Thermal stratification in cfd modelling for wind resource assessment

0

40

80

120

160

200

0,0 0,1 0,2 0,3 0,4 0,5

He

igh

t (m

)

S04_model

S05_model

measurement_100<L<400


measurement_20<L<60

1. C. Albrecht, D. Delaunay, A. Grözner, M. Kohiert et L. Pauscher, «Atmosphericstability: general classification and simulation in Meteodyn WT,» in EWEA, 2014.

2. A. Grachev, E. Andreas, C. Fairall, P. Guest et P. Persson, «The critical richardsonnumber and limits of applicability of local similarity theory in the stable boundarylayer,» Boundary Layer Meteorology, vol. 147, pp. 51-82, 2013.

3. P. J. Hurley, «An evaluation of several turbulence shemes for the prediction of meanand turbulent fields in complex terrain,» Boundary Layer Meteorology , vol. 83,n° 1, pp. 43-73, 1997.

4. S. S. Zilitinkevich, T. Eleperin, N. Kleeorin, I. Rogachevskii, V. L'vov, I. Esau et R.Kouznetsov, «Turbulence closure for stably stratified flows: local and non-localformulations,» in 8th European conference on applied climatology, Zürich,Switzerland, 2010.

Up to know CFD computations in wind resource assessment mainly focused on wind statisticstreatment, and then considered average thermal structure of the atmosphere. With anincreasing demand for a more accurate description of these statistics, including time series,there is a need for considering more specific situations, and particularly stable thermalstratifications (ref 1). We present here a new turbulence model allowing to consider thestrongly stable cases in CFD computations. This model is implemented in Meteodyn WT 5.2

Applying a k-L multi-layer model in CFD computations is a promising way to reproducebehavior of the Stable Boundary Layer, especially when the height of the SBL becomesinferior to the wind turbine hub height. This is an important step forward for a betterconsideration of the atmospheric stability in wind resource assessment.

Abstract

Objectives

Conclusions

References

The objective of this work is to develop a new turbulence model for the Stable BoundaryLayer, by considering measurements at two sites with well documented measurementsup to 200 m height above ground.

A new approach for CFD modelling in stable situations

Figure 2: Meteodyn WT simulations of wind profile at Cabauw using the new turbulent multi-layer viscosity model

The main effect of stable stratification in the atmospheric boundary layer is to reduce thevertical turbulent exchanges. In CFD computations these exchanges are modelled viathe concept of turbulent viscosity, which is the product of a wind speed scale and aturbulent length scale. The model is focusing on the determination of the length scalevertical profile, allowing a multi-layer approach, adapted to stable situations.

The model considers 3 layers: The Monin-Obuknov similarity theory (MOST) layer, thelocal z-less model, and the layer above the SBL (ref 1, 2,3).

A first calibration has been made with 3 years data at Cabauw (NL) (flat open terrain),with data sorted according to a Stability Index (z1= 10 m and z2= 80 m), given by:

A verification has been conducted on a more complex site at Rödeser Berg (hilly andforested) with 1 year data sorted according to the measured Monin-Obukhov length L at40 m height. A dense forest lies in the South-West and South sectors. In the East sector,the roughness is more moderate.

21212

1212

1 )/())()((

)/())()((

zzzVzV

zzzzgSI

S03 S04 S05 S06 S07 S08 S09 S10 S11 S12 S13 S14SI range 0.00 ‐ 0.01 0.01‐0.05 0.05‐0.10 0.10‐0.15 0.15‐0.20 0.20‐0.25 0.25‐0.32 0.32‐0.40 0.40‐0.50 0.50‐0.70 0.70‐1.00 > 1.00

frequency 7.3% 9.4% 14.4% 11.2% 9.0% 6.9% 6.7% 4.5% 3.4% 3.1% 1.6% 1.9%

power contribution 9.1% 24.1% 21.4% 9.6% 6.0% 4.0% 3.5% 2.1% 1.4% 1.0% 0.4% 0.2%

(V80+V140)/2 (m/s) 8.1 11.3 9.5 8.0 7.4 7.1 6.8 6.5 6.2 5.8 5.2 4.3

Table 1: SI classes and characteristics at Cabauw (wind direction: 255-285 deg)

Considering thermal stratification in CFD modelling for wind resource assessment

Didier DELAUNAY1, Ru LI1, Soner HARNUBOĞLU2

1 METEODYN, 14 bd Winston Churchill, 44100 Nantes, France2 UCYEL ENERJI, Yasamkent Mahallesi 3118 Cadde No 33, Cankaya 06810 Ankar, Turkey

PP008PP

008

Comparison with measurements at Rödeser Berg

Comparison with measurements at Cabauw

Figure 1: Views of Rödeser Berg and Cabauw sites

Figure 4: Meteodyn WT simulations of wind profile and TKE in Rödeser Berg (Southwest Direction)

0

40

80

120

160

200

0,0 0,1 0,2 0,3 0,4 0,5

He

igh

t (m

)

S04_model

S05_model



measurement_20<L<60

√ / _

0

40

80

120

160

200

0,0 0,2 0,4 0,6 0,8 1,0

He

igh

t (m

)

Normalized wind speed (m/s)

S04_model

S05_model



measurement_20<L<60

Figure 6: Meteodyn WT simulations of wind profile and TKE in Rödeser Berg (East Direction)

0

40

80

120

160

200

0,0 0,2 0,4 0,6 0,8 1,0

He

igh

t (m

)

Normalized wind velocity (m/s)

S04_model

S05_model

S06_model



measurement_20<L<60

Figure 5: Meteodyn WT simulations of wind profile and TKE in Rödeser Berg (South Direction)

0

40

80

120

160

200

0,0 0,2 0,4 0,6 0,8 1,0

He

igh

t (m

)

Normalized wind velocity (m/s)

S04_model

S05_model



measurement_20<L<60

Acknowledgements

In the East direction, the wind speed gradients increase according to the thermal stability(L decrease) as computed in Meteodyn WT, but the measuured TKE are less dependanton stability than expected in the computations. Calibrations on the dissipation terms inthe TKE equation will have to reduce this discrepancy.

In the W and SW direction, the surface friction generated by the tree weakens thethermal effect for the wind speed gradient, but the TKE is strongly affected by stability.Globally, the results of the Meteodyn WT computations are quite satisfactory.

We acknowledge Anselm Grötzner (Cube Engineering) for his help for Cabauw dataanalysis and Tobias Klaas (IWES Fraunhöfer) for the data at Röderser Berg.

Rödeser BergCabauw

0

40

80

120

160

200

0 0,05 0,1 0,15 0,2

He

igh

t (m

)

S05_model

S06_model

S07_model

measurement_S05

measurement_S06

measurement_S07

√ / _

Figure 3: Meteodyn WT simulations of TKE at Cabauw using the new turbulent multi-layer viscosity model

0

40

80

120

160

200

0 0,05 0,1 0,15 0,2

He

igh

t(m

)

S08_model

S09_model

S10_model

measurement_S08

measurement_S09

measurement_S10

√ / _

0

40

80

120

160

200

0,0 0,1 0,2 0,3 0,4 0,5

He

igh

t (m

)

S04_model

S05_model

S06_model



measurement_20<L<60

√ / _