基于GPS/BDS桥梁短期变形预警方法研究  

Research on Early Warning Methods of Short Term Deformation Based on GPS/BDS Bridge

作　　者：陈浩[1] 岳东杰[1] 陈健[1] 杨国庆[1] Chen Hao;Yue Dongjie;Chen Jian;Yang Guoqing(School of Earth Science and Engineering,Hohai University,Nanjing 211100,China)

机构地区：[1]河海大学地球科学与工程学院,江苏南京211100

出　　处：《甘肃科学学报》2018年第5期36-41,共6页Journal of Gansu Sciences

基　　金：江苏省产学研前瞻性联合研究项目(BY2015002-04)

摘　　要：针对环境因素对桥面变形的影响,桥梁动态位移的时间序列表现出高度不确定性的问题,提出了基于主成分和多变量时间序列模型耦合的桥梁位移预报模型,将该模型应用于苏通大桥的实测数据。试验结果表明:采用主成分分析的方法,提取3个对桥面变形有影响的主成分,3个主成分的累计贡献值达95.088%。建立主成分-时间序列模型,拟合精度优于多变量时间序列法,监测点dx,dy,dH 3个方向的均方根误差分别为2mm、2mm、8mm,且采用均值控制图可以识别出由于结构损伤所引起的位移10mm的异常变化。因此,该模型具有一定的有效性和适用性。In view of the effect of environmental factors on bridge deck deformation, the time series of bridge dynamic displacement shows a high uncertainty. This paper proposes a bridge displacement prediction model based on the coupling of principal component and multivariable time series model, and such model is applied to the measured data of Sutong