Civil Engineering Reference
In-Depth Information
examples showing how other researchers modeled the steel bridges and the
tal and numerical investigation on high-performance steel I-shaped bridge
girders. The investigations were assessed against the American bridge spec-
ification (AASHTO) provisions for cross-sectional compactness and ade-
quate bracing. The study showed that the specifications may be
inadequate owing to intense interactions between local and global buckling
modes in the high-performance steel I-shaped bridge girders. An alternate
bracing requirement was proposed by the authors for use with high-
performance steel bridge girders. The proposed bracing scenario did not
require any additional costs in fabrication or materials. The authors used
ABAQUS [1.29] in the numerical investigations, which considered both
geometric and material nonlinearities using the modified Riks method.
The authors performed an eigenvalue buckling to predict the first buckling
mode, which was factored by an assumed maximum initial displacement of
the girder (L/1000). The models of the bridge girders considered in this
study are constructed from a dense mesh of four-node nonlinear shell finite
elements (S4R). The loading in the finite element modeling was imposed as
a concentrated load at the midspan of a simply supported (SS) I-shaped girder
assembly. The concentrated force simulates the pier reaction of the investi-
gated bridge and the simple supports are placed at the approximate points of
inflection on either side of the pier. In the finite element models, an addi-
tional length of girder was present beyond the support locations to help sim-
ulate the torsional-warping restraint provided by the adjacent beam
segments in the actual bridge. The length of the additional beam segments
was chosen to be 7.625 m, which corresponds to the distance to the next
diaphragm member occurring after the point of inflection, as measured
along the longitudinal axis of the girder. The top compression flange had
a width of 406 mm and a thickness of 45 mm thickness. The bottom tension
flange had the same width but with a modified thickness of 84 mm. A uni-
axial representation of the A709 steel grade was used.
ical study on the ultimate load behavior of plate girders curved in plan. The
investigated girders were medium-sized girders built using rolled steel plates
and were tested to failure. The girders were supported at the ends and sub-
jected to a concentrated load applied at the midspan. The behavior of web
panels was closely studied in order to investigate the tension-field action.
The numerical investigation employed the elastoplastic finite element
method and the results were compared with that measured experimentally.
Search WWH ::
Custom Search