Geoscience Reference
In-Depth Information
sort and classify them in order to understand the links between them and, only then,
to model their influence on the projected structure. In keeping with his approach,
this chapter will analyze the seismic behavior of dams through an initial appraisal of
observed accidents and assessed failures before launching any modeling.
Jean Biarez was among the first soil mechanics in France to focus on the specific
non-linearities of the ground. Full of admiration for the pioneering work of Duncan
and Chang, he promoted the development and use of constitutive laws to fit the
observed performance of earthen structures. He showed that they were essential to
predict the movements of dams. With the collaboration of Al Issa, he showed the
way to model the construction and first filling of the Mont-Cenis dam (1968). He
never lost sight of the main pedagogical and practical purpose, however, which was
to identify, among all the possible reactions of the soil, the key phenomenon that
governed the behavior of the structure under the studied loading.
Therefore, over the years, his disappointment with the inflation of increasingly
complex numerical models became such that, on behalf of the practitioner, he
demanded simpler models that were less ambitious but more dependable. Towards
the end of his career, in order to help the “numerical analysts” simplify their models,
he combined the acquired experimental soils to extract their “key behaviors”, a set
of behaviors unique to granular or cohesive soils (“soils with glue”, as he liked to
say).
Biarez never ceased to repeat that any numerical approach had to be validated by
its ability to reproduce the “collapse” of any earthen structure, the ultimate phase of
the behavior studied. The underlying idea was to find the main phenomenon, which
triggered the failure of dams under seismic loading, and to have it implemented into
the analysis. In accordance with this line of thought, a seismic modeling of an
embankment dam is described in these pages by a deliberately simplified approach.
Again, the basic idea is to provide the engineer with a rapid method with which to
reinforce his/ her judgment on the dam performance, allowing him/her to improve
his/her design of such works as quickly as possible. The most critical phenomenon
leading to failure under seismic hazard is liquefaction. A coupled analysis will be
presented in the last part of this chapter, illustrated by an application to
embankments and their foundation in gravelly alluviums.
10.1.2.
Definitions
In the field of dam engineering, failure is the word used in accordance with the
practices of the International Commission on Large Dams (ICOLD). It means the
loss of water storage capacity. This definition is adopted in this text. As a corollary,
a sliding with no impact on the level of the reservoir is not considered a failure. In
Search WWH ::
Custom Search