Icon EnvelopeFacebook iconIcon LinkedInIcon Youtube
HUESKER Geosintéticos S.A.U.
C/ O. Nave 8
31110 Noain (Navarra)
España

Analysis of a self-regulating foundation system for embankments on soft soils by means of centrifuge tests and numerical simulations

Abstract

A new innovative foundation system for embankments on soft soils is currently being analyzed at the Ruhr-Universität Bochum, Germany, in cooperation with HUESKER Synthetic GmbH, Germany. The foundation system consists of two parallel vertical walls, which are installed into the soft subsoil and connected at ground level via a horizontal geosynthetic reinforcement which acts as a tension membrane (further called tension membrane). The embankment is constructed on top of the tension membrane. The system ensures global stability, controls the horizontal deformation and reduces the hori-zontal thrust in the subsoil on either side of the embankment. The complex interactive system behaviour is analyzed by centrifuge model tests and numerical simulations. Results of the centrifuge model tests are presented and interpreted in this paper. Conclusions and presumptions from the measurements of the cen-trifuge tests will be analyzed and verified thereafter by numerical simulations. One of the main findings is the activation of an arching mechanism within the embankment even with the relatively low ratio of 0.25 between the final embankment height and base width. The relevance of this mechanism for the design will be discussed at the end of this paper.

Conclusion

Comparing the relation of embankment height and base width, which is 0.25 for the final embankment height, with the relation mentioned in the literature for the formation of a full arch (70% to 100% for an uniform thick layer), it becomes obvious that no full arch has developed within the embankment. How-ever, the decrease of the vertical stresses within the consolidation phases can clearly be attributed to an arching mechanism probably enhanced by spreading forces.

For the safe system design, the decrease of the vertical pressure beneath the embankment axis is less important. The consideration of this effect would lead to a better prediction of settlement.

More important is the consideration of the load redistribution towards the vertical wall, since this leads to an increased load along the vertical wall, which does have a significant influence on the design for ULS as well as the SLS of the system.

There are still disagreement between measurement and numerical simulation, which may result from the ideal assumed boundary conditions, i.e. no friction at the side wall of the strong box on the axis of symmetry. The next step is to do the numerical simulation with more realistic values.