908
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
paper, which is believed to be a more feasible and economical
solution under certain boundary conditions. The new concept
and its application areas are presented in the following sections.
2. NEW SELF-REGULATING FOUNDATION SYSTEM
2.1. Description and basic ideas
The new self-regulating foundation system consists of two
vertical parallel walls (e.g. sheet pile walls) which are installed
at a certain distance between each other into the soft soil and
connected to each other by a horizontal tension membrane (e.g.
geotextile). The tension membrane is assumed to cover the
whole area in between the vertical walls. The vertical walls may
end within the soft soil layer or reach further down into a firm
layer. The soft soil beneath the embankment is therefore
confined by the membrane on top and the vertical elements
(Figure 3).
The embankment will be constructed above the tension
membrane, which is connected to the walls. This surcharge
generates vertical and horizontal pressures and corresponding
strains in the soft soil. The horizontal thrust tries to move the
walls outwards. At the same time tension forces are mobilized
in the tension membrane: first due to settlements (deflection)
beneath the embankment and second due to the outward
movements of the vertical walls tensioning the connected
membrane additionally.
Figure 3. Sketch of deformed tension membrane foundation system
The basic ideas of the system are on the one hand to confine
the soft soil by the vertical and horizontal elements to prevent
excessive lateral deformation or even extrusion of the soft soil.
This confinement results also in reduced vertical deformation.
On the other hand a self-regulating mechanism of the system
takes place. Each load increment provokes an increased
horizontal pressure on the vertical walls and therefore a further
outward deformation. This deformation results in a larger strain
of the tension membrane and a corresponding higher tensile
force. Thus the later provides an increased resistance to the
outward displacement tendency of the walls. Say, the system
reacts to a higher surcharge with a higher lateral restrain.
The foundation system not only ensures the global stability
of the embankment but also “automatically” prevents or reduces
deformations.
2.2. Overview on related systems
The use of a geotextile basal reinforcement is a well established
and documented method for the construction of embankments
on soft soils. Many authors have reported about research and
cases studies, as e.g. Rowe and Li (2005)
.
This will be the most
economic solution, if there are no restrictions regarding the
settlements, the horizontal “spreading”, the time for
consolidation etc.
Wager and Holtz (1976) used in the 1960’s very short sheet
pile walls connected via tie-rods to capture spreading forces of
embankment on soft soils. The tie-rods and sheet pile walls
acted like a basal reinforcement mechanism and were just
placed at the base of the embankment, not being embedded into
the soft ground. It is reported that several projects applied this
method. This solution was not followed further when geotextile
reinforcements became readily available, mainly for cost
sed deformation. Design approaches have not been
me
ion or failure while
ear
sign approaches or system
dep
ribed foundation system for
embankments on soft soils.
3. RESEARCH STRATEGY
orithm for serviceability
and ultimate limit state of the system.
3.2
reasons.
Harata et al. (2008) reported about the use of sheet pile
walls at the toe of embankments on soft soils to cut off the
settlement depression. Due to the installation of the sheet pile
walls into the ground a stress discontinuity between the
embankment and the surrounding ground is generated, which
leads to a reduction of the vertical deformation outside the
embankment. In the design concept of Harata et al. only the
equilibrium of the vertical forces is considered. Ochiai et al.
(1991) studied in small scale laboratory tests different
arrangements of two parallel sheet pile walls at the toes of the
embankment, where the wall length and inclination were varied.
Additionally in two of the tests the influence of a connection via
tie-rods between the walls has been investigated. As a result of
the tests the authors rated the different arrangements in respect
of the deformation outside the embankment. The use of tie-rods
led to decrea
ntioned.
Adalier et al. (2003), Elgamal et al. (2002) and Tanaka et al.
(2000) reported about the use of tie-rod connected sheet pile
walls beneath embankments on loose, saturated sandy
foundation soils to prevent earthquake-induced liquefaction.
Adalier et al. (2003), analyzed the behaviour with centrifuge
tests and Elgamal et al. (2002) performed numerical simulations
based on these results. Tanaka et al. (2000) performed shaking
table tests and numerical simulations. All researchers confirmed
the benefit of tie-rod connected sheet pile walls beneath the
embankment with respect to deformat
thquake-induced liquefaction occurs.
In both applications only single tie-rods are used, thus the
embankment weight has to be carried only by the subsoil. A
restraining tensile force as with the membrane foundation
system is not generated by the embankment weight. Long time
consolidation processes are not relevant in the case of the
liquefaction issue and of little relevance where a stress
discontinuity is of interest. De
endencies are not addressed.
Cofferdams do have a similar set-up but they are mainly
constructed above the existing ground level. The infill is a well
draining granular material, which provides the stability of the
system. Cofferdams are mostly loaded horizontally from one
side, so the construction sequence as well as the interaction
between the structural elements and soil are completely
different to above desc
3.1. Aim of the research
The aim of this research project is to demonstrate the
applicability of the system, the self-regulation mechanism and
to develop an analytical calculation alg
. Theoretical system behaviour
The stress and strain of the different system components,
vertical walls, tension membrane and soft soil, are strongly
influenced by their interaction. Due to consolidation processes
in the soft soil the interactions are time dependent. The stiffness
of the soil as well as the total stress on the walls are changing
with the consolidation from undrained conditions at the
beginning of the embankment construction to drained
conditions in the final state. The system behaviour depends on
