On the Large-Eddy Simulation modelling of wind turbine dedicated airfoils at high Reynolds numbers

Calafell, J.; Lehmkuhl, O.; Rodríguez, I.; Oliva, A.

doi:10.1615/ichmt.2012.procsevintsympturbheattransfpal.1460

Cited by 5 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…10,11,32 In general, mesh resolution and the size of the domain are crucial factors. 8,[33][34][35] Until now most investigations at high Reynolds numbers concern low aspect ratio (AR) studies, 8,[33][34][35][36][37][38][39] although it is generally agreed that high AR simulations are required if the wake characteristics are to be captured. SC simulations also need to be of sufficiently high AR, 17,18,[40][41][42] if the phenomenon needs to be captured.…”

Section: Introductionmentioning

confidence: 99%

Investigation of the three-dimensional flow past a flatback wind turbine airfoil at high angles of attack

Μανωλέσος

Papadakis

2021

Physics of Fluids

View full text Add to dashboard Cite

Flatback airfoils are airfoils with a blunt trailing edge. They are currently commonly used in the inboard part of large wind turbine blades, as they offer a number of aerodynamic, structural, and aeroelastic benefits. However, the flow past them at high angles of attack (AoA) has received relatively little attention until now. This is important because they usually operate at high AoA at the inboard part of Wind Turbine blades. The present investigation uses Reynolds averaged Navier-Stokes (RANS) and hybrid RANS þ large eddy simulation predictions to analyze the flow in question. The numerical results are validated against previously published wind tunnel experiments. The analysis reveals that to successfully simulate this flow, the spanwise extent of the computational domain is crucial, more so than the selection of the modeling approach. Additionally, a low-drag regime observed at angles of attack before stall is identified and analyzed in detail. Finally, the complex interaction between the three-dimensional separated flow beyond maximum lift (stall cells) with the vortex shedding from the blunt trailing edge is revealed.

show abstract

Section: Introductionmentioning

confidence: 99%

Investigation of the three-dimensional flow past a flatback wind turbine airfoil at high angles of attack

Μανωλέσος

Papadakis

2021

Physics of Fluids

View full text Add to dashboard Cite

show abstract

“…Figure 8. On the left, a DU-93-W-210 mesh of 19 × 10 6 CV used in previous works [14] is shown. On the right, the current mesh of 3.9 × 10 6 is displayed together with the wall domain mesh.…”

Section: The Du-93-w-210 Case Resultsmentioning

confidence: 99%

“…Hence, it can be concluded that in this case, the use of the dynamic wall model causes a significant improvement of the results. The DU-93-W-210 wind turbine airfoil, has already been studied in previous works [14] at various angles of attack (AoA) and at Reynolds number of Re = U 0 c/ν = 3 × 10 6 , where U 0 is the free stream velocity, c is the airfoil chord and ν is the fluid viscosity. In those previous works, a fine mesh of 19 × 10 6 CV was used together with the VMS-WALE as a subgrid-scale model.…”

Section: Conclusion From the Ahmed Car Testmentioning

confidence: 99%

A dynamic wall model for Large-Eddy simulations of wind turbine dedicated airfoils

Calafell¹,

Lehmkuhl²,

Carmona³

et al. 2014

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Abstract. This work aims at modelling the flow behavior past a wind turbine dedicated airfoil at high Reynolds number and large angle of attack (AoA). The DU-93-W-210 airfoil has been selected. To do this, Large Eddy Simulations (LES) have been performed. Momentum equations have been solved with a parallel unstructured symmetry preserving formulation while the wall-adapting local-eddy viscosity model within a variational multi-scale framework (VMS-WALE) is used as the subgrid-scales model. Since LES calculations are still very expensive at high Reynolds Number, specially at the near-wall region, a dynamic wall model has been implemented in order to overcome this limitation. The model has been validated with a very unresolved Channel Flow case at Reτ = 2000. Afterwards, the model is also tested with the Ahmed Car case, that from the flow physics point of view is more similar to an stalled airfoil than the Channel Flow is, including flow features as boundary layer detachment and recirculations. This case has been selected because experimental results of mean velocity profiles are available. Finally, a flow around a DU-93-W-210 airfoil is computed at Re = 3 × 10 6 and with an AoA of 15• . Numerical results are presented in comparison with Direct Numerical Simulation (DNS) or experimental data for all cases.

show abstract

“…Numerical study of the flow past FB airfoils is particularly challenging due to its unsteady and three-dimensional (3D) nature, which includes impingement, separation and vortex shedding. Prediction of force coefficients and wake characteristics depend heavily on the fidelity of the numerical approach, the mesh resolution and the size of the domain (Calafell et al, 2012;Lehmkuhl et al, 2014;Stone et al, 2009). It is generally agreed that two-dimensional and Reynolds-Averaged Navier Stokes (RANS) simulations nonphysically damp the flow (Stone et al, 2006) and that higherfidelity approaches might be more suitable (Standish and Van Dam, 2003), especially if the flow frequencies in the wake are of interest.…”

Section: Introductionmentioning

confidence: 99%

“…It is generally agreed that two-dimensional and Reynolds-Averaged Navier Stokes (RANS) simulations nonphysically damp the flow (Stone et al, 2006) and that higherfidelity approaches might be more suitable (Standish and Van Dam, 2003), especially if the flow frequencies in the wake are of interest. There have been a number of investigations at high Re numbers (Calafell et al, 2012;Kim et al, 2014;Lehmkuhl et al, 2014;Standish and Van Dam, 2003;Stone et al, 2009;Xu et al, 2014). However, most deal with 2D or low-aspect-ratio simulations and not with TE devices for flow control.…”

Section: Introductionmentioning

confidence: 99%

The flow past a flatback airfoil with flow control devices: benchmarking numerical simulations against wind tunnel data

Papadakis

Μανωλέσος

2020

Wind Energ. Sci.

View full text Add to dashboard Cite

Abstract. As wind turbines grow larger, the use of flatback airfoils has become standard practice for the root region of the blades. Flatback profiles provide higher lift and reduced sensitivity to soiling at significantly higher drag values. A number of flow control devices have been proposed to improve the performance of flatback profiles. In the present study, the flow past a flatback airfoil at a chord Reynolds number of 1.5×106 with and without trailing edge flow control devices is considered. Two different numerical approaches are applied, unsteady Reynolds-Averaged Navier Stokes (RANS) simulations and detached eddy simulations (DES). The computational predictions are compared against wind tunnel measurements to assess the suitability of each method. The effect of each flow control device on the flow is examined based on the DES results on the finer mesh. Results agree well with the experimental findings and show that a newly proposed flap device outperforms traditional solutions for flatback airfoils. In terms of numerical modelling, the more expensive DES approach is more suitable if the wake frequencies are of interest, but the simplest 2D RANS simulations can provide acceptable load predictions.

show abstract

On the Large-Eddy Simulation modelling of wind turbine dedicated airfoils at high Reynolds numbers

Cited by 5 publications

References 9 publications

Investigation of the three-dimensional flow past a flatback wind turbine airfoil at high angles of attack

Investigation of the three-dimensional flow past a flatback wind turbine airfoil at high angles of attack

A dynamic wall model for Large-Eddy simulations of wind turbine dedicated airfoils

The flow past a flatback airfoil with flow control devices: benchmarking numerical simulations against wind tunnel data

Contact Info

Product

Resources

About