Upp í vindinn - 01.05.2017, Page 53

The overturning moment demand obtained from time history analysis
is shown in Figure 3 b) as a function of normalized period (pulse period
normalized by the fundamental period of the structure). The results indicate
a strong dependence of overturning moment on the normalized period. The
maximum overturning moment is around 325 MNm which is close to what is
reported in other studies [5]. The values exceed moment demands of extreme
wind loads previously reported as 98 MNm based on extensive simulations [3], it
is even higher than the wind load reported by NREL [4] when using extreme load
factors of 1.35 where a maximum overturning moment demand of 153 MNm was
reported.
These results indieate that the earthquake loads may be design-driving loads
for large wind turbines and particularly in the near-fault region. The results also
indicate that the most critical ground motions for the wind turbines of this type
are near-fault earthquakes with moment magnitude in the range 6.5 to 6.9. it
seems that seismic loads due to near-fault ground motions are the largest when
the pulse period is between about 0.5 to 1.5 times the fundamental period of
the structure. It can be observed from Figure 3 a) that the pulse period is, on
the average, close to the fundamental period of the structure in the second
magnitude bin, which also explains the higher demand due to earthquakes in
this bin. The average overturning moment demands in each bin are shown in
Figure 3 b). If response spectralshapes from EC8 scaled with PGA of eachground
motion is used, the results vary significantly frorn those obtained by time history
analysis. For earthquakes that have a moment magnitude larger than 6.5, the EC8
spectra shows serious underestimation of base moment demand. In conclusion,
the results indicate that the EC8 model is not suitable to evaluate seismic action
on tall wind turbine towers. This is due to the inability of the model to account
for long-period energy content in near-fault ground motions. The RR2011 model
was found, on the average, to represent the results obtained from time history
analysis very well. It is observed that response spectral analysis using tlie SRSS
combination rule gives satisfactory results as long as a proper response spectrum
is used. More details on the models and the results can be found in Sigurðsson
2015) [8] .
References
[1] Bir. G.. S Jonkman. J. (2008). Modal dynamics of large wind
turbines with different support structures. In ASME 2008 27th
International Conference on Offshore Mechanics and Arctic
Engineering (pp. 669-679). American Society of Mechanical
Engineers.
[2] Butterfield. S.. Musial. W.. fi Scott. G. (2009). Definition of a
5-MW reference wind turbine for offshore system development.
[3] Fogle. J.. Agarwal. P.. fi Manuel, L. (2008). Towards an
improved understanding of statistical extrapolation for wind
turbine extreme loads. Wind Energy, 11(6). 613.
[4] Jonkman. J. M. (2007). Dynamics modeling and loads
analysis of an offshore floating wind turbine. ProOuest.
[5] Prowell. I. (2011). An experimental and numerical study of wind
turbine seismic behavior.
[6] Ris0 (2002). Guidelines for Design of Wind Turbines. Det
Norske Veritas fi Wind Energy Department of Ris0 National
Laboratory.
[7] Rupakhety, R., fi Sigbjörnsson. R. (2011). Can simple pulses
adequately represent near-fault ground motions? Joumal of
Earthquake Engineering, 15(8), 1260-1272.
[8] Sigurðsson. G. Ö. (2015). Seismic response of wind turbine
structures in the near-fault region. MSc. Thesis. University of
lceland. Iceland.
[9] Standard. B. (2005). Eurocode 8= Design of structures for
earthquake resistance.
10] Wind. G. L. (2005). Guideline for the Certification of
Offshore Wind Turbines. Germanischer Lloyd Industrial Services
GmbH.