Abstract:
Based on observed damage patterns from previous earthquakes and a rich history of analytical studies, asynchronous input motion has been identified as a major source of unfavourable response for long span structures, such as bridges. Frequently long span bridges provide deep valley crossings, which require special consideration due to the possibility of local amplification of the ground motion as a consequence of topographical irregularities and local soil conditions. This does in fact cause locally enhanced seismic input with the possibility for the bridge piers to respond asynchronously. This introduces special design requirements so that possible out of phase ground displacements and the associated large relative displacements of adjacent piers can be accommodated without excessive damage. Assessment of the local variability of the ground motion due to local lateral heterogeneities and to attenuation properties is thus crucial toward the realistic definition of the asynchronous motion at the base of the bridge piers.
Normally the bridges are designed with synchronous earthquake (same earthquake time-history input to all piers). But this assumption is not fully correct when a bridge has more than one pier and the spans are not very small. When the earthquake reaches at the first pier, then at the same time there is no earthquake at other piers. The earthquake reaches at other piers after some time lag from first pier. So the bridges are required to be designed according to asynchronous earthquake data.
The main objective of the present work is to study the response of box girder bridges due to time differences in dynamic motions at different piers and to study the response of box girder bridges due to variations of soil conditions at different piers. For this, a detailed finite element model of a straight concrete box girder bridge consisting of long spans (695.625 metre length for first bridge and 845.625 metre length for second bridge) has been developed and has been subjected to synchronous earthquake data and asynchronous earthquake data of 0.01 second time lag, asynchronous earthquake data 0.10 second time lag, asynchronous earthquake data 0.10 second time lag with 45 degree angle, asynchronous earthquake data 0.10 second time lag with 90 degree angle, asynchronous earthquake data 0.5 second time lag and asynchronous earthquake generated for different types of soil layers beneath the piers.
Analyses have been carried out to determine displacement variation due to synchronous and asynchronous motions with time lags of 0.01 second, 0.10 second and 0.5 second. Z-0 degree and 45-degree incident angles were considered for the case of 0.1 second time lag motion. In addition, the effect of different sub soil conditions at two portions of the bridge was also investigated.
For this box girder bridge, the finite element analysis shows that there is significant effect on the response of the bridge in case of 0.01 second or longer time delays. Similar effects can be observed due to asynchronous motion caused by different soil profiles at different portions of the bridge. Thus, asynchronous earthquake input should be considered for bridge design for earthquakes with long epicentral distance.
