wind resource, turbulence and electricity production in...
TRANSCRIPT
![Page 1: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/1.jpg)
Mattias Mohr, Johan Arnqvist, Hans Bergström
Uppsala University (Sweden)
Wind resource, turbulence and electricity
production in forests
![Page 2: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/2.jpg)
Project:
Wind Power in Forests II - Forestwind
• Continuation of Wind Power in Forests project
• Better estimation of wind resource
• Better estimation of turbine loads (wind shear/veer,
turbulence, inhomogeneous forests)
• Quantify effects on energy production
Use and develop models for these purposes
![Page 3: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/3.jpg)
Problems
• What is the wind resource over Swedish forests at
150 m height above ground?
• What about the loads?
• Is energy production affected?
Forest height depends on mainly latitude and altitude.
New laser scans give very detailed picture of Swedish forest.
Swedish forest is inhomogeneous (”patchwork”)!!!
![Page 4: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/4.jpg)
Old mast: Ryningsnäs
140m high mast
18m high
mast
T1, T2 = wind turbines
![Page 5: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/5.jpg)
New mast: ”Mossen”, Småland
One important question: Can Sodar & Lidar replace masts in forests?
180 m mast
with
turbulence
instruments
![Page 6: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/6.jpg)
Wind resource assessment
• Better estimations of roughness (z0),
displacement height (d) and leaf area density
improves models (mesoscale and CFD)
• 42 sites in Wind power in forests I (final report) :
d = 16 m; z0 = 1.4 m (median values)
![Page 7: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/7.jpg)
Wind profile over forest
• Correct input
parameters crucial
for models!
• Especially: How does
inhomogeneous
influence parameters?
![Page 8: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/8.jpg)
Modelling of forest:
Different approaches
1. Mesoscale models simulate flow down to
displacement height (with drag = f(z0))
2. CFD models simulate flow all the way down
(with drag = f(leaf area density))
Can CFD-type forest parameterisations improve mesoscale
models? Maybe
Can mesoscale models be used instead of CFD models
for micro siting Maybe
For small clearings/forests approach 1 does not work!
![Page 9: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/9.jpg)
Drag parameterisations
𝜕𝑢
𝜕𝑡= …− Cd · LAD · | 𝑢horizontal | · 𝑢 (same for v-component)
𝑢 = horizontal wind vector, 𝑢 = wind component W-E
![Page 10: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/10.jpg)
Example results: WRF mesoscale
model
x [km]
z [
m]
Wind Speed Reduction over Forest
-10 -5 0 5 100
50
100
150
200
250
300
0
1
2
3
4
5
6
7
8
9
10
11
WIND
Forest edge
![Page 11: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/11.jpg)
Forest leaf area density
problematic
How should
we average
data from all
pixels?
Forest leaf area density profile different to theoretical profile!
Theoretical profile
(for mature
homegeneous
forest)
Figur: Johan Arnqvist
![Page 12: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/12.jpg)
Better input parameters for models
• Nationwide maps of
– Surface roughness
– Displacement height
– Leaf are density = f(height)
from airborne laser scans
• Suggestions of averaging
methods
Tree height from laser scan
Figur: Johan Arnqvist
![Page 13: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/13.jpg)
Turbulence & loads
• IEC standard exceeded, but does it matter???
Ryningsnäs
Figure: Johan Arnqvist
![Page 14: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/14.jpg)
Turbulence simulations
• LES = Large Eddy Simulation model (simulates
largest turbulent vortices)
• Forced by data
from WRF
mesoscale
model
• Methods have
to be deve-
loped for this Figure: Bastian Nebenführ (Chalmers)
Wind field during ≈ 1 second
![Page 15: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/15.jpg)
Energy Production in Forests
Power Curve Working Group identified four main
factors:
1. Air density
2. Vertical wind shear
3. Vertical wind veer
4. Turbulence intensity
5. (Directional variation)
6. (Vertical inflow angle)
Especially high above
forests and in Sweden!!!
Factors 2 - 4 probably most important in South.
Factor 3 probably most important in North.
![Page 16: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/16.jpg)
Wind shear effects on power curve
• As turbines get bigger and bigger, important to
consider shear and veer! Ekman layer
• Rotor equivalent wind speed with wind veer:
𝑈𝑒𝑞𝑢𝑖𝑣 = (𝑈 𝑖cos(𝜙𝑖))3 ∙𝐴𝑖𝐴
𝑖
1/3
![Page 17: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/17.jpg)
Turbulence effects on power curve
• Energy gain/loss depending on curvature of
power curve
𝑃 = 𝑃 𝑢 +1
2
𝑑2𝑃(𝑢 )
𝑑𝑢2𝜎𝑢2
Vestas V90-3.0 MW
+
-
≈ 0
Power curve
valid for 10
minute
averages,
but wind
speed
variations
due to
turbulence
wind speed variance
![Page 18: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/18.jpg)
Förutsätt- ningar
Miljö & Samhälle
Drift &
Underhåll
Nätintegration
Projektering & Etablering
Konstruktion & Produktion
THANK YOU FOR YOUR ATTENTION!
![Page 19: Wind resource, turbulence and electricity production in ...standupforwind.se/digitalAssets/491/491178_3vindk... · Wind resource assessment • Better estimations of roughness (z](https://reader030.vdocument.in/reader030/viewer/2022040411/5ed67c5eff0e593c0b63ee05/html5/thumbnails/19.jpg)
References
• Halldin, S. (1985): Leaf and bark area distribution in a pine forest. In The forest atmosphere interaction, edited by B.
A. Hutchison and B. B. Hicks (Dordrecht: Reidel Publishing Company), p. 39–58.
• Lalic, B. and D. T. Mihailovic (2004): An Empirical Relation Describing Leaf-Area Density inside the Forest for
Environmental Modeling. Journal of Applied Meteorology, Notes and Correspondence, Vol. 43, p. 641-645.
• Hans Bergström, Henrik Alfredsson, Johan Arnqvist, Ingemar Carlén, Ebba Dellwik, Jens Fransson, Hans
Ganander, Matthias Mohr, Antonio Segalini and Stefan Söderberg: Wind power in forests. Winds and effects on
loads Elforsk rapport 13:09.
• Power Curve Working Group: http://www.pcwg.org/
• Matthias Mohr, Wasana Jayawardena, Johan Arnqvist and Hans Bergström: Wind energy estimation over forest
canopies using WRF mesoscale model. EWEA Annual Event 2014. Poster presentation.