Civil Engineering Reference
In-Depth Information
geometric imperfections can be found in any region of the outer or inner
surfaces of metal structural members and are in the perpendicular directions
to the structural member surfaces. On the other hand, initial overall geomet-
ric imperfections are global profiles for the whole structural member along
the member length in any direction. Initial local and overall geometric
imperfections can be predicted from finite element models by conducting
eigenvalue buckling analysis to obtain the worst cases of local and overall
buckling modes. These local and overall buckling modes can be then fac-
tored by measured magnitudes in the tests. Superposition can be used to pre-
dict final combined local and overall buckling modes. The resulting
combined buckling modes can be then added to the initial coordinates of
the structural member. The final coordinates can be used in any subsequent
nonlinear analysis. The details of the eigenvalue buckling analysis were
highlighted in
Section 5.5.2
. Accurate finite element models should incor-
porate initial local and overall geometric imperfections in the analysis; oth-
erwise, the results will not be accurate. Efficient test programs must include
the measurement of initial local and overall geometric imperfections.
Residual stresses are initial stresses existing in cross sections without the
application of an external load such as stresses resulting from manufacturing
processes of structural steel members. Residual stresses produce internal
membrane forces and bending moments, which are in equilibrium inside
the cross sections. The force and the moment resulting from residual stresses
in the cross sections must be zero. Residual stresses in structural cross sec-
tions are attributed to the uneven cooling of parts of cross sections after
hot rolling. Uneven cooling of cross-section parts subjects to internal stres-
ses. The parts that cool quicker have residual compressive stresses, while
parts that cool lower have residual tensile stresses. Residual stresses cannot
be avoided and in most cases are not desirable. The measurement of residual
stresses is therefore important for accurate understanding of the performance
of metal structural members.
Extensive experimental investigations were conducted in the literature
to determine the distribution and magnitude of residual stresses inside cross
sections. The experimental investigations can be classified into two main
categories, which are nondestructive and destructive methods. Examples
of nondestructive methods are X-ray diffraction and Neutron diffraction.
Nondestructive methods are suitable for measuring stresses close to the out-
side surface of cross sections. On the other hand, destructive methods
involve machining/cutting of the cross section to release internal stresses
and measure resulting change of strains. Destructive methods are based
Search WWH ::
Custom Search