ALE and CEL formulations of pile installation in sand

Numerical simulation of large deformation problem is challenging due to various factors including large soil deformation, non-linear behavior of soil, the potential influence of pore water, and high computation cost. This work presents a a sophisticated numerical model which addresses the aforementioned aspects. A key feature of the numerical model is the capability to simulate potential pile tip damage during the installation process. Then numerical model employs Multi-Material Arbitrary Lagrangian-Eulerian (MMALE) method to solve the issues associated with large deformation. The performance of MMALE is evaluated against several benchmarks for which analytical and experimental results are available. An advanced constitutive model, based on hypoplasticity concept, is used to model sandy soils commonly encountered in the German northern sea bight. These models offer the advantage of capturing complex soil behavior, including changes in stress and density states. The material model is verified using element tests and subsequently integrated with the MMALE element formulation to simulate various pile installation problems. The accuracy of the numerical model is checked against the measurements. In case of any discrepancies, the underlying reasons are discussed. In addition, a hydro-mechanically coupled formulation for the MMALE method is developed to simulate a saturated soil, i.e., soil whose pores are filled with water. The theory of the coupled formulation is discussed in detail, and assumptions made to develop the current formulation are explained. Later, the performance of the coupled formulation is investigated in various benchmarks ranging from one-element tests to 2D and 3D consolidation problems. The numerical model is further adapted for parallel computation to enable calculation on high-performance computing platforms. Following verification, this parallelized numerical tool is employed to simulate large-scale problems. The use of parallel computation significantly reduces the computational time and provides accurate results.

Author dc:creator

• Daryaei, Reza

Advisors dc:contributor.advisor

• Rackwitz, Frank
• Aubram, Daniel

Licence dc:rights.uri

• https://creativecommons.org/licenses/by/4.0/

Language dc:language.iso

en

Identifier URI

https://doi.org/10.14279/depositonce-23319

OAI identifier oai:identifier

oai:depositonce.tu-berlin.de:11303/24503