Numerical modelling of time-dependent negative skin friction on single pile and pile group in soft soils

Abstract

Over the past few decades, increasingly taller and larger structures have been constructed, making piles a crucial foundation solution. However, the adoption of pile foundations in consolidating soil can pose significant design challenges, particularly the mobilization of the negative skin friction (NSF). NSF induces additional axial forces on the pile, imposing a detrimental rather than a beneficial load. While empirical methods derived from field observations for estimating NSF can be applied in preliminary foundation design, some design codes remain overly simplistic. Moreover, the creep effect on NSF is often ignored in these design methods. The soft soil creep effect on the long-term development of NSF has remained poorly understood. This thesis aims to incorporate the elasto-viscoplastic model into the NSF analysis using the rigorous finite element method (FEM) for both single pile and pile groups, providing insights into current design methods while refining them through parametric studies. In this study, the two-dimensional axisymmetric single pile-soil interaction model is first established and calculated to examine various degrees of creep effects on the variation of NSF and neutral plane (NP) during and after the primary consolidation periods. According to the findings, a high creep coefficient of the soil results in an increase in NSF and a descending trend of the NP. The creep induced delay of NSF is observed attributed to the increase in excess pore pressure during the early stage of consolidation. The NP position varies drastically at the commencement of consolidation when taking creep into consideration. An exponential prediction model to reflect the time dependence of the location of the NP is proposed. Subsequently, numerical investigations further extend to three-dimensional analysis, validated through centrifuge tests, are conducted to examine the effect of sacrificial piles on the dragload reduction in the center pile (termed as group effect) at varying pile spacings. A parametric investigation is conducted to quantify the group effect under different variables, including pile spacing, end-bearing layer stiffness and creep coefficient. Results reveal that beyond the spacing of 7 d (where d is the diameter of sacrificial piles), the group effect can be neglected. Furthermore, the group effect is highly dependent on the site-specific creep behavior, becoming less significant under high creep conditions. The reduction in effective stress of soil within a pile group is identified as the primary cause of the NSF pile group effect. Pile penetration in soft ground involves complex mechanisms, including significant alterations to the surrounding soil state, which influence NSF over time. However, pile penetration is often excluded from finite element analysis. The impact of pile penetration on the NSF generation is thus analyzed. A novel stable node-based smoothed particle finite element method (SNS-PFEM) framework is introduced for two-dimensional axisymmetric conditions and coupled consolidation, incorporating the ANICREEP model for soft soil with a modified cutting-plane algorithm. A field case study with penetration process is simulated to verify the numerical model's performance, followed by a parametric analysis on the effect of penetration rate on NSF during consolidation. Results indicate that excluding pile penetration from NSF analysis can result in an unsafe underestimation of NSF and dragload magnitudes. The penetration rate affects dragload only at the initial consolidation stage. As consolidation progresses, dragload converges to nearly the same magnitude across different rates. Additionally, current design methods inadequately predict the β value (where β is an empirical factor correlating vertical effective stress of soil with the pile skin friction) and its time dependency, for which a new empirical formula for the time-dependent β value is proposed and successfully applied to other field cases. Finally, conclusions are drawn, and future work is outlined.