In recent years efforts have been made to increase the production of renewable energy such as wind energy. The industry increases rapidly and wind turbines continue to grow in both size and numbers. In addition new building sites are incorporated as large areas are required in order to build the wind farms. This means that oshore wind farms are being built increasingly farther from the coast and in deeper waters. The turbines are often placed at water depths of 15 - 30 m. The most common oshore foundation for wind turbines are monopiles. These monopiles often have an embedded length of 20 - 30 m and a diameter of 4 - 6 m.
When designing monopiles in regard of lateral loading, current design guidances, i.e. DNV (2010) and API (2007), use the method of p -y curves. The p -y curves are based on a few static and cyclic tests on a few
flexible, slender piles, as described in (Cox et al., 1974).The p -y curves are formulated depending on very few properties of the sand and the pile, respectively. For the sand, the angle of internal friction, the relative density, and the specific weight is considered. The dimensions of the pile are considered in terms of length and diameter. However, the general behaviour of the pile is assumed that of slender piles. The monopiles today have a slenderness ratio < 10 and so, this will give the piles a more rigid response which is not accounted for in the current design guidances. Another subject where the design guides are not up to date, is their limited implementation of issues regarding long-term cyclic, lateral loading. This effect may change the stiffness of the soil-pile system and cause a tilting rotation of the wind turbine. In recent years 3D finite element analysis has become a tool in the investigation of complex geotechnical situations, such as the laterally loaded monopile. In this paper a 3D FEA is conducted as basis for an evaluation of the p-y curves of the design guides. It is found that the applied material models have a significant influence on the stiffness of the obtained p-y curves. p-y curves are obtained by evaluation of soil response during a prescribed displacement and applied load respectively. The responses are not in clear agreement. The p-y curves evaluated by means of FEA are compared to the conventional p-y curve formulation which provides a much stiffer response.
In order to evaluate the effect of cyclic lateral loading a small-scale test of a pile placed in saturated sand is conducted. The pile is 100 mm wide and has a slenderness ratio of 6. The cyclic load affecting the pile is found from the lateral bearing capacity which is defined at a rotation of 3. The cyclic load is determined as 35 % of this load. Force and displacement is measured as the pile is loaded to evaluate the rotation of the pile. The cyclic test shows decreasing displacement increments with increasing number of load cycles, but a stabilised situation does not occur. A literature study on state of the art knowledge within the field of cyclic loading is conducted. Theories on degradation of the stiffness of the soil-pile system by Long and Vanneste (1994) and Lin and Liao (1999) are presented as well as recent experimental work on cyclically loaded piles by Peng et al. (2006), Peralta and Achmus (2010), LeBlanc et al. (2010) and Roesen et al. (2011). The measured test results are compared with the theoretical formulations as well as other cyclic load
tests. Long and Vanneste (1994) and Lin and Liao (1999) suggest formulations that compared to the measured results give simple estimates on the accumulated rotation of the pile. The measured
result agree with recent experimental work that rotation of the pile will keep increasing with increasing number of load cycles. However, in contrast to the measured results, Roesen et al. (2011) finds that the system stabilises. After 15000 load cycles no further increase in rotation occurs.