- About this Journal ·
- Abstracting and Indexing ·
- Advance Access ·
- Aims and Scope ·
- Annual Issues ·
- Article Processing Charges ·
- Articles in Press ·
- Author Guidelines ·
- Bibliographic Information ·
- Citations to this Journal ·
- Contact Information ·
- Editorial Board ·
- Editorial Workflow ·
- Free eTOC Alerts ·
- Publication Ethics ·
- Reviewers Acknowledgment ·
- Submit a Manuscript ·
- Subscription Information ·
- Table of Contents

Advances in Mechanical Engineering

Volume 2013 (2013), Article ID 175684, 9 pages

http://dx.doi.org/10.1155/2013/175684

## Comparison of Structural Properties between Monopile and Tripod Offshore Wind-Turbine Support Structures

^{1}Key Laboratory of Coastal Disaster and Defence, Ministry of Education, Hohai University, Nanjing 210098, China^{2}College of Harbor, Coastal, and Offshore Engineering, Hohai University, Nanjing 210098, China

Received 28 June 2013; Revised 22 September 2013; Accepted 23 September 2013

Academic Editor: Luigi Cappelli

Copyright © 2013 Da Chen et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

#### Abstract

Offshore wind power provides a new kind of green energy. This paper presents a comparison study on the structural properties of monopile and tripod wind-turbine support structures, which are used extensively in offshore wind farms. Both structures have the same upper tower, but different lower structures, one with a monopile and the other with a tripod. Static, fatigue, and modal analyses indicate that both the tripod and monopile structures are feasible in the field, but that the tripod structure is superior to the monopile structure. Static analysis reveals that the location of maximum stress in the monopile structure is different from that in the tripod structure, and that the tripod structure shows higher stiffness and greater stress-control capacity than the monopile structure. Fatigue analysis indicates that the tripod structure has a longer lifetime than the monopile structure. Modal analysis indicates that the two structures exhibit large differences in their natural frequencies. Unlike the monopile structure, the third and first modes both have a substantial influence on the dynamic response of the tripod structure.

#### 1. Introduction

In recent years, clean energy strategies have been given great importance in environment protection and durable development. In the case of electricity, offshore wind farms promise to become an important source of energy in the near future to decrease reliance on traditional coal-fired power. In the past three decades, nations around the world with wind energy have led the way in the development of offshore wind farms. Thirty to forty percent of all new installed power generation capacity in Europe and the United States is now associated with wind energy. It is expected that wind energy farms with a total of 80 GW will be installed in Europe and the United States by 2020 [1].

It is worth noting that the support structures of a wind-turbine system act as the main structural members of a wind electricity farm and are closely related to the structure’s safety, stability, and durability. In general, the upper tower is composed of a conical steel pipe, whereas various types of substructures can be designed according to field conditions. The substructure is used to anchor the support structure to the seabed and typically belongs to one of six types: gravity, monopile, tripod, jacket, suction, and floating foundation [2, 3]. Based on current design philosophy, gravity-based structures are preferred for shallow waters (up to 5 m), whereas the monopile foundation is used for wind farms in water depths up to 20 m. For deeper water, tripod or jacket support structures are often considered. Floating support structures remain a challenge due to their high cost, but this challenge will need to be met for countries with fewer regions of shallow water [4]. However, Scharff and Siems [5] have explored the application of monopile foundations in water depths of up to 20–40 m and have provided two detailed discussions of design examples.

The support structures (tower and foundation) are subjected to a variety of combined static and dynamic loads such as gravity, wind, waves, tides, and earthquakes. The design and analysis of these support structures are key parts of the design of the whole wind-turbine system. A fatigue property analysis of a monopile structure by Mo et al. [6] indicated that turbine-system vibration was a main source of load, resulting in structural fatigue damage. Torcinaro et al. [7] presented a structural optimization design for a tripod support structure through stress and stability analysis. Agbayani [8] reviewed damage to monopile structures subjected to high-cycle fatigue and proposed a series of repair measures. Bazeos et al. [9] investigated the static, seismic, and stability capabilities of the monopile support structure of a wind turbine and found that refined finite-element models were necessary at specific critical locations for a more accurate structural analysis.

