Design limits and solutions for very large wind turbines

In relation to the coming IEC standard on performance
verifi cation with the use of nacelle anemometry [8],
nacelle anemometry has been studied, both the
theory and its practical application [9]. An alternative
to nacelle anemometry has also been developed, the
so-called spinner anemometer [10, 11]. This type of
sensor seems to avoid the draw-backs of nacelle anemometry
because the sensors are positioned in front
of the rotor on the spinner. Finally, mast correction
measurement has also been developed to improve
current measurements [12].
With respect to new wind measurement technologies
such as remote sensing (LIDAR & SODAR), a substantial
number of projects have been undertaken over the
last fi ve years. Ground based LIDARS have been developed
signifi cantly, and in the past few years calibration
and traceability issues have also been tackled and are
now being included in the IEC standard revision [5].
An overview is provided by [6], where signifi cant metrology
contributions are described.
Focus on LIDAR technology
LIDAR technology is relatively recent and requires
testing. Testing against a traditional meteorological
mast has been shown to be effi cient for gaining
confi dence and trust in measurement accuracy.
In principle, LIDAR measurements could be made
traceable through the fundamental measurement
principles, but at this stage of development it is not
feasible. Instead, traceability is secured through
comparison with meteorological masts that are
themselves traceable through wind tunnel calibrations
of cup anemometers. LIDARs can fulfil different
objectives:
For resource assessment, and replacing the traditional
measurement masts, UpWind demonstrated
that the ground-based LIDAR measurement principle
is effi cient in fl at terrain. In complex terrain and
close to woods, the measurement volume is disturbed
by a vertical turbulence component. Due to a
large measurement volume, ground-based LIDARs
perform a spatial averaging which has the effect of a
low pass filter on turbulence measurements. This
effect requires special attention and correction
methods, which were analysed within the UpWind
project by the use of the WAsP engineering tool,
which is a computer based program for the estimation
of extreme wind speeds, wind shears, wind profiles
and turbulences in complex (and simple) terrain.
LIDAR measurements were made from a rotating
spinner. The analysis show good perspectives for
scanning the incoming wind, which may lead to better
controlled wind turbines.
LIDARs have also been used to scan the wake of
wind turbines. These measurements show the
curvy wake pattern. Progress has been made with
large-scale site wind scanning with three coordinated
LIDARs.
Verification of anemometer calibrations
In the context of MEASNET 9 a cup anemometer calibration
round robin and acoustic noise measurement
round robin were performed [13] with good results for
the associated MEASNET institutes. Together with
the MEASNET experts, UpWind worked to try and
find a procedure for calibrating sonic anemometers.
The operational characteristics of 26 sonic anemometers
have been investigated over a long period of time
by [17]. Improvements were required in order to use the
sensors for wind energy applications. These improvements
are being implemented by the manufacturer.
9
Measnet: International network for harmonised and recognised measurements in wind energy.
http://www.measnet.com, MEASNET Round Robin (RR) on anemometer calibration is one method of measurement.
Design limits and solutions for very large wind turbines
29