Abstract:
The Horsetail Creek (HC) bridge is an example of an Oregon bridge that
was classified as structurally deficient and was not designed to withstand
earthquake (EQ) excitations. A fiber-reinforced polymer (FRP) rehabilitation was
performed on the HC bridge to increase flexural and shear capacities for traffic
loads. However, a seismic retrofit has not yet been accomplished for this bridge.
Fully three-dimensional finite element (FE) models are developed to
simulate and examine the structural behavior of both full-size reinforced concrete
(RC) beams and the HC bridge using ANSYS. FE analyses are compared with
tests of full-scale beams replicating the transverse beams of the HC bridge before
and after FRP strengthening from linear and nonlinear ranges up to failure. The FE
models can effectively predict the behavior of the beams, and analytical and
experimental results correlate very well.
For the FE analyses of the HC bridge, soil-structure interface modeling is
incorporated to replicate the actual bridge boundary conditions. Truck loadings are
applied to the FE model at different locations, as in the actual bridge test. A
sensitivity study is performed by varying uncertain bridge parameters to develop an
FE bridge model best representing the actual bridge conditions. The optimal FE
model obtained from the sensitivity study can accurately predict the magnitudes
and trends in the strains.
After an optimal FE bridge model is established, a performance evaluation
on the FRP strengthening of the HC bridge is conducted. Both unstrengthened and
FRP-strengthened bridge models are subjected to two different types of loading;
i.e., scaled gravity and scaled truck loads to failure. Comparisons of results show
the improvement in structural performance due to FRP strengthening.
A seismic risk-related investigation of the HC bridge is also carried out.
Nonlinear time-history analyses are performed using EQ acceleration-time histories
applied to the HC bridge model. The ground motions are appropriate to the Pacific
Northwest site and scaled so that the response spectrum, within natural periods of
interest, matches the 1996 AASHTO design response spectrum. Based on the
analytical results, colunm confinement is recommended to increase ductility and
reduce potential for substructure collapse in future seismic events.
