Centrifuge modelling of piled embankments

Abstract

It is becoming increasingly necessary to construct on land that was previously considered inappropriate for construction, such as soft clay. The properties of soft clay make it highly compressible and low in shear strength, meaning that bearing capacity failure and excessive settlement are of concern.Piled embankments are a ground improvement technique that can provide a solution for this problem. Piled embankments have the ability to transfer the greater part of the embankment load and any surcharge to more competent material at greater depth due to the 'arching' concept. Consequently, the soft foundation soil has little direct impact on the performance of the embankment. The concept of 'arching' of granular soil over an area where there is partial loss of support from underlying strata has long been recognised in the study of soil mechanics (e. g. Terzaghi, 1943). However, a number of competing theories exist to quantify this behaviour in piled embankments.In addition, the use of geotextile reinforcement in piled embankments placed above the pile caps in principle provides a number of technical as well as economical benefits. As the embankment fill is placed, tension is created in the reinforcement and it is the vertical component of this tension that transfers the embankment load onto the piles and reduces the load carried by the soft clay hence transferring the load of the embankment on to the piles.Differential settlement can be a problem for piled embankments of low height.Significant differential settlement can cause undesirable effects on any structures constructed on the embankment. 'Arching' limits the amount of differential settlement in embankments and the use of geotextile geogrid can also potentially have additional benefits.This thesis presents a series of centrifuge tests examining the performance of unreinforced and reinforced piled embankments constructed over soft subsoil in terms of stress acting on the subsoil, and differential movement at the surface of the embankment. A large range of embankment heights are considered, and the results for stress on the subsoil are compared with existing predictive methods, allowing generic conclusions to be drawn regarding the predictions of various methods.The effect of a 'working platform' below pile cap level and thus directly loading the subsoil is also considered, and used to examine the concept of a 'Ground Reaction Curve' (Iglesia et al, 1999) for arching in the embankment. In principle this can be used to consider compatibility of displacements at the base of the embankment, and thus improve design simultaneously considering the effect of arching in the embankment and underlying support from the subsoil and layers of reinforcement acting in tension.