However, few studies have yet been performed on the differences in structural properties among the different types of wind-turbine support structures. Moreover, a new planned wind-power installation in Donghai, China, is in the discussion stage, waiting for a decision on the type of foundation. In China, grouped pile foundations are used in most wind-turbine support structures. However, compared with monopile and tripod foundations, the grouped pile foundation involves more complex construction and higher cost. Given these points, to provide a reference for foundation-type selection, this paper presents a study on a support structure where the lower support structure is designed as a monopile or tripod structure, but the upper one remains the traditional tower structure. Results of static, fatigue, and modal analysis using the SOLIDWORKS2012 general FEM software were compared between monopile and tripod structures.

#### 2. Description of Support Structures

In the present study, two types of offshore wind-turbine support structures were investigated. In these structures, the upper towers were the same, but the lower structures were different, one being a monopile and the other a tripod. Moreover, the support structure was made of high-strength steel S620M with a yield strength of 620 MPa.

##### 2.1. Upper Tower Structure

Figure 1 shows a sketch of the upper tower. This upper tower consisted of four pieces of steel shell with diameters linearly varying along their height. The upper tower was 60 m in height, and each piece of shell had the same length, 15 m. The bottom diameter of the tower was designed as 5400 mm, which is identical to that of the upper tower in a similar wind-turbine system built in Donghai. The cross-sectional thickness of the tubular shell was designed based on design standard NORSOK N-004, where the ratio of diameter to the thickness of the tubular shell is required to be [10]. Note that the thickness of steel tubular was determined according to the bottom diameter of each piece and the ratio of 110. These four pieces of tubular shell were labeled as A to D in sequence according to decreasing diameter. Details of all the pieces of tubular shell are summarized in Table 1.

Two steel transition pieces, as shown in Figure 1, were used as the end configuration of each steel shell and were welded onto the steel shell. The transition pieces of adjacent steel shells were connected together by rivets to make up the upper tower.

##### 2.2. Lower Support Structures

Figure 2 shows details of the monopile and tripod substructures. Both types of substructures were 40 m long. The monopile structure had a constant diameter of 3600 mm, or in other words, the diameter decreased from 5400 mm in the upper tower to 3600 mm in the lower monopile. The thickness of the steel-pipe pile was 45 mm. The tripod structure consisted of a central column, three diagonal bracings, and three supporting pile sleeves. For the central column, the diameter of the top was 5.4 m and then decreased to 4.08 m. Each pile sleeve was 13 m high, was located at 12.5 m from the central column, and had a diameter of 2.4 m. The thickness of the pipes in the tripod substructure was constant at 40 mm. A diagonal transition member was used to connect the upper tower and the lower support structure together. All the joints between the steel shells were welded together.

Note that for the lower monopile support structure, its diameter is supposed to increase with water depth because of the increased bending moment and compressive force. However, it is also true that the wave loading on a pile, as one of the main loads considered in structural design, becomes larger with increasing pile diameter. Moreover, the pile-sinking construction of the monopile foundation is more difficult with the increasing diameter from the viewpoint of the local stability. Given these two points, the authors used a monopile foundation as the lower support structure that was made of high-strength steel with a yield strength of 620 MPa, and thus a diameter (3600 mm), smaller than the 5400 mm bottom diameter of the upper tower, was utilized.

##### 2.3. Applied Load

Offshore wind-turbine structures are not only exposed to highly corrosive environmental conditions, but are also subjected to various quasi-static, periodic, stochastic, and transient loads. For convenience in structural design, it is necessary to perform a reasonable load simplification. In this research, the main loads experienced, including horizontal wind load, wave load, and vertical gravity, were simplified as described below.

###### 2.3.1. Wind Load

Wind conditions are important not only in defining the loads imposed on turbine structural components, but also in designing the support structures of the wind-turbine system. The measured on-site wind parameters strongly influence the design of wind-turbine support structures. The wind load can be obtained through a formula based on lift theory:
where is the force of the wind on the blades (kN), is the density of air (1.225 kg/m^{3}), is the wind speed (m/s), is the surface area of the blades (m^{2}), and is the portance coefficient, assumed to be 0.8 for a classic blade.

According to observed data and information, the surface area of the blades was 50 m 1.5 m 3 blades 225 m^{2}, and the maximum speed before installation of the wind farm was 34 m/s. Therefore, the wind load calculated by (1) was 127 kN. Furthermore, assuming a safety coefficient of 1.35 (as recommended in DNV-OS-J101 [11]), the horizontal force applied on the structure by the wind would be 172 kN. In addition, according to the structural offshore wind-turbine optimization method, the wind exerted a pressure of 5 kN/m on one side of the upper tower.

