Environmental Engineering Reference
In-Depth Information
change of the gears, two gearbox bearings were replaced in all the wind turbines. The nacelle crane
was used, and average downtime was 48 h per turbine.
8.4 WAKE EFFECTS
The wake is expanding and the wind speed is reduced downwind, so if there are multiple wind
turbines, how far apart should they be placed? Also, vortices are generated from the tips of blades,
trailing edges of blades, and by the tower, and they increase the turbulence of the wake. So tips of
airfoils and trailing edges are designed to reduce the vortices and to reduce the noise accompanying
some vortices. The three primary methods of wake and array loss research have been numerical
modeling, wind tunnel simulations, and field measurement. A database of literature on wind turbine
wakes and wake effects through 1990 is available [8].
The wakes from wind turbines create turbulence and, along with the wind speed deficit,
result in array losses, which are reflected in reduced annual energy production. Therefore, the
placement of wind turbines in a wind farm is a trade-off between energy production and cost
of installation. There will be reduced energy produced by downwind units, so the question is
how much reduction for what spacing (within a row and between rows). In fairly flat areas, the
rows will be placed perpendicular to the predominant wind direction, and within row spacing
is two to four rotor diameters, and between row spacing is five to ten rotor diameters. Offshore
wind farms generally have larger spacing; for example, Horns Rev in the North Sea off the coast
of Denmark has a seven-rotor diameter spacing (within row and between rows). The physical
factors controlling wake interference are downwind spacing, power extracted by the wind tur-
bines, turbulence intensity, and atmospheric stability. Wind turbine wakes develop according to
fairly well-defined regions at different downwind distances, and wake geometry models show
this information [9]. Field tests on single and multiple wind turbines measured the velocity and
power deficit downwind. The wake effects are still noticeable at ten rotor diameters downwind
from a rotor. Wind turbines had close spacing between rows at wind farms in San Gorgonia Pass,
California, due to the high cost of land, and energy production was reduced for the second row
and even more for the third row, as it had the wake effects from both the first and the second
row. Field measurements of wake effects inside of wind farms have generally been limited to
two to four rows of wind turbines. Energy deficits of 10-15% in row 2 and 30-40% in row 3 have
been reported for densely packed wind farms. Measurements of wake deficits downwind of large
arrays indicate that the losses may be larger and extend farther downwind than expected. Energy
deficits of 15% were estimated at 5 km downwind from a 50 MW array [10]. Early wind turbines
were small, 25-100 kW, and later some larger wind turbines on taller towers were interspaced
within a row.
It is more difficult to predict output and array losses without an extensive wind measurement
program within the wind farm. There is an exception, and that is for offshore wind farms, as ocean
waves provide data on wind speeds at 10 m height determined from satellite data (see Section 4.4 ).
High-resolution data are used to estimate the wind resource of the Danish Seas. There has been
some comparison of those data with met tower data taken offshore. Ocean wind maps covering the
Horns Rev wind farm (400 m grid cells) in the North Sea and the Nysted wind farm (1.6 km grid
cells) in the Baltic Sea were used to quantify the wake effect [11]. The Horns Rev wind farm has
eighty turbines (80 m diameter, 2 MW) with an 8 by 10 array and a distance of 560 m between the
turbines (7D spacing). The Nysted wind farm has seventy-two turbines (82.4 m diameter, 2.3 MW)
with a 9 by 8 array, and the distance between turbines within a row is 480 m and between rows is
850 m (5.8D by 10.3D). The velocity deficit is around 10% at 0 to 3 km downwind, and the wind
recovers to 2% of the upstream values at around 5 to 20 km downstream, which depends on the
ambient wind speed, atmospheric stability, and the number of operating turbines [12]. The recovery
is faster for unstable than for near-neutral conditions. In calm winds the turbines are clearly visible
in the ocean wind speed maps.
Search WWH ::




Custom Search