Refined 3D finite element modeling of partially restrained connections including slip

The effect of partially-restrained (PR) connections on the behavior of steel frames
and their potential economical benefits is well recognized [18]. However, many structural
analysis and design approaches still consider connections as either fixed or
pinned. This assumption is mainly due to convenience and the lack of common
analysis and design approaches that address PR connections. Despite many full-scale
experimental studies that have been conducted to date, there is still a need for a
better understanding of the mechanisms that effect the non-linear behavior of PR
connections [8].
Non-linear moment rotation response of connections was recognized in the early
1930s. Standardized functions have been developed starting from basic linear and
bilinear approximations to more sophisticated models based on polynomials, cubic
B-splines and power functions fitted to available experimental data. Sherbourne and
Bahaari [13] have recently presented a review of these functions. Frye and Morris
[9] were among the first to incorporate these standardized moment–rotation functions
in steel plane frame analysis to investigate the effect of the connections on the
frame behavior.
Moment rotation functions can be useful for designers in practice. These usually
include small number of parameters taken into account from limited test data. The
lack of a large and parametrized experimental database does not allow for generating
standardized functions. Thus, there is a need to be able to analytically generate a
reliable moment–rotation response of PR connections that can be used in analysis
and design.
Non-linear finite elements are an attractive tool for modeling connections. Early
attempts to use finite elements for analysis of PR connections was by Krishnamurthy
[11]. As in many early studies using finite elements, many simplifications are made
due to the limitations of computational power. More recent studies using finite
elements in modeling connections have focused on end plate connections
[6,7,10,13,14]. In these studies 2D and 3D models are used with various simplifications
in the geometry of members, the bolts, and contact conditions. The effect
of friction on the response of end plate connections is usually neglected in these
models [7,10].
Azizinamini [3–5] preformed an extensive and detailed experimental study for top
and bottom seat angle connections with double web angles along with pull tests. In
addition, simplified 3D FE models for the pull tests are also studied. One quarter of
the top angle in the connection is modeled with 3D elements to simulate a pull test.
The force–displacement relation was converted to a moment–rotation relation in
order to examine the role of the top angle on the behavior of the connection and
approximate the overall response of the connection with a pull test. Different assumptions
and simplifications are made in order to avoid detailed modeling and reduce
the computational effort.
Yang et al. [19] consider a double web angle connection where the angles are
bolted to the column flanges and welded to the beam web. The bolts and angle are
modeled using 3D finite elements and wedge elements are used to model the weld
region. Contact is included between the bolt head and angle. However, the contact
between the bolt shank and hole is ignored.
In the studies on end plate connections and the double web angle connection, the
bolts are transferring the loads axially, thus eliminating the need for combined contact
and friction modeling between the bolts and members. These models are therefore
limited to these types of PR connections. The bolted connections tested by
Azizinamini et al. [4,5] are investigated in this study. These top–bottom bolted seat
angle connections transfer the forces by friction by clamping the parts together with
the bolts. Modeling such a mechanism requires the inclusion of contact and slip
between the connection members.
In this study, a refined 3D modeling of PR bolted connections are performed
recognizing contact and friction effects. The modeling approach is general and capable
of modeling various types of geometries of PR connections by using parametric
meshing techniques. Therefore the time of generating detailed 3D geometries is
almost eliminated. A calibration method for the pretension of the bolts is presented.
In this method, parametric solutions are first generated separately for a single bolt
clamping semi-infinite plates. These solutions are used to specify initial pretension
values for the bolts in the full connection. The correct pretension values are then
examined and corrected in the full connection model to achieve accurate final values.
It is shown in this study that the response of the bolted PR connections are sensitive
to the pretension of the bolts, thus correctly modeling the pretension and slip is
important.
Download File