Page 189 - Fluid Power Engineering

162 Chapter Eight

4. Lightning strikes. Direct lightning strikes can take a turbine
offline until it is checked for damage.
5. Change in roughness because of growth. Over time, growth
of forests and tall trees will increase the roughness in the
area leading to lower energy production.
Curtailment of energy production. Energy production may be

curtailed because of either grid constraints or high de-
gree of turbulence from adjacent turbines. The subcategories
are:
1. Grid operator may order curtailment of wind energy pro-
duction for a variety of reasons, including, low demand,
high supply from other sources, grid failure, etc.
2. Wind sector management. Layout of turbines may be such
thatonrareoccasionsthedirectionofwindcausesturbines
in the wake to experience a high degree of turbulence.
Under such conditions, the turbines subject to high tur-
bulence are shut down. As an example, consider a wind
farm in which the shortest distance between turbines is
three times the rotor diameter along the direction that is
perpendicular to the primary wind direction. In this exam-
ple, wind sector management would be deployed when
thedirectionofwindisperpendiculartotheprimarydirec-
tion of wind and the wind speed is large enough to cause
high degree of turbulence. Wind sector management im-
proves the longevity of the turbine by eliminating large
fatigue loads.
Others. A variety of locale-specific losses may exist including:

1. Earthquakes. An earthquake will cause a wind farm to shut
down for safety reasons. It may not resume until an in-
spection is completed to check the integrity of foundation
and other structural elements. Although there is uncer-
tainty in the timing of earthquakes, over a lifetime of the
wind farm, the average loss may be predicted.

Estimation of the losses may require extensive research into locale-
specific performance of equipment, weather conditions, grid condi-
tions, and others. Objective estimates of losses may be substituted
for subjective estimates of local experts, but a temptation to broadly
use industry averages should be avoided when preparing a bankable
resource estimate. For preliminary wind resource assessment, a loss
estimate of 10% is commonly used as a placeholder. Table 8-5 lists
generic estimates for losses. These losses are applied to the annual en-
ergy production estimate that are computed from wind speed estimate
and power performance curve.