Wind Turbine Aerodynamics Part A: Basic Principles - Wind Turbine Technology: Fundamental Concepts of Wind Turbine Engineering - page 323

HAWT Aerodynamic Performance Tests
Testing Methods
One of the aerodynamicist's few windows for observing and understanding the physical
phenomena of wind turbine aerodynamic behavior comes from performance data obtained
during field testing of HAWTs. There are three basic methods of testing wind turbine rotors,
each method having its own advantages and disadvantages.
Wind Tunnel Testing
Wind tunnel testing, the mainstay of the aircraft industry, has been of limited value in
wind turbine rotor development, although it is useful for obtaining basic two-dimensional
airfoil lift and drag data. While tests in a facility such as the NASA Ames Research Center's
80-ft by 120-ft Wind Tunnel have provided useful data [Simms et al . 2001 and Schreck 2002],
financial limits have restricted the extent of wind tunnel tests. Scaling and blockage problems
occur in most wind turbine tests in wind tunnels. For the test and full-scale Reynolds Num-
bers to be approximately equal, a large test model is wanted. However, because the rotor
wake expands, the blockage ratio (swept area/cross-sectional area of the wind tunnel) should
be small, preferably less than 1/10.
Tow Testing
Tow testing, whereby the wind turbine is pulled or pushed through static air, can relieve
the scaling and blockage problems of wind tunnels, but the rotor size is limited. The tow test-
ing of a 15-m diameter wind turbine, for example, would be a formidable task. Tow testing
also shares with wind tunnel testing the problem of failing to subject the wind turbine to the
unsteady nature of the wind.
Field Testing
Field testing presents the proper wind environment, but it brings new challenges in mea-
suring and recording test data. Field test data acquisition and recording vary with the type of
test to be made. For load tests , it is desirable to sample the wind at several circumferential
stations upwind of the rotor ( e.g. with an array of anemometers as shown in Fig. 3-25), with
time resolution sufficient to distinguish small-scale turbulence . Load measurement is made
with strain gages , using a sampling rate sufficiently fast to determine the frequency response
of the rotor blades. Performance tests , in which the power output of the wind turbine is de-
termined, use time-averaged measurements. Discussion of testing in this chapter focuses on
performance data, since loads data are covered in later chapters.
Test Data Acquisition
Performance data from wind turbines are stored and plotted using the well-proven method
of bins [Akins 1978]. Output power and free-stream wind speed are sampled over periods of
time, on the order of 5 minutes, and average values for each period are stored in wind-speed
“bins.” The bins are on the order of 1 m/s in range. The power and wind speed samples in
each bin are averaged for plotting. This method greatly smooths the resulting graph of power
vs. wind speed or the power curve .