Numerical study of basal reinforced embankments supported on floating/end bearing piles considering pile–soil interaction

Abstract

Construction sites consisting of soft soils may require ground improvement to prevent excessive settlements or bearing capacity type failures and shear movements, which results in construction delays and premature failures. Among the various ground improvement techniques, the Geosynthetic Reinforced Piled Embankment Systems (GRPES) provide a practical and efficient solution due to the low cost and short construction times. Most of the piled embankments are constructed on end bearing piles. At large depths of foundation soil, floating piles are more economical and technically feasible than the end bearing piles. The design of floating piles involves complex soil–structure interaction and there are no clear uniform guidelines available for the design of embankments supported on floating piles. This paper presents the results of numerical investigation into the performance of geosynthetic reinforced embankments supported on end bearing as well as floating piles considering the pile–soil and geosynthetic–soil interaction. 3-D Column models are employed to carry out the parametric studies on factors such as the development of arching, skin friction distribution along the pile length and axial force distribution. Full three-dimensional analyses are carried out to study the overall behavior of the GRPES system and the results obtained from the analyses were compared with those from British Standard BS8006-2010. The results indicated that the use of floating piles could considerably reduce the settlements and the embankment load transferred through the piles to the foundation soil is found to depend very much on the length of the piles. This aspect needs to be accounted for while calculating the arching factor in the empirical equations.

Introduction

Rapid development of infrastructure projects like highways, railways and other industrial activities created large pressure to utilize even marginal lands for construction purposes. The construction of these structures, which exerts large amount of load on a wide area on soft foundation soils, is a challenging task because of the low bearing capacity, high compressibility and tendency for lateral flows, etc. Suitable ground improvement/treatment methods have to be used to improve the engineering properties of such foundation soils. If no treatment of the ground is possible due to the urgency of the project, suitable foundation techniques need to be adopted to directly transfer the loads to the hard stratum at some depth below the ground level. In the recent years, the geosynthetic reinforcement has been used in combination with pile or column systems to support embankments over soft clay soils. The Geosynthetic Reinforced Piled Embankments Systems (GRPES) are preferred over conventional column embankments as the inclusion of the geosynthetic reinforcement at the base of the embankment eliminates the need for raked piles to resist large lateral pressures (Han and Gabr, 2002) and provides an economical solution (Chen et al., 2008, Zheng et al., 2009). Field data indicated that the use of basal reinforcement limits the lateral shear displacement of the subsoil at the edge of the embankment (Liu et al., 2007). No procedure is currently available to predict the effect of GRPES on the lateral displacement of the subsoil.

The load transfer mechanism in GRPES is a combination of the following three phenomena (a) soil arching (2) tensioned membrane action of the geosynthetic reinforcement and (3) stress transfer from the soft soil to the pile due to the difference in their stiffness. The design of GRPES includes the design of embankment geometry, reinforcement strength, pile dimensions and spacing. Based on the embankment load transfer to piles through arching, the reinforcement tensile strength requirements are determined. Different design methods are currently used to determine the load on the piles due to arching (Hewlett and Randolph, 1988, Russell et al., 2003, Collin et al., 2005, BS8006, 2010, EBGEO, 2010, van Eekelen et al., 2013). It is clearly evident from the published literature that the design procedures give conservative designs due to the simplifying assumptions made and therefore they are good for preliminary design of piled embankments (Russell and Pierpoint, 1997, Stewart and Filz, 2005, Abdullah and Edil, 2007, Smith and Filz, 2007). These design methods can be refined further to reduce the construction costs (Abdullah and Edil, 2007).

Most of the piled embankments are constructed on end bearing piles. Providing end bearing piles in very thick soft clay soils is not economically feasible (Poulos, 2007). In such cases, floating piles, where the pile tip does not reach the hard stratum and the embankment load transfer to piles is mainly due to skin friction are required. However, the current design procedures are suitable only for the design of embankments on end bearing piles. Though BS8006 (2010, Vol. 1) mentions about floating piles, it gives only a cursory procedure for the design of embankments supported on floating piles. As per BS8006 (2010, Vol. 1), the arching coefficient, C_c used for end bearing piles is reduced to consider the reduced arching action of the floating piles. The reinforcement load is then calculated with the assumption that the reinforcement is spanning the rigid piles with no subsoil support. In reality, the reinforcement layer feels the support given by the subsoil (van Eekelen et al., 2011, Van Eekelen et al., 2015). Van Eekelen et al., 2015 observed that in extreme situations the subsoil support can be lost. Such situations occur due to the lowering of water table in the foundation soil or the settlement of soft foundation soil under the weight of working platform below the reinforcement.

The German design standard (EBGEO, 2010) states the need for further research for the design of embankments on floating piles. Similar recommendation is also made by Satibi (2009) and van Eekelen et al. (2011). The frictional force along the length of the pile affects the GRPES behavior and in the case of embankments supported on floating piles, the relative deformation between the pile and the surrounding soil is a function of the complex soil-structure interaction between the reinforcement, foundation soil and the pile (Abusharar et al., 2009, Jenck et al., 2009, Bhasi and Rajagopal, 2013). The present study explores the time dependent behavior of geosynthetic-reinforced embankments supported on end bearing as well as floating piles considering pile-soil and geosynthetic–soil interaction. In this work, 3-D column models are developed for carrying out the parametric studies to investigate key factors such as the development of arching, skin friction distribution along the pile and the axial force distribution. Full three dimensional models are developed to study the overall behavior of the system. Numerical values of the load transferred to the piles and the tensile forces developed in the geosynthetic are compared with the recommendations given in British Standard BS8006 (2010, Vol. 1). Based on the studies, modified arching coefficients are proposed.