Environmental Engineering Reference
In-Depth Information
load change occurs during wind speeds below rated wind speed [
18
]. Therefore, it
can be concluded that half of the fatigue loading is caused by periodic tower
vibrations.
12.2.2 Seismic Loading
Annual installations of wind power have increased constantly across the world
over the last decades, expanding the market toward seismically active areas. If a
project is located at sites with relevant seismic hazard, the wind unit must be
designed considering a reasonable likelihood of earthquake occurrence during the
operational state or an emergency shutdown. In some cases, seismic plus opera-
tional loads may govern tower and foundation design. Figure
12.2
shows the
representation of the seismic hazard in Europe in combination with the average
wind speed at 50 m from the ground. The latter is a key issue for the site suitability
assessment for wind power and must be greater than 5-6 m/s. The map reports also
annotations of the wind power installations in European countries by the end of
2012 according to GWEC [
16
]. A large part of the south European coastal areas
presents high seismic hazard and such wind conditions, which are sufficiently
suitable for financial returns from modern wind turbines.
Norms and guidelines come to the aid of practitioners, providing general
suggestions for the seismic design of wind turbines. However, an accurate
approach for the superposition of earthquake and wind effects is still missing.
This topic has aroused the interest of researchers all over the world. Seismic
design loads are less crucial than standard design loads under extreme wind
conditions [
7
,
25
,
26
]. Therefore, the load combinations prescribed by the Inter-
national Electrotechnical Commission (IEC) [
19
] provide also a seismic safe
design.
Also, Ritschel compared the loads occurring during earthquakes with IEC
design loads for the case of a 60 m hub height wind turbine [
28
]. Modal and
transient analyses were also compared. The earthquake response was covered by
the design load at most of the tower sections. However, a peak acceleration of
0.3 g may be considered the maximum seismic excitation, a 60 m hub height wind
turbine can withstand. For the blades, earthquake loads are much lower (about
70 %) than the design loads, governed typically by the 50-years gust load case,
therefore are not decisive. They found that the modal approach yields relatively
conservative results near the tower base with respect to the results of transient
analysis. This work confirmed the general rule that a frequency domain analysis
produces more conservative results than a time domain analysis, because of the
smaller number of considered participating modes. However, in the frequency
domain, phase information gets lost and possible in-phase summation of wind and
seismic signals cannot be detected. Transient analyses are generally preferred
because of the higher accuracy and more realistic representation of the problem.
