Centrifuge Modeling of Seismic Performance of Reinforced Earth Structures - Reinforced Soil Engineering: Advances in Research and Practice - page 407

tests on the vertical retaining walls reinforced with metal strips and discussed the
limit states of this type of wall and applicability of various analyses to evaluate
the collapse limit state using observed failure types (friction failure and tension
failure) and other data such as vertical stresses on the base of reinforced portion
and tensions of the strips. Shen et al. (1982) used the centrifuge model tests to
confirm the shape of the failure plane assumed in a stability analysis on the
cut reinforced with soil nailing. These two centrifuge model tests are good
examples showing the high potential of centrifuge modeling; namely, it can
provide very useful information about the behavior of soil structures, for
which very limited information is available under the conditions assumed in
some design methods.
Earth reinforcement technology has a relatively long history compared
with the time since centrifuge modeling has been applied to this type of structure.
However, many aspects still exist for many for which the centrifuge can
contribute to earth reinforcement technology. The performance of reinforced
soils during an earthquake is one of the typical examples about which very few
reliable field records are available in the literature. Although intensive field
observations conducted after previous large earthquakes have provided useful
data on the seismic performance of various structures, the data are normally
limited in the final figures after the earthquakes, not including responses of
structures during the earthquakes and accurate input ground motions. From the
study on the damage of soil retaining walls for railway embankments after the
1995 Hyogoken-Nanbu earthquake, Tatsuoka et al. (1996) reported that geogrid
reinforced soil retaining walls performed very well even in one of the most
severely shaken areas, while gravity-type retaining walls showed a low stability
against strong seismic motion. Nishimura et al. (1996) also concluded from the
investigation of the geogrid reinforced soil walls after the 1995 Hyogoken-Nanbu
earthquake that the geogrid reinforced soil wall could have a much higher seismic
resistance than that predicted by two methods based on the pseudo-static
limit equilibrium approach. Tatsuoka et al. (1996) concluded in their report
that the performance of the geogrid reinforced soil wall observed in their
investigation would foster confirmation and development of aseismic design
procedures. In order to develop more rational design procedures and confirm
the applicability, observing the behavior of the reinforced soils under well-
controlled or recorded shaking motion, which can be done by physical modeling,
is most important.
The following sections outline advantages of centrifuge modeling in the
study of reinforced earth structures especially for the seismic performance, as
well as some limitations to this technique. As an example of the application,
centrifuge model tests on the vertical geogrid reinforced wall are described, and
results and discussions on the tests are presented.