Civil Engineering Reference
In-Depth Information
provided transient nodal temperatures with respect to time. In the second
phase, a structural analysis was performed, and the nodal temperatures were
read from the thermal analysis. For the thermal analysis, DC3D8 was
employed, which is a 3-D eight-noded linear heat transfer brick element
with one degree of freedom per node. For the structural analysis, element
C3D8 was used, which is a 3-D eight-noded solid continuum element with
three degrees of freedom per node that is compatible with DC3D8 element.
Finite element analyses included geometric and material nonlinearities.
Since there is no structural connection between the concrete slab and the
girder (the bridge was not a composite bridge), the slab was included in
the thermal phase of the analysis, but then was deactivated in the structural
analysis. In this manner, only the thermal impact of the slab was considered.
A finer mesh was used near the supports and the stiffeners because these are
areas of high stress and more susceptible to local buckling. The finite element
model had 533 nodes and 6560 solid elements. The efficiency of the mesh
and finite element model was tested by checking that the difference between
the stresses and deflections due to dead loads at ambient temperature given
by the beam theory and the FE model was negligible and by checking that an
increase of the number of elements in the areas where the mesh was coarser
did not have any significant influence in the thermal and structural results.
Appropriate boundary conditions were used at the midspan section of the
bridge to consider that only half of its structure was modeled. Specifically,
midspan section had free vertical displacement, but it was restrained from
rotating and from translating on the longitudinal axis. In addition, a vertical
support was provided along the surface of the bottom flange beneath the
stiffener. Finally, and only for the “fix” analyses, a rigid solid block was cre-
ated at a distance from the outer cross section of the bridge equal to the width
of the expansion joint. This rigid solid block simulated the existence of an
adjacent span or abutment, and its goal was to ensure that axial expansion of
the nodes of the outer cross section of the bridge was restrained once their
horizontal displacement equaled the width of the expansion joint.
Shifferaw and Fanous [1.51] investigated fatigue crack formation in the
web gap region of multigirder steel bridges. The authors have shown that the
region has been a common occurrence of fatigue crack formation due to
differential deflections between girders resulting in diaphragm forces that
subject the web gap to out-of-plane distortion. The study investigated
the behavior of web gap distortion of a skewed multigirder steel bridge
through field testing and finite element analyses. The study also investigated
different retrofit methods that include the provision of a connection plate
Search WWH ::
Custom Search