868
Proceedings of the 18
th
International Conference on Soil Mechanics and Geotechnical Engineering, Paris 2013
distorted layers of coloured sand, or time-lapse exposures that
reveal particle trajectories. Scaling laws are never referred to,
aside from commentary by Meyerhof on the effect of stress
level on friction angle, and the resulting scale effect between the
bearing capacity of small models and field scale shallow
foundations.
Over the past 50 years, physical modelling technology has
evolved through to two key developments. The development of
the centrifuge in the 1970s allowed the realism of physical
modelling to be enhanced, through the correct modelling of self-
weight stresses. The subsequent development of miniaturised
electronics and micro-computers has led to enhanced methods
of data acquisition, control, and image analysis. The refinement
of these techniques continues to yield dramatic improvements in
the utility of physical modelling. More realistic simulations can
be conducted, and more detailed observations can be gathered.
Reviews of recent technological developments within
physical modelling are described by Mayne et al. (2009) and
White (2008).
Meanwhile, continuing cross-disciplinary efforts have led to
wider recognition of definitive scaling laws to allow small scale
physical models to be related to field scale conditions. TC104
has overseen the cataloguing of scaling law research. The initial
publication of a TC2 (as TC104 was previously known) Scaling
Law Catalogue (Garnier et al. 2007) has been followed by
continuing development of this resource.
Proper application of scaling laws is vital when interpreting
physical model tests for purposes (2) and (3) given on the
previous page. If physical model test data is to be correctly
linked to a field scale prototype, or to a numerical simulation,
correct account must be made of the influences of size and
timescale effects.
Scaling laws are well-established and straightforward to
adhere to for small scale modelling of many geotechnical
problems, particularly in fine-grained saturated soils. It is
therefore no coincidence that physical modelling is a well-
accepted technology in both research and design practice in
offshore geotechnics, where soft normally- or lightly over-
consolidated soils predominate. Recent state-of-the-art review
papers have summarised many such applications (Martin 2001,
White 2008, Gaudin et al. 2010).
For purpose (1), listed above, similitude and scaling laws are
less significant, since the physical model might be an abstract
component of the full system, or precise scaling of particular
conditions may not be important. In this categorisation, we are
referring to studies that aim to uncover a building block of the
geotechnical system behaviour. Perhaps the overall system is
not being modelled, or perhaps there is established theory to be
tested. In the words of the title of the first Schofield Lecture, the
aim might simply be to
“expect the unexpected”
(Bolton 2013).
It is this exploratory nature that makes physical modelling an
attractive tool for many researchers and educators in
geotechnics. By observing geotechnical systems in action, an
intuitive understanding of soil mechanics can be gained,
complementing the study of theory. Physical modelling tools for
geotechnical education range from the venerable Hele-Shaw
cell (Hele-Shaw 1898) to modern miniature – but highly
instrumented – geotechnical centrifuges.
The Hele-Shaw cell provides solutions to seepage flow
according to Darcy’s Law as a consequence of the Navier-
Stokes equations simplifying for narrow planar flow. Henry
Darcy studied in Paris – the venue of this 18
th
ICSMGE – at the
Ecole Polytechnique and then Ecole des Ponts et Chausées,
before joining the Corps des Ponts et Chausées. His most
famous public work was the water supply system for Dijon.
Appendix D of his published account of the Dijon works
contains his report on the classic physical model tests in which
the linear relationship between head gradient and flow velocity
was identified. This work was published in 1856 and two years
later Darcy died of pneumonia here in Paris.
Undergraduates studying geotechnics and fluid mechanics
have faced Hele-Shaw cells for almost a century. Falling head
permeability tests of the form analysed by Darcy in his classic
work are standard undergraduate geotechnical laboratory
experiments. Some of today’s undergraduates also have access
to more sophisticated apparatus. These include miniature
experiments that provide detailed measurements and
observations of geotechnical constructions – often via bench-top
centrifuges. These facilities provide opportunities to apply the
analysis tools taught in lectures, completing the learning cycle
through practical experience (Wartman 2006, Kolb 1984).
Meanwhile, to inspire the next generation of geotechnical
engineering students, physical models are the most commonly
called-upon facility within university departments to enliven
events for school students. Physical models show engineering
systems in action in a way that is immediately comprehensible.
Figure 1. Darcy’s physical model for investigating flow in porous media
(Darcy 1856)
The following review summarises and discusses many of the
papers contributed to the 18th ICSMGE that fall under the
TC104 session. The contributed papers have been divided into
the physical modelling categories listed at the start of this
report.
2 INSIGHTS INTO SOIL-STRUCTURE INTERACTION
AND GEOTECHNICAL BEHAVIOUR
2.1
Reinforced ground
Several papers have focused on the performance of ground
improvement systems utilising reinforced columns – either of
cemented material, sand or stone. These systems are difficult to
analyse, due to the complexity of the construction process and
