Incremental dynamic analysis (IDA) applied to seismic risk assessment of bridges - Seismic Risk Analysis and Management of Civil Infrastructure Systems - page 575

Table 21.3 Deformations at damage states for the bridge columns
with D = 2.0 m and H = 7 m

Curvature
ductility

Displacement
ductility Damage state Drift

Yielding 0.54% 1 1

Serviceability 1.05% 4.1 2.0

Bar buckling 4.33% 23.1 8.1

inelastic response during a major event, the bridge superstructure is designed
to remain elastic and is typically modelled using elastic beam elements.
Rigid elements were defi ned to model the superstructure depth, as shown
in Fig. 21.7. The heights of these elements were considered as half of the
superstructure depth. The drift and ductility capacities at different damage
states are presented in Table 21.3 for the bridge columns studied. Therefore
to defi ne the backbone curve shown in Fig. 21.2, the drift ratios at yielding,
beginning of strength degradation and collapse (i.e., zero strength) were
taken as 0.56%, 4.33% and 14.33%, respectively.

21.7.3 Abutment modelling

The simplifi ed abutment model developed by Aviram et al. (2008a) was
used to model the infl uence of the abutments on the seismic response of
the bridges in the longitudinal direction. The longitudinal response in the
simplifi ed model is a function of the system response including the gap, the
abutment backwall and the soil backfi ll material. In the simplifi ed model
the effects of the bearing pads on the responses are ignored. After gap
closure, the superstructure bears directly on the abutment backwall and
mobilizes the full passive backfi ll pressure. Two rigid elements were con-
nected to the abutment ends to model the superstructure width as recom-
mended by Aviram et al. (2008a). More details of modelling the abutments
and the infl uence of the abutments on the longitudinal response are avail-
able in Tehrani and Mitchell (2012c).

21.7.4 Record selection criteria and spectral shape effects
For crustal earthquakes, the 'basic far-fi eld' records selected by Haselton
and Deierlein (2007) were used in this study. This fi xed record set includes
39 crustal records from the PEER-NGA database. Some minimum limits
on event magnitude, PGA and PGV were imposed in the record selection
to make sure that the selected records are strong enough to cause structural