Page 109 - Fluid Power Engineering

W i nd Measurement 87

8%

7% 2.0%
Wind speed error at hub height 5% 1.5%
6%

1.0%
4%

3%

2%

1%

0%
10 20 30 40 50
Lower anemometer height [m]
FIGURE 6-10 Error in wind speed at 80 m that is computed using wind shear
based on different quality anemometers at an upper height of 50 m and
different lower heights. (From Taylor, M., Mackiewicz, P., Brower, M. C.,
Markus, M., An Analysis of Wind Resource Uncertainty in Energy Production
Estimates, EWEC, London, 2004.)

ment heights, and terrain. Figure 6-9 indicates the impact of
6
measurement uncertainty of the anemometer, and Fig. 6-10
indicates the impact of measurement heights. In this case, the
upper anemometer is at 60 m and the hub height is 80 m. The
conclusion is that a larger difference in height yields higher
accuracy of wind speed extrapolation. As indicated by the
points annotated in Fig. 6-10, error decreases from about 4.7
to about 2.9% as the height of lower anemometer is changed
from 50 to 40 m. The error is further decreased to 2.4% when
lower anemometer is at a height of 30 m. The reason for higher
accuracy is explained by considering the power law for shear
in the following format:

(6-3)
γ = log v 1 − log v 2 / log h 1 − log h 2
Since γ is estimated from 10-min wind speeds, it is instructive to ob-
serve that γ is the slope in the log v, log h space. A better estimate of
the slope is achieved when the two measurement points are separated.
However, there is a limit to the separation in terms of height of the
lowest anemometer. At some point, the uncertainty in readings of the