Wind turbine acoustics - Wind Power Generation and Wind Turbine Design - page 167

Figure 14: Sketch of the board position (left) and the arrangement of the micro-
phones on the board (right) (after [16]).

microphone. The spatial accuracy of the proximity microphones depends on the
distance from the source, but, unlike acoustic parabolas, is independent of the
acoustic wavelength.

5.1.4 Acoustic antenna
Acoustic antennas are composed by a set of microphones in a row or in a matrix
which can be mounted on a ground plate or in the proximity of the wind turbine.
By computing correlations of the signals provided by different microphones, the
location of the acoustic sources can be determined.
Oerlemans et al. [16] used for example an array of 148 microphones mounted
on a board of 15 m 18 m. The array had an elliptic shape to account for the cir-

cular motion of the blades and for the tilted angle of the board (see Fig. 14). Cross-
correlating the microphones data made possible to determine the distribution of
the noise sources in the rotor plane and along the rotor blades (individually for the
three blades). It was observed that most of the noise sources are located on the
right-hand side of the rotor plane, indicating that most of the noise is generated
during the downward motion of the blades. Furthermore, the contribution of the
hub was also clearly visible.

×

5.2 Wind-tunnel measurements

Wind tunnels have the advantage of providing better controllable fl ow conditions
and thus are preferred for systematic studies. Due to their size limitations, how-
ever, wind tunnels are limited to downscaled model studies. The large-scale facil-
ity at the National Renewable Energy Laboratory in the USA is equipped with
a wind tunnel with a measurement cross section of 24.4

36.6 m 2 which made
possible to carry out fl ow measurements on a full-scale (10 m rotor diameter) wind
turbine [17]. The wind tunnel, however, is not anechoic, thus is not well suited for
acoustic measurements.

×