Wind Turbine Acoustics - Wind Turbine Technology: Fundamental Concepts of Wind Turbine Engineering - page 448

In Figure 7-34, each data point represents observations by one or two people and de-
fines the distance at which the wind turbine noise is heard intermittently. The two aural
curves in the figure are then estimated from these observations and from a limited number
of sound pressure measurements. Both curves are foreshortened in the upwind direction and
elongated in the downwind direction. With one exception, broadband noise was the domi-
nant component perceived for both HAWTs. The exceptional case is that of noise
downwind of the downwind-rotor machine, for which low-frequency impulses are the
dominant component. This accounts for its longer downwind detection distance as
compared with that of the upwind-rotor turbine.
Background Noise
Because background noise is an important factor in determining people's responses to
wind turbine noise, it must be carefully accounted for by site measurements without the
wind turbines operating, and preferably prior to their construction. Sources of background
noise are the wind itself; its interaction with structures, trees, and vegetation; human
activities; and, to a lesser extent, birds and animals. Natural wind noises are particularly
important because they can mask wind turbine noise, as a result of the fact that their
broadband spectra are similar to those of wind turbines. Measuring background noise, at
the same locations and with the same techniques used for measuring wind turbine noise, is
an integral part of assessing receiver response.
Noise Exposure Inside Buildings
People who are exposed to wind turbine noise inside a building experience a much dif-
ferent acoustic environment than do those outside. The noise transmitted into the building
is affected by the mass and stiffness characteristics of the structure, the dynamic response
of structural elements, and the dimensions and layouts of rooms. People may actually be
more disturbed by the noise inside their homes than they would be outside [Kelley et al.
1985]. Indoor background noise is also a significant factor.
Data showing the reductions in outdoor noise provided by typical houses are given
in Figure 7-35 as a function of frequency. The hatched area shows experimental results
obtained from a number of sources [Stephens et al. 1983]. The noise reduction values of
the ordinate are the differences between indoor and outdoor levels. The most obvious
conclusion here is that noise reductions are larger at higher frequencies. This implies that
a spectrum measured inside a house will have relatively less high-frequency content than
that measured outside. These data are derived from octave-band measurements but are
generally not sensitive to frequency bandwidth.
Very few data are available on outdoor-to-indoor noise reduction at the lowest frequen-
cies ( i.e. , below 50 Hz). In this range, wavelengths are comparable to the dimensions of
rooms, and there is no longer a diffuse sound field on the inside of the building. Other
complicating factors are low-frequency building resonances and air leaks. The inside distri-
bution of sound pressure can be nonuniform because of structure-borne sound, standing
wave patterns, and cavity resonances in rooms, closets, and hallways.
Data relating to the noise-induced vibration responses of houses are summarized in
Figure 7-36, in which RMS acceleration levels are plotted as a function of external sound
pressure level. The trend lines for windows, walls, and floors are averaged from a large
number of test measurements of aircraft and helicopter noises, sonic booms, and wind
turbine noise.