基于毫米波雷达的桥梁动挠度监测数据分离技术研究  

Research on Data Separation Technology for Bridge Dynamic Deflection Monitoring based on Millimeter-Wave Radar

作　　者：徐文城 王淼[2,3] XU Wen-cheng;WANG Miao(School of Civil Engineering,Southwest Jiaotong University,Chengdu 610031,China;CCCC Highway Consultants Co.Ltd.,Beijing 100010,China;CCCC Engineering Big Data Information Technology(Beijing)Co.Ltd.,Beijing 100010,China)

机构地区：[1]西南交通大学土木工程学院,成都市610031 [2]中交公路规划设计院有限公司,北京市100010 [3]中交公规土木大数据信息技术(北京)有限公司,北京市100010

出　　处：《公路》2024年第11期144-155,共12页Highway

摘　　要：在全寿命周期内,桥梁受温度、车辆荷载等多种作用的耦合,主梁产生挠度变化。基于毫米波雷达的健康监测系统,可精准获取桥梁在多种作用耦合下的动挠度数据,这为获取不同因素影响下产生的挠度数据分离提供了基础。基于某枢纽动荷载试验过程中采用毫米波雷达所获取主梁挠度数据,采用小波分析及局部加权回归方法,开展温致长期效应及车载效应下挠度数据的分离工作;通过与试验过程中获取的校验数据开展相关性分析,进行有效性验证。结果表明,小波分析及局部加权回归方法在分离长期效应及车辆荷载所引起挠度数据上具有良好的适用性。局部加权回归方法由于其可以更好适应数据的局部及整体变化,在桥梁动挠度分离中可取得更好效果。During the life cycle of bridges,the main beams undergo deflection changes due to the coupling of multiple effects such as temperature and vehicle loads.Based on the millimeter-wave radar health monitoring system,the dynamic deflection data of bridges under multiple effects can be accurately obtained,which provides a foundation for separating the deflection data under different effects.Based on the main beam deflection data obtained by millimeter-wave radar during a dynamic load test of a hub,wavelet analysis and locally weighted regression method are utilized to separate the deflection data under temperature-induced long-term effect and vehicle load effect.Calibration data are obtained during dynamic load test,and correlation analysis is carried out to verify the validity.The results indicate the practicality of wavelet analysis and locally weighted regression method in separating the deflection data caused by long-term effect and vehicle load.Due to better adaptability to local and overall changes in data,the locally weighted regression method can achieve better results in dynamic deflection separation for bridges.

关 键 词：桥梁监测 动挠度数据 毫米波雷达 动荷载试验 数据分离 

分 类 号：U446.2[建筑科学—桥梁与隧道工程]