基于动应变识别桥梁移动荷载  

On the Dynamic-Strain-Based Identification of the Moving Loads of a Bridge

作　　者：余佳代 王赞芝[1] 王丹[1] 于世龙[1] 

机构地区：[1]广西科技大学土木建筑工程学院,广西柳州545006

出　　处：《国防交通工程与技术》2017年第5期9-14,共6页Traffic Engineering and Technology for National Defence

基　　金：广西自然科学基金项目(2015GXNSFBA139229)

摘　　要：移动荷载识别可作为桥梁健康监测的研究基础,提出了基于BP神经网络技术识别桥上移动荷载的新方法。基于相似理论,建立车桥耦合试验模型,测试与数值仿真了一简支T梁的动应变数据,两者吻合较好。将测试的应变、速度与荷载数据作为训练样本,采用BP神经网络识别预测结果。结果表明:通过传感器采集桥梁截面的动应变数据,运用BP神经网络技术识别车重方法可靠有效,可以广泛运用在桥梁的荷载识别中,适合工程应用。As the identification of the moving loads of a bridge may be used as the research grounds for monitoring the health of a bridges new BP-neural-network-technology-based method capable of identifying the moving loads of a bridge is put forward in the paper. Based on the similarity theory, an automobile-bridge-coupled test model is established in the paper to test and nu-merically simulate the dynamic strain data of a T-shaped simply-supported girder,and it is found that the two are in quite good agreement with one another. Then, the measured strain,speed and load data are used as the training samples, and the BP-neural network technology is used to identify the forecast results. The results show that the dynamic strain data of the cross-section of the bridge collected with the sensors and the method of identifying the weight of automobiles with the BP neural network technology are both reliable and effective, which shows that the technology may be only widely applied to the identification of the moving loads of bridges but also used for other projects.

关 键 词：桥梁 荷载识别 BP神经网络 动应变 

分 类 号：U441.2[建筑科学—桥梁与隧道工程] U446.2[交通运输工程—道路与铁道工程]