Abstract:
Current wind turbine performance codes are not yet able to predict the
rotor aerodynamic behaviour with sufficient certainty. This has led to
both the over-design of blades and to operational restrictions in
certain wind conditions.
Essentially the problem is one of aerodynamic stall. Steady
3-dimensional stall can occur near the blade root in high wind
conditions and may produce more power than predicted. Dynamic stall
can also be expected due to the effects of yawed operation,
turbulence, tower shadow and the earth's boundary layer.
The main aim of this work is to provide a coherent set of measured
aerodynamic data accounting for both axial/non-axial flow and stall in
high winds. These measurements are designed to highlight the effects
of both steady and dynamic stall on the rotor aerodynamic performance.
In addition, the data will enable current performance prediction codes
to be developed and validated.
A completely new turbine has been designed and built at Cranfield to
make aerodynamic measurements using pressure transducers. The design
has been dominated by the requirements of accommodating the transducer
signal processing equipment and allowing variation of many of the
rotor parameters. Three commercial glass fibre blades were installed
and performance curves measured on a conventional field site at a
height of 11.5m for three rotor speed settings. These measurements
show the turbine to give adequate power performance.
A mobile trailer has been used to tow the turbine at a height of 4m
along the Cranfield runways. Mobile testing facilitates an accelerated
test schedule and allows aerodynamic data to be acquired under
controlled wind conditions. A fully instrumented blade, fitted with
forty transducers, has been tested under these circumstances and
produced a large database of pressure measurements covering operation
in winds up to 25 iq/s and yaw angles between -4511 and +55°.
Analysis of the data has shown it to be of good quality and allowed
some of the effects of yaw and stall to be identified. The use of the
data base for performance prediction code validation has also been
established.
