Offshore wind turbines are a highly competitive source of renewable energy. The
development of these offshore installations has advanced quickly, therefore the
elaboration of a specific regulation and theoretical method for the design of wind
turbine foundations is needed.
Several types of foundation concepts can be employed for offshore wind turbines, i.e.
gravitational foundation, monopile foundation, bucket foundation, tripod foundation,
etc. For water depths of 15 to 30 m, the monopile foundation concept has been
employed in many projects. The focus of this report is therefore on monopile
foundations. The goal of the project has been to understand and investigate the
uncertainties in the design method currently being used for monopile foundations.
Small-scale tests have been carried out in a pressure tank with the objective of
minimising small-scale effects. Afterwards, these tests have been modelled by means of
three dimensional numerical methods by the programs FLAC3D and Plaxis 3D 2010. A
dimensionless analysis is realised through the project with the goal of analysing the soil
exponent and deriving a general equation which relates the lateral load to the lateral
displacement of the pile. A full-scale model is also modelled in Plaxis 3D 2010 and
FLAC3D.
The method followed to design monopile foundations in sand is based on the p-y curve
method, but this method does not fit the requirements of the monopile foundations
properly. The reason of the problem is that p-y curve formulations were developed for
lateral loaded piles with higher diameters and slenderness ratios, designed for offshore
petroleum industry, where most of the discoveries were located in areas with soft clays
primarily, extended later to the knowledge of the behaviour in sands.
Therefore, the methods developed for the oil industry, the jag piles, are based on slender
piles with different geometries and properties than the wind turbine monopiles, such as
the slenderness ratio. Slenderness ratios for wind turbine monopiles are considered less
than 10, due to the fact that they have bigger diameter and shorter embedded length than
the jags piles, which are considered slender piles.
Another uncertainty in the actual method is that the different diameters of the piles and
the initial stiffness have not been taken into account due to the fact that they are
designed regarding to the ultimate limit state, where it has been assumed that possible
initial deformations of the soil are not included in the calculations. Thereby, in the
service limit state, the initial stiffness is very important for the design of monopile
foundations. In the case that the monopile rotates a bigger angle than the maximum
allowed by API design regulations, the efficiency of the wind turbines decreases.
The experimental tests are realised on a monopile with strain gauges installed on it
which can measure the strain at eleven different levels through the embedded length of
the pile. In total 22 tests with different slenderness ratios and pressure levels are carried
out since 2009, but only 10 have been conducted by Borobia, Mikalauskas and Troya
during the present year. Afterwards, the same tests have been modelled numerically by
FLAC3D and Plaxis 3D 2010 in order to analyse the possible uncertainties between
different methods and a dimensionless analysis is performed to derive a general
equation which relates the lateral displacement of the pile with the load applied.