Sam Austin

Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Civil Engineering

First Advisor

Sukhvarsh Jerath


The goal of this research is to add additional data to an expanding body of knowledge in the field of seismic engineering of wind turbines through a series of numerical analyses and also comparing them to current guidelines. This data can be used to further study the dynamic behavior of wind turbines under seismic loads and also provides answers to some of the questions in this field. This is important considering the expansion of wind farms into seismically active regions, increasing size of wind turbines, and lack of specific guidelines for seismic design of wind turbines. This research also emphasizes the need for a comprehensive research on the seismic behavior of larger turbines. This study will outline the shortcomings of current design guidelines and will result in safer and more economic designs in the wind industry.

In the following chapters, first a review of existing codes and articles on the topic is presented. Then theoretical formulations associated with time and frequency domain methods are presented. These methods will be used in the numerical procedure in the following chapters. Next using experimental shake table results on an industrial scale wind turbine, the finite element model is first validated. Then a series of modal and transient finite element analyses are performed on three horizontal axis tubular steel wind turbines towers on four types of foundations. Effect of soil is added using both implicit and explicit techniques. Seismic response of wind turbines with different sizes is then analyzed and effect of different design parameters including damping ratio, load direction, natural frequencies, size, foundation type, and soil model is investigated. Finally, accuracy of the response spectrum method (first mode approximation) suggested by current design codes is evaluated.

The results of this study show that the acceleration response in the vertical direction is sensitive to damping values and in design of connections for vertical forces, lower damping ratios should be considered. It was also observed that for all turbine sizes, displacement and stress values in the horizontal directions were significantly higher compared to vertical direction which means tower design is governed by horizontal seismic forces. When it comes to critical seismic direction, it was seen that increasing the size of turbine, tends to change this direction therefor, turbines should be designed for both seismic directions. Another observation was that larger turbines (5 MW) with lower damping values can have vertical accelerations higher than horizontal and it shows the need for three dimensional design of connection components. When it comes to seismic resonance, it was shown that natural frequency of the wind turbines can be close enough to earthquake frequencies to cause amplification. Therefore, it’s recommended to separate structural natural frequencies of wind turbines from both operational and seismic frequencies. Regarding the seismic design factor of safety, it was shown that factor of safety of wind turbines designed with current guidelines decrease as the wind turbine size increase. This is an important finding and reiterates the need for updated design codes considering the current trend in increasing size of turbines. Analysis results showed that including soil and foundation can increase the acceleration and displacement up to 13%. Therefore, it’s recommended to include the soil-structure interaction. When it comes to foundation, it was shown that different foundation types shift the structural frequencies unequally. This shows the importance of foundation type selection to avoid resonance. This study concludes that for wind turbines that have a frequency-based foundation design, the soil can be modeled faster and easier using the K-model. The results of this study also showed that moment demand value for the larger 5 MW turbine calculated based on IEC 61400 design guideline was smaller than what was seen in the transient FEA. This is a significant finding considering that size of modern wind turbines is increasing and they are being installed more frequently in seismic regions. This suggests the need for reevaluation of the current design standards for wind turbines.