###### 2.3.2. Wave Load

In the ocean environment, wave force is also a major load imposed on the structure. The maximum horizontal wave force was calculated based on Airy’s linear theory. In this theory, the horizontal and vertical water-particle velocity at coordinates and time can be expressed as [12] where and are the horizontal and vertical velocity of water, denotes wave height, and and represent the wave length and wave angular frequency. Based on Airy’s linear theory, the correlation between and is given by the dispersion equation

Furthermore, the water-particle accelerations and can be obtained based on and using the corresponding velocity (2): where is the distance from the vertical position of the wave surface to the seabed and can be calculated as where is the water depth, the distance from the seabed to the still-water level, and is the distance from the vertical position of the wave surface to the still-water level and can be obtained by [13]

For slender offshore structures such as monopiles, tripods, or offshore template structures, the Morison equation can be used to convert the velocity and acceleration terms into wave forces [14]. The Morison equation can be written as where denotes water density, and denote the drag and inertia coefficients, and is the diameter of the structural member. The first term on the right-hand side of this equation is referred to as the drag term and is proportional to the square of the water velocity. The second term is referred to as the inertial term and is proportional to water acceleration. For the case under study, the parameters were taken as 10.66 m, , and . The predicted wave force was 357.46 kN. Assuming a safety factor of 1.04 (according to the DNV-OS-J101 standards), the horizontal wave force for the current case was 371.76 kN.

###### 2.3.3. Blade and Rotor Loads

According to a real offshore wind farm such as M5000, the mass of the top structure (blades and rotor) is approximately 49.5 tons (or 493 kN). This load can be equivalently applied on the top of the tower in the vertical direction for calculating the wind-turbine support structure. Detailed data on the M5000 wind farm can be obtained on the AREAVA website [15].

###### 2.3.4. Gravity Load of the Structure

The acceleration of gravity was assumed to be 9.81 m/s^{2}. The total load condition of the monopile structure is shown in Figure 3 and is identical to that of the tripod structure.

#### 3. Numerical Analysis and Discussion

##### 3.1. Static Analysis

###### 3.1.1. Stress Comparison

Figure 4 shows the calculated stress nephograms for the monopile and tripod support structures. Comparing the stress nephograms in Figures 4(a) and 4(b), it is apparent that the tripod structure shows the maximum stress at the transition junction from the lower support structure to the upper tower, whereas the monopile structure has its maximum stress value at the end of the lower pile. In addition, as shown in Figure 4, the maximum stress of the monopile structure is approximately 2.35 times greater than that of the tripod structure, but it is still far less than the yield strength of 620 MPa, and therefore these two structures remained in the elastic stage. For the monopile structure, the maximum tensile and compressive stresses were, respectively, 486.6 MPa and 499 MPa, while the corresponding values for the tripod structure were 279.2 MPa and 211.8 MPa. Given these points, the tripod structure was more effective than the monopile structure from a stress-control point of view.

Figure 5 shows the tensile and compressive stresses distribution along the longitudinal height of the two support structures. It is clear that the stress distributions in the upper tower are similar for the tripod and monopile foundations, whereas a large difference is apparent along the lower support structures. The stress was only 27 MPa at the foot of the tripod structure, where the maximum stress was observed for the monopile structure. Moreover, as the diameter increased from 4.08 m to 5.4 m, the stress rapidly dropped from the maximum value to 70 MPa for the tripod foundation structure. In addition, the maximum stress for the tripod foundation was observed in the transition section, a value close to the stress of 274 MPa at the same location in the monopile foundation. Judging from these points, the tripod structural system can be said to be superior to the monopile structural system from the structural stress distribution viewpoint.

As shown in Figure 5, note that the maximum stress in the tripod support structure and the tower structure was lower than the yield strength of 355 MPa for S355 steel. Therefore, the high-strength steel S620M can be replaced with the normal steel S355 to make full use of tensile strength of steel.

