In the low-lying parts of The Netherlands challenging ground conditions are often encountered. Thick layers of soft material such as peat and clay are present and pile foundations into deeper sand layers are required when objects are constructed that have to transfer significant loads to the subsurface without inducing settlement (differences). This also counts for pile foundations of wind turbines which, apart from compression and tension loads also are subjected to horizontal loading. The horizontal translational stiffness is an important and depending on the wind turbine manufacturer governing aspect in the required dimensions of the pile foundation (number and type of piles). In order to improve this horizontal translational stiffness, the foundation designer usually opts for the use of battered (raked) piles. The basic principle is that the axial stiffness of the pile will, in this way, contribute to the horizontal stiffness of the foundation as a whole. The verification of wind turbine requirements usually involve the use of generally accepted analytical 2D methods, simplifying the more complex 3D effect, to model the interaction between soil-pile and the group effect for closely spaced piles. Although the analytical model generally provides good results, investments can reduce significantly when only a small amount of piles can be saved upon. Also the effort in the field is reduced when using vertical piles...

TheNetherlands is a densely populated country and is located in an areawhere the shallowsubsurfacemainly consists ofHolocene and Pleistocene soils. To build permanent structures, enough bearing capacity is needed to provide sufficient support for these structures. However, Holocene (soft) soils typically do not provide sufficient support. In the western part of the country, the soft soils from the Holocene overlay the stiffer Pleistocene soils. To still be able to construct at locations where soft soils are found, pile foundations are needed to transfer the loads through the soft soil onto the stiff soil in order to provide for sufficient bearing capacity. The normativemethod for determining pile bearing capacity in the Netherlands uses a relatively straightforward semi-empirical approach known as the 4D/8D or the Dutch method. This method was based on the work of [vanMierlo and Koppejan, 1952] who introduce the logarithmic spiral theory. Apart from small modifications and elaborations, the method has been used ever since. To get to a reasoned solution on how to calculate and design pile foundations the Eurocode [Normcommissie 351 006 ’Geotechniek’, 2017a] was introduced. One of the (original) aims of the Eurocode was to harmonise rules and regulations regarding, amongst others, pile foundations. To prepare this harmonisation, Belgium, France and the Netherlands took a closer look into their calculation methods. For the Netherlands this research was performed by a committee who presented their findings in CUR 229 - “Axiaal belaste palen” [CUR B&I, 2010]. From CUR 229 it was found the pile tip capacity was overestimated. In other words, the calculated pile tip capacity was higher than the measured pile tip capacity. This overestimation led to a reduction of ®p of 30% starting on the 1st of January 2017. (®p is the factor which reduces the pile tip resistance for different pile types, see Appendix A.) A point of interest is the overestimation of the pile tip capacity, which seems to become more prominent when the pile goes deeper into a non-cohesive soil. It was even found, a lower ®p is not required for piles installed less than 8D into the bearing layer. The reduction of ®p seems to contradict with practice, because no cases of damage are known regarding the bearing capacity of pile foundations. Several explanations for this might be valid. For example, hidden safeties may prevent overall safety issues to arise. In addition, errors may occur in the determination of the pile capacity from pile load tests or in the Dutch method for calculating pile tip capacity. This leads to the two main subjects for this thesis. On the one hand, the hidden mechanism of residual loads is considered and on the other hand, the zone of influence at the pile tip during failure is analysed. To analyse the above-named subjects, the 4D/8D method was put under scrutiny. First, it was compared to other (international) analytical methods, which determine the bearing capacity directly from CPT data. Furthermore, the zone of influence around the pile tip is considered inmore detail by analysing the analytical approach and comparing this to the results of numerical models. This led to the conclusion that the observed overestimation can partly be explained by the inaccuracy of the assumed zone of influence around the pile tip, especially regarding the extent of this zone above the pile tip (8D). As this extent is too large, the 4D/8D method results in a too low average cone resistance in the 8D zone (for piles installed less than 8D into the bearing soil layer). In combination with the ‘old’ ®p , this leads to a reasonably accurate estimation of the pile tip capacity compared to the measured tip capacity. However, the observed extent of the zone of influence above the pile tip, in the FEMmodels, is in the order of 1 to 1.5D. The average cone resistance in this zone will in most cases be higher than in a zone extending 8D above the pile tip. The hidden mechanism of residual loads is implemented in a numerical model, with the use of a static load on top of the pile. The residual loads are caused by the installation of the pile and so, they are considered to be installation effects. No analytical or numerical models or procedures were found in literature to quantify the xiii xiv 0. Abstract residual load. Residual loads were implemented using a procedure defined in this thesis, taking into account the maximum pile tip capacity as indicator for the order of magnitude for the residual load. This is done, because a drawback of FEM models is; they are not able to calculate large strains, which occur during installation of a pile. Therefore the installations effects (like residual loads) have to be implemented indirectly. The horizontal compression due to installation is modelled as an installation effect according to the procedure of [Broere and van Tol, 2006]. Both installation effects are validated using data from CUR 229 and they are subjected to a sensitivity analysis. Furthermore, the softening or peak behaviour of soils might have a significant influence on the zone of influence around the pile and on the residual loads, as for both cases, the soil is loaded up to failure. Most regular constitutive models do not take into account this behaviour. Due to the fact the Hypo Plasticity (HP) model takes into account this peak behaviour, it was used to performthe FEM analysis. During the final stages of this thesis, scaled pile load tests were performed. The results of these tests were analysed to find the presence and order of magnitude of the residual load in a pile. However, as the test results were incomplete, only first assumptions were made that indicated the presence of residual loads. No substantiated conclusion could be drawn regarding the magnitude. Due to the observed importance of residual loads in this thesis, the residual loads should be measured during load tests more extensively using Osterberg cells or fibre optics. The installation effects are modelled in the Hypo Plasticity model with reasonable confidence, but future research is needed on models that can model the installation process (for example theMaterial PointMethod). Furthermore, the small strain parameters of the HP model should be taken into account. This thesis questions the 4D/8D method and shows some of its inaccuracies. However, more extensive research should be done considering the extent of the zone of influence and the limiting value defined by in the Dutchmethod. Also, the interaction between the shaft and the pile tip resistance has to be evaluated, as this thesis indicates they interfere with each other if the Dutch method is considered. Besides the zone of influence and the residual loads, the failure criterion stated by NEN 9997-1 is questioned. Failure of a pile happens when the pile tip has moved 10%D [Normcommissie 351 006 ’Geotechniek’, 2017a], but the FEM models show more displacements at failure, where failure is defined by themoment the models cannot stabilise anymore and therefore fail to calculate for a certain load step.