisovský²,

Hujer³

et al. 2020

Topical Problems of Fluid Mechanics 2020

View full text Add to dashboard Cite

An appropriate choice of physical resolution for the Scale-Resolved Simulations (SRS) method is crucial for the accurate prediction of flow dynamics. The present study deals with numerical simulations of the flow past a circular cylinder using the Partially-Averaged Navier-Stokes bridging method. A subcritical Reynolds number 3900 is investigated at various physical resolution (the ratio of unresolved to total resolved turbulent kinetic energy f k) involving two approaches to estimate the ratio of unresolved to total resolved eddy dissipation rate fε, in particular a low Reynolds number fε = f k and a high Reynolds number fε = 1 approaches. Good agreement with experimental and DNS results was achieved for a higher physical resolution f k ≤ 0.5, fε = 1. The low Re approach led to the earlier transition of the separated boundary layer which resulted in a short recirculation region. Additionally, predicted results from simulations with the low Re approach were not influenced by a refinement of the computational grid.

show abstract

Simulation of the flow around a circular cylinder at Re =3900 with Partially-Averaged Navier-Stokes equations

Cited by 72 publications

References 38 publications

Investigation of the near-wake flow topology of a simplified heavy vehicle using PANS simulations

Investigation of the near-wake flow topology of a simplified heavy vehicle using PANS simulations

Comparison of PANS and LES of the flow past a generic ship

Partially-Averaged Navier-Stokes Simulation of the Flow Around Circular Cylinder at Re=3900: Comparison of Low and High Reynolds Number Approaches

Contact Info

Product

Resources

About