中文摘要：

提出了基于有限元模型修正的单车通过多梁式桥梁的移动荷载识别方法。首先采用Butter-worth低通滤波器对现场采集到的24h内所有过桥车辆产生的桥梁动位移信号进行滤波处理,提取静力响应极值,并严格按照车型进行分类统计;其次,对观测桥梁进行基于静力试验的有限元模型修正,建立能够反映桥梁真实状态的基准有限元模型;最后将修正后的有限元模型输入至自行研发的BDANS软件中的多梁式车-桥耦合振动模块,以车型为单位,依据该车型车辆在桥面横向移动时各主梁竖向位移响应分配关系,结合多梁式车-桥耦合振动模块以及实测车辆过桥时各主梁静力极值响应,识别出车辆在桥面行驶的横向位置,然后根据识别出的车辆横向行驶位置和实测桥梁响应识别出车质量。结果表明：该识别方法较为可靠,识别精度较高,能按照车型批量进行识别,可大规模处理交通荷载数据。

英文摘要：

Based on the FE model updating, a new identification method for the moving load on the multi-girder bridge was proposed. Firstly, a Butterworth low pass filter was designed forfiltering dynamic displacement signals of vehicles moving across the bridge in 24 hours, extreme values of static response were extracted, and classification and statistics were performed in termsof vehicle type rigorously. Secondly, the reference finite element model, reflecting the real state of bridge, was developed by updating finite element model based on static test on observedbridge. Finally, the updated finite element model was taken as an input of self-developed multigirder vehicle-bridge coupling vibration program in BDANS. With vehicle type as unit, transversevehicle moving location identification was realized according to vertical displacement assignment relationship when vehicles moving in transverse direction, and with the combination of multi-girder vehicle-bridge coupling vibration program as well as measured static extreme value of each girder when vehicle moving across the bridge. Vehicle weight identification was realized based onidentified transverse locations and measured bridge responses. The results show that the method is of good reliability and accuracy, besides, it can make batch identification in terms of vehicletype, which makes it possible to process data of traffic load on a large scale.