Abstract

In this study we investigate the development of shear zones due to the settlement of shallow foundations and their load-settlement behavior. Firstly, a well-documented experiment of shallow penetration into sand is used for the validation of the soft particle code (SPARC). For these simulations a hypoplastic material model for sand with calibration for the model sand is implemented in SPARC. In order to deliver a more comprehensive investigation, the shape of the shear zones predicted by SPARC is also compared with the analytical solution. Secondly, the penetration of shallow foundation into clay is investigated by means of SPARC and the finite element method. For this purpose, barodesy for clay with the calibration for Dresden clay is implemented in both numerical methods. The simulations are carried out for six different surcharges, corresponding to a range of over-consolidated clay to normal-consolidated clay. Furthermore, the load-settlement behavior and the shape of shear zones for both methods are compared and the weaknesses and strengths of each numerical approach are discussed. Finally, the peaks of the load-settlement curves for all surcharges are compared with the analytical solution. Results show that SPARC performs better at predicting the trajectories of particles under the foundation, which consequently leads to better estimation of the load-settlement behavior.

Highlights

  • Since the emergence of the FE methods, numerical methods have become the so-called third pillar besides theory and experiment in understanding the response of structures and analysis of engineering problems

  • Evaluation of incremental volumetric strain of the experiments in Aubram (2013), p. 291, show that no extreme densification occurs in the wedge below the foundation, which corresponds to the prediction of Soft PArticle Code (SPARC) in comparison to the predictions made by the arbitrary LagrangianEulerian (ALE) method in Aubram (2015) where a densification up to e = 0.482 in the wedge has been predicted

  • The results show that SPARC has been able to predict the load-settlement behavior well and the peak of the curve is predicted at the same relative penetration z/B of the experiment

Read more

Summary

Introduction

Since the emergence of the FE methods, numerical methods have become the so-called third pillar besides theory and experiment in understanding the response of structures and analysis of engineering problems. One of the shortages of most numerical methods is the inability to simulate large deformations due to settlements, installation processes (e.g. penetration) and excavation, the simplified wished-in-place method is vastly employed in numerical simulations. Several studies have proven that the final deformations in geotechnical projects have been larger than those estimated by the numerical methods and the unforeseen deformations are mainly caused during the installation phases, e.g. Settlements due to the penetration of shallow foundations cannot be simulated by means of numerical methods, since they are associated with large and non-topological deformations. In contrast to FE methods, meshfree methods have no fixed connectivities between the points and are more appropriate for problems associated with penetration. The meshfree code (SPARC) applied in this study has been earlier introduced in Ostermann et al (2013), Schneider-Muntau et al (2017), Michel et al (2017), Chen (2014) and Polymerou (2017)

Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call