###### 3.1.2. Deformation Comparison

Figure 6 shows the deformation nephogram for the monopile and tripod support structures. It is apparent that the deformation increased with the height above the seabed and that the maximum displacement occurred at the top of the structure in the lateral direction. The monopile structure showed a maximum deformation of 486 mm, whereas the tripod structure had a maximum displacement of 368 mm. In addition, the displacement of the tripod structure was less than that of the monopile structure at the same longitudinal location. In other words, the tripod structure had greater flexural stiffness than the monopile structure.

With respect to the maximal lateral deformation and the height of the structure , the deflection factor can be calculated based on DNV-OS-201J [11]:

In the cases studied, when = 100 m, the computed values of for these two structures were 206 and 271.7, respectively, which both satisfied the requirements of DNV-OS-201J [11].

In summary, the tripod structure is clearly more stable and resistant to applied loads than the monopile structure.

##### 3.2. Fatigue Analysis

Figure 7 shows the number of cyclic loads carried up to fatigue failure for the monopile and tripod structures. The most fatigue-affected locations still corresponded with the locations of the maximum tensile stress in the static analysis. However, the number of loading cycles carried by the tripod structure was far larger than that of the monopile structure. With respect to the most damaged structural position, the maximum number of load cycles was only 210,000 for the monopile structure, whereas the tripod structure was able to endure at least 410,000. In other words, the tripod support structure had a lifetime 48.5% greater than that of the monopile support structure. Judging from these points, the tripod structure is more resistant to fatigue than the monopile structure.

Both support structures satisfied the minimum cycle number of recommended in DNV-OS-J101 [11]. However, it is also necessary to determine the lifetime of these two structures. The parameters were defined according to the study by Agbayani [8]. First, for these two structures, the number of loading cycles was estimated to be 40 per day. It should be highlighted that the stress on a structure subjected to different loading cycles varies between the minimum and maximum stresses, and some cycles must result in a stress lower than the maximum value. Indeed, all loads are not equal in intensity and always occur in a random time distribution. However, the forces and pressures applied in the present study have been set to their maximum values. Therefore, to perform a realistic calculation, the number of loading cycles has been assumed to be 40 per day.

Second, the working time ratio (proportion of working hours in each 24 hours) had to be determined for each structure. According to investigations of the structures under study, the maximum working time ratio was set to 0.85 for the tripod structure and 0.8 for the monopile structure [16]. Note that these ratios are defined in a conservative way because of safety considerations and that the actual ratios are probably less than these values.

Knowing all these factors, the lifetime can be calculated for each structure. For the monopile structure, the number of load cycles is 11,680 per year considering the assumed working ratio. The lifetime durability can then be obtained as where is the lifetime (in years), is the minimum number of loading cycles, and is the loading cycles per year. For the case studied, the lifetime durability obtained by (9) was approximately 18 years. For the tripod support structure, the number of load cycles per year was 12,400 assuming a working ratio of 0.85. The lifetime durability calculated based on (9) was then approximately 33 years. From these results, it can be concluded that the tripod support structure can survive longer than the monopile structure, which is advantageous over the long term.

##### 3.3. Modal Analysis

When a structure is subjected to dynamic loads, its natural structural frequencies should be adjusted to be far from the dynamic load frequency to avoid the resonance. If the natural structural frequency is close to the load frequency, even small driving forces can produce large-amplitude oscillations.

Modal analysis is the basis of dynamic analysis. A wind-turbine support structure is subjected to many periodic actions such as winds and waves. In general, these loads have progressive and changeable frequencies within a known and specific domain. Therefore, the risks of resonance can be estimated and eliminated through calculating and adjusting the natural frequencies of the structures.

Table 2 lists the natural frequency and period of the first three vibration modes for the monopile and tripod structures. The tripod structure was found to have higher natural frequencies and shorter periods than the monopile structure. To estimate whether resonance will occur, the dynamic load frequency must be established. In this research, only the main load (wind load) was taken into account to simplify the analysis. In fact, the vibration of structures subjected to wind load arises from rotary turbulences due to the impact of wind on the structure. The frequency of these rotations can be determined using the Strouhal number, a dimensionless number describing oscillating flow mechanisms.

Based on the theory for calculation of the Strouhal number, the turbulence frequencies can be obtained as where is the frequency of the vortex, is the wind velocity, and is the characteristic length of the structure. In the case under study, the Strouhal number was in the range from 0.1 to 0.2. The minimum wind velocity for the structure was 11 m/s and the maximum 34 m/s. The characteristic length was taken as the length of the upper tower structure (60 meters), and then the calculated vortex frequency was in the range from 0.0183 Hz to 0.1130 Hz. As listed in Table 2, the lowest natural frequencies of the two structures were 0.390 Hz and 0.797 Hz. Therefore, resonance of these structures due to rotary turbulence can be eliminated.

Figures 8 and 9 show the first three mode shapes for the monopile and tripod structures. These mode shapes were similar for both structures, whereas the variance law of displacement coordinates showed a large difference. The maximum displacement coordinates are listed in Table 2. It is apparent that the maximum displacement gradually increased with increasing frequency for the monopile structure, whereas the tripod structure showed an increase first and then a drop with frequency.

Based on the mode shape, the mass participation factor can be calculated as where is the mass of the th mass particle and is the displacement coordinate of this point for the th mode. Figure 10 shows the mass participation factor of different modes for these two types of structures. For the monopile structure, the first mode shape, with a frequency of 0.390 Hz, has a maximum mass participation factor of 46%, approximately 3.5 times that of the second mode. This implies that the first mode plays a controlling role in the dynamic response of monopile structures. However, for the tripod structure, the third mode, with a frequency of 7.108 Hz, showed the maximum mass participation factor of 30.2%, slightly larger than the value of 24.5% for the first mode, but far greater than 9.6% for the second mode. It can be concluded that both the third and first modes play key roles in the dynamic response of tripod structures, which is clearly different from the response of monopile structures.

Moreover, the frequency separation can be further defined by the difference between any two adjacent frequencies. For the monopile structure, the first two modes presented the smallest frequency separation, 2.111 Hz, and the sum of the corresponding mass participation factors was 59%. In the tripod case, the smallest frequency separation was 2.920 Hz, between the second and third modes, and the total mass participation factor of these two modes was 39.8%. It can be concluded that the tripod is the safer structure because of its wider frequency separation and smaller mass participation factor.

The static, fatigue, and modal analyses of monopile and tripod support structures described above have demonstrated that both the monopile and tripod support structures are applicable under the field conditions considered. Note that the designed pile diameter and the use of high-strength steel are feasible for the support structures described in this paper. In contrast, a previous investigation by Scharff and Siems [5] indicated that for a wind-turbine system similar to those studied here, the pile diameter was increased to 7500 mm from 5500 mm of the upper tower for a water depth of 20–40 m. Comparing Scharff’s study with the current one, several differences are evident. First, the type of steel used in Scharff’s study was S355, with a yield strength of 355 MPa, whereas high-strength steel with a yield strength of 620 MPa was used in this study. High yield strength of steel is supposed to result in a smaller diameter based on the design of the ultimate limit state. Moreover, because the pile diameter was smaller than that in Scharff’s study, the wave load applied to the pile was less in the current study, which is helpful to know when decreasing pile diameter.

#### 4. Conclusions

This paper has presented an investigation into the structural properties of monopile and tripod wind-turbine support structures. Based on a comparison of the results of static, fatigue, and modal analyses results for these two structures, the following conclusions can be drawn.(1)The maximum stress in the monopile structure was at the base, whereas the transition location from the lower central column to upper tower showed the maximum stress for the tripod structure. Moreover, for the case under study, the maximum stress and displacement for the monopile structure were approximately 2.35 and 1.32 times greater than those of the tripod structure. The tripod showed greater stiffness and better stress-control capacity than the monopile structure. (2)Under the same cyclic loading, the tripod structure has a longer lifetime than the monopile structure. This is beneficial for saving engineering cost in the long run. (3)The modal analysis indicated that the first three natural frequencies of the tripod structure were higher than the corresponding frequencies of the monopile structure. The first mode played a controlling role in the dynamic response of the monopile structure, whereas both the third and first modes had a strong influence on the tripod structure. The results for frequency separation and mass participation factor indicated that the tripod structure was more reliable than the monopile structure. (4)The analytical results indicated that both monopile and tripod structures can be used as wind-turbine support structures in the environment considered. However, comparison of structural responses using static, fatigue, and modal analyses revealed that the tripod structure is superior to the monopile structure. Moreover, it is effective to use high-strength steel in the monopile structure for optimizing the monopile diameter to be smaller than that of the upper tower bottom.

