Page 114 - Fluid Power Engineering

92 Chapter Six

(example Optimize in WindPRO) with wind resource map
from step (1) and a variety of constraints. Constraints include
maximum wind-farm capacity, minimum capacity factor,
turbine tower height and rotor diameter, distances between
turbines, and other setback criteria like distance between tur-
bine and property boundary, public roads, transmission, and
inhabited areas. In this phase, the constraints are guidelines
rather than precise. For a single turbine case, a location with
the highest wind resource that satisfies all the constraints is
the location of the met-tower. For a wind farm, go to step (3).
3. In the preliminary WTG layout, form clusters of WTGs. If the
wind farm is in a complex terrain, then 5 to 7 WTGs may be
grouped into one cluster. If it is a simpler terrain with very
little changes in elevation and roughness, then 10 to 12 WTGs
may be grouped into one cluster. The clusters will be based
on distance. Clusters are best formed visually; the borders of
clusters may be drawn manually on a Computer Aided De-
sign (CAD) drawing or on paper. The ratio between WTG
and met-towers of 5 to 7, or 10 to 12 are normal guidelines for
determining number of fixed met-towers for wind measure-
ment. Rarely will all clusters have the same number of WTG.
4. In each cluster, find the median wind speed WTG. This WTG
location or a location in the vicinity would be a location for
placement of met-tower. Measuring at the best or worst wind
resource location in the cluster would yield wind measure-
ments that have to be either extrapolated down or extrap-
olated up to all points in the cluster, which would lead to
higher inaccuracies. Normally, a set of two or three locations
is chosen in each cluster. For instance, the three locations are:
Location with median wind speed and the two locations with
the smallest difference with the median wind speed.
5. The goal of this step is to pick one location in each cluster such
that, for the wind farm as a whole, the met-tower locations are
sufficiently spread out geographically. This step is best done
visually, starting with the median wind speed location in each
cluster and then examining the proximity of these locations. If
the median locations of two clusters are geographically close,
then alternate locations are chosen from the set.

Thedurationofmeasurementisaminimumof1yearwithapreference
of 2 to 3 years. One year is sufficient to capture the seasonal variations
inwindspeed,direction,turbulence,andshear.However,year-to-year
variations are significant. In order to predict the energy output of a
wind project accurately over its lifetime of 20 or more years, statistical
methods like MCP are used. This method is described in detail in
Chapter 7. One of the key steps in this process is the computation of