Analytical Study of NEESR-SG 4-Span Bridge Model Using OpenSees

Report No. CCEER-07-03This study focuses on the development of a 4-span bridge analytical model using OpenSees. Three ¼ -scale 4-span bridge models are included for testing on the University of Nevada, Reno (UNR) shake tables as a part of a collaborative Network for Earthquake Engineering Simulation Research-Small Group (NEESR-SG) study entitled Seismic Performance of Bridge Systems with Conventional and Innovative Design. The columns of the first bridge model include conventional steel and concrete, while in the other two models, innovative materials will be incorporated in the bridge columns. This pre-test analytical study is intended to predict the performance of the first bridge model during shake table testing and develop motions to be used at the abutments during shake table testing. The bridge piers are of drop-cap type with a continuous post-tensioned deck connected to three, two-column bent caps and roller type connections at the abutments. The bridge model will be tested under bilateral horizontal motions applied by the UNR shake tables at the bridge pier footings and longitudinal motions applied at the abutments through actuators to simulate the abutment interaction with the bridge. To estimate the bridge-abutment interaction and its effect on the bridge response, three different models are developed. The first model (Model 1) represents a bridge with no abutment interaction. In this model the bridge deck ends are supported on rollers so the bridge deck is free to move in both horizontal directions. The abutment interaction is included in the second bridge model (Model 2). In this model, the bridge deck ends are assumed to be supported on the roller. The abutment consists of the backwall and the backfill soil behind it. The soil stiffness is represented by a nonlinear spring. An initial gap of 0.5 in (12.7 mm) is assumed between the deck and the backwall. The third bridge model (Model 3) is developed to represent the actual bridge test setup with the abutment springs replaced by the actuators at the ends. The displacement histories recorded at the bridge deck end nodes in Model 2 are used as the actuator input at the bridge deck end nodes in Model 3. This report presents the modeling assumptions and the predicted results. To evaluate the adequacy of the analytical modeling method, an analytical model of a two-span bridge for which test data were available was used. Results were found to be reasonably close. In addition to the response of the two-span bridge, this report summarizes important calculated response parameters for the 4-span bridge