#### Acknowledgments

This research was supported by National Natural Science Foundation of China (51137002) and Natural Science Foundation of Jiangsu Province (BK2011026).

#### References

- A. R. Henderson, C. Morgan, B. Smith, H. C. Sørensen, R. J. Barthelmie, and B. Boesmans, “Offshore wind energy in Europe—a review of the state-of-the-art,”
*Wind Energy*, vol. 6, no. 1, pp. 35–52, 2003. View at Publisher · View at Google Scholar · View at Scopus - S. Malhotra, “Selection, design and construction guidelines for offshore wind turbine foundations,” in
*Wind Turbines*, I. Al-Bahadly, Ed., pp. 231–264, InTech, Zagreb, Croatia, 2011. View at Publisher · View at Google Scholar - S. Malhotra, “Design and construction considerations for offshore wind turbine foundations in North America,” in
*GeoFlorida 2010: Advances in Analysis, Modeling and Design*, D. O. Fratta, A. J. Puppala, and B. Muhunthan, Eds., pp. 1533–1542, ASCE, West Palm Beach, Fla, USA, 2010. View at Publisher · View at Google Scholar - A. R. Henderson, R. Leutz, and T. Fujii, “Potential for floating offshore wind energy in Japanese waters,” in
*Proceedings of the 12th International Offshore and Polar Engineering Conference (ISOPE '02)*, Kitakyushu, Japan, May 2002. View at Scopus - R. Scharff and M. Siems, “Monopile foundations for offshore wind turbines—solutions for greater water depths,”
*Steel Construction*, vol. 6, no. 1, pp. 47–53, 2013. - J.-H. Mo, Y.-P. He, Y.-G. Li, and E.-H. Zhang, “Fatigue analysis of offshore wind turbine mono-pile support structure,”
*Journal of Shanghai Jiaotong University*, vol. 45, no. 4, pp. 565–569, 2011. View at Scopus - M. Torcinaro, F. Petrini, and S. Arangio, “Structural offshore wind turbines optimization,” in
*Earth and Space 2010: Engineering, Science, Construction, and Operations in Challenging Environments*, G. Song and R. B. Malla, Eds., pp. 2130–2142, ASCE, Honolulu, Hawaii, USA, 2010. View at Publisher · View at Google Scholar - N. A. Agbayani, “Defects, damage, and repairs subject to high-cycle fatigue: examples from wind farm tower design,” in
*Forensic Engineering Congress 2009: Pathology of the Built Environment*, S. Chen, A. D. de Leon, A. M. Dolhon, M. J. Drerup, and M. K. Parfitt, Eds., pp. 546–555, ASCE, Washington, DC, USA, 2009. View at Publisher · View at Google Scholar - N. Bazeos, G. D. Hatzigeorgiou, I. D. Hondros, H. Karamaneas, D. L. Karabalis, and D. E. Beskos, “Static, seismic and stability analyses of a prototype wind turbine steel tower,”
*Engineering Structures*, vol. 24, no. 8, pp. 1015–1025, 2002. View at Publisher · View at Google Scholar · View at Scopus - N. NORSOK-004,
*NORSOK Standard for Design of Steel Structures*, Standards Norway, Lysaker, Norway, 2004. - Dnv-Os-J101,
*Offshore Standard- Design of Offshore Wind Turbine Structures*, Det Norske Veritas, Oslo, Norway, 2010. - T. H. Dawson,
*Offshore Structural Engineering*, Prentice Hall, New Jersey, NJ, USA, 1983. - S. K. Charkrabarti,
*Nonlinear Methods in Offshore Engineering: Developments in Marine Technology*, Elsevier, New York, NY, USA, 1990. - A. R. Henderson, M. B. Zaaijer, and T. R. Camp, “Hydrodynamic loading on offshore wind turbines,” in
*Proceeding of the Offshore Windenergy in Mediterranean and Other European Seas*, Naples, Italy, 2003. - The M5000 wind turbine (AREVA), http://www.areva.com/FR/activites-4430/l-eolienne-m5000.html.
- P. Schaumann, S. Lochte-Holtgreven, and S. Steppeler, “Special fatigue aspects in support structures of offshore wind turbines,”
*Materials Science and Engineering Technology*, vol. 42, no. 12, pp. 1075–1081, 2011. View at Publisher · View at Google Scholar · View at Scopus