Geotechnical centrifuge modelling - The Mechanics of Soils and Foundations

Civil Engineering Reference

In-Depth Information

behaviour observed. The requirements for control of loading and measurement of

load and displacement in a model are broadly similar to those for laboratory tests

described in Chapter 7.

Communication with the rotating model is through slip rings, as shown in Fig. 27.4.

These may transmit fluids (e.g. water, gas or hydraulic oil) or power to operate motors

or valves, and they will transmit signals from force, pressure and displacement trans-

ducers and from closed circuit television cameras set to observe critical points in the

model.

Before conducting a test the model should be allowed to come into equilibrium under

the increased self-weight stresses at constant centrifuge acceleration; larger models

of fine-grained soils may require the centrifuge to be run continuously for several

days to reach equilibrium. Often a small ground investigation will be carried out in

flight using model cone penetration or shear vane tests similar to those discussed in

Sec. 17.5.

A very large number of different events and construction activities can be modelled.

Design and manufacture of model loading and construction devices taxes the ingenu-

ity of the engineer and a number of sophisticated and novel examples can be found

in the literature of centrifuge modelling. Some typical examples include: vertical and

horizontal loading of foundations, piles and anchors; modelling excavation and tunnel

construction by draining heavy fluids or by reducing pressures; embankment construc-

tion in stages by dropping sand from a hopper; earthquakes simulated by vibrating

the base of the model; formation of craters and blast loading on buried structures

simulated by detonating small explosive charges.

27.7 Summary

1. Modelling geotechnical structures can be used to examine mechanisms of defor-

mation and collapse, to validate numerical analyses and for parametric studies.

Models can occasionally be applied to site-specific cases, but this is usually very

difficult.

2. For correct scaling of stresses and soil properties geotechnical models should be

tested while under acceleration in a centrifuge. An n -scale model should be tested

at an acceleration of ng , where g is the acceleration due to Earth's gravity.

3. At a scale factor of n , rates of loading should be raised by a factor of n and rates

of consolidation will be increased n 2 times.

Reference

Taylor, R. N. (1987) Modelling in ground engineering, Chapter 58 in Geotechnical Engineers

Reference Book, F. G. Bell (ed.) Butterworth, London.