### Filtern

#### Dokumenttyp

- Beitrag zu einem Tagungsband (4)
- Vortrag (3)
- Zeitschriftenartikel (1)

#### Schlagworte

- Suction bucket (3)
- Aberfan flowslide (2)
- Finite element method (2)
- Generalized plasticity (2)
- Numerical modelling (2)
- Numerical simulation (2)
- Offshore geotechnics (2)
- Offshore wind energy converter (2)
- SPH (2)
- Sand (2)

#### Organisationseinheit der BAM

Suction Bucket Jackets (SBJ) are found as a suitable alternative to driven piles for the support of foundations for offshore wind energy converters. In the case of jackets or multipods, a predominant vertical load is to be expected. The effect of such a tensile loading is the generation of suction in the soil inside the bucket which leads to an increment of tensile capacity. This paper aims to study the bearing behaviour of a suction foundation by taking into account how the soil permeability and the loading rate influence the foundation behaviour. Moreover, after submitting the structure to a storm load, the bearing capacity is studied again, in order to see the effect of such a load on the bucket's bearing behaviour. This study is carried out by means of Finite Element numerical simulations based on the formulation of Biot's equations combined with a constitutive model that reproduces the key aspects of cyclic soil behaviour in the frame of Generalized Plasticity.

Suction Bucket Jackets (SBJ) are found as a suitable alternative to driven piles for the support of jacket or tripod foundations for offshore wind energy converters. Offshore wind energy turbines are characterized by a small self weight and they can be subjected to different load combinations. The work presented here aims to show the numerical investigation on the behavior of suction bucket foundations under different kind of loads as well as load combinations. In order to do so, a suitable numerical model is much needed. The theoretical basis of the model lies on the Swansea formulation of Biots equations of dynamic poroelasticity combined with a constitutive model that reproduces key aspects of cyclic soil behavior in the frame of the theory of generalized plasticity. An adequate FE formulation, the representation of appropriate soil-structure interfaces and the computational efficiency are key aspects in order to successfully model such complex systems. The 3D numerical simulation allows a special insight into the fundamental behavior of the founding of Suction Bucket Jackets such as the evolution of the pore water pressure or the occurrence of the so called soil liquefaction.

Landslides can cause major economic damage and a large number of casualities as it is possible to see from past events occurred all over the world. Being able to predict these kind of hazards would then suppose the achievement of great benefits. Here a model that combines a depth integrated description of the soil-pore fluid mixture together with a set of 1D models dealing with pore pressure evolution within the soil mass is presented. The mathematical model is based on the Biot-Zienkiewicz equations, from where a depth averaged model is derived. Concerning the material behaviour, the approach used is the one suggested by the Perzyna viscoplasticity, which has been extensively used in the past to model solid behaviour prior to failure. In this framework, a simple shear rheological model is derived, providing the basal friction needed in depth integrated models. The Smoothed Particle Hydrodynamics (SPH) has been the numerical technique chosen to spatially discretised the depth integrated equations of the mathematical model. The purpose of this work is to apply the SPH depth integrated numerical model, together with the sub-model that predicts the evolution of the pore water pressure inside the landslide, to simulate the propagation phase of the Aberfan flowslide occurred in 1966.

The choice of a pure cohesive or a pure frictional viscoplastic model to represent the rheological behaviour of a flowslide is of paramount importance in order to obtain accurate results for real cases. The principal Goal of the present work is to clarify the influence of the type of viscous model—pure cohesive versus pure frictional—with the numerical reproduction of two different real flowslides that occurred in 1966: the Aberfan flowslide and the Gypsum tailings impoundment flowslide. In the present work, a depth-integrated model based on the v-pw Biot–Zienkiewicz formulation, enhanced with a diffusion-like equation to account for the pore pressure Evolution within the soil mass, is applied to both 1966 cases. For the Aberfan flowslide, a frictional viscous model based on Perzyna viscoplasticity is considered, while a pure cohesive viscous model (Bingham model) is considered for the case of the Gypsum flowslide. The numerical approach followed is the SPH method, which has been enriched by adding a 1D finite difference grid to each SPH node in order to improve the description of the pore water evolution in the propagating mixture. The results obtained by the performed simulations are in agreement with the documentation obtained through the UK National Archive (Aberfan flowslide) and the International Commission of large Dams (Gypsum flowslide).

The purpose of the work presented in this paper is to analyze locally (at the element level) the contact behavior of a soil-pile contact problem. Therefore, a 2D shear test is modeled using the Finite Element Method. The formulation of a 4 nodded zero-thickness interface element of Beer is chosen with a linear interpolation function. Four constitutive contact models adapted for contact problems have been implemented. The Mohr-Coulomb and Clough and Duncan models were chosen initially, due to the ease of implementation and few number of parameters needed. After, more complicated models in the framework of elasto-plasticity such as: Lashkari and Mortara were implemented for the first time into the finite element code of the shear test problem. They include other phenomena such as: relative density of soil, the stress level and sand dilatancy. From the results the relation between shear displacement and shear stress has been deduced. Finally, a discussion of the advantages and the drawbacks during computation of each model is given at the end.