Wind Turbine Acoustics - Wind Turbine Technology: Fundamental Concepts of Wind Turbine Engineering - page 442

pressure level. For nonzero atmospheric absorption, the effect of additional rows is less sig-
nificant at all receiver distances. Doubling the number of rows results in an increase in the
sound pressure level of less than 3 dB.
Figure 7-28 shows similar data for two different row lengths. For these comparisons,
the turbine spacing is constant and the row lengths are doubled by doubling the number of
machines per row. When the receiver is at shorter distances, the predicted sound pressure
levels are equal because of the equal turbine spacing. At longer distances, the levels for the
double-length row are higher by 3 dB because the acoustic power per row is doubled.
Figure 7-28. Calculated noise propagation downwind of wind turbines in rows of two
different lengths. ( a = 0.54 dB/100 m) [Shepherd and Hubbard 1986]
Computations were also made [Shepherd and Hubbard 1986] for a configuration similar
to that of Figure 7-25, except that the row spacing was reduced from 200 m to 100 m. At
all distances to the receiver, the computed sound pressure levels were higher for this more
compact array.
Effect of Turbine Rated Power
Shepherd and Hubbard [1986] calculated the effect of the turbine's rated power on
noise emissions by increasing the power of each turbine and the total station power. The
turbine and row spacings were adjusted from those of Figure 7-25 to more appropriate
values for larger machines (four times the rated power). Sound pressure levels from rows
of 16 400-kW wind turbines were compared with levels from the same number of rows
of 31 100-kW machines. This approximately doubled the rated power of the station. The
reference spectrum for the larger turbines was assumed to have the same shape as that of
the smaller turbines (Figure 7-10), although the levels were all 6 dB higher. This implies
four times the acoustic power for four times the rated power. The computed sound pressure
levels are 3 dB higher for the array of larger turbines because the acoustic power is doubled
for each row of the array. Different results would be obtained if the reference spectra of
the two sizes of turbines had different shapes.