Centrifuge Modelling Of Displacement Piles and Installation Effects

"This paper presents some recent advancements in centrifuge modelling, which enable the installation of displacement piles to be performed in-flight while measuring ground responses. Details and results are provided from a series of experiments where 21 full-displacement piles were installed into a sample of consolidated kaolin clay at 40g. The piles were pushed into the sample at a constant rate of penetration, with installation reaction loads and pore pressures measured. Insight into the installation effects of the full-displacement piles was obtained by undertaking in-situ testing pre- and postinstallation of piles. The findings show that centrifuge testing has the potential to play an important role in furthering the current understanding of displacement pile behaviour and installation effects.INTRODUCTIONDisplacement piles and displacement ground improvement columns cause soil to undergo significant distortion and shearing during installation, which alters the soil properties. It is essential to consider these changes in soil properties to estimate the behaviour of pile foundations with higher degree of accuracy. However, the installation effects of displacement piles are widely considered as one of the most difficult geotechnical mechanisms to model, both analytically and numerically, and experimental data is needed to advance the current state of knowledge.Theories, analytical and semi-analytical, have been developed to describe the displacement pile installation process. Randolph et al. (1979) applied the cylindrical Cavity Expansion Method (CEM) to model the installation of a displacement pile in a fine grained cohesive soil. However, cylindrical CEM only considers the field variables in radial coordinates, and hence the soil-pile interface can only be modelled as a frictionless material. Baligh (1985) introduced the Strain Path Method (SPM), which assumes soil flows around the tip of a displacement pile. SPM was shown to model the stress and strain fields, particularly within close proximity to the pile, more accurately compared to the cylindrical CEM. While these methods account for the installation effects of a displacement pile and can be used to estimate a resulting ultimate bearing capacity, they are unable to describe the load-settlement behaviour of such piles. With the added complexity of group effects, existing analytical methods do not provide a means to accurately calculate the ground response to the installation of a group of displacement piles in a cohesive soil.The advancement of computational power, numerical techniques and constitutive models has improved the accuracy of numerical analysis, particularly in regard to modelling pile installation. However, even with advanced techniques, Finite Element Method (FEM) suffers a number of disadvantages due to the large deformations, which results in mesh distortion and tangling. Mesh-free methods such as Discrete Element Method (DEM), Smoothed Particle Hydrodynamics (SPH) and Material Point Method (MPM) are the next generation of computational techniques that are able to deal with large deformation problems. However, these advanced techniques are computationally expensive, complex and yet to be incorporated into commercial packages. Such methods are considered to still be in the development stage and are yet to provide a complete solution to pile installation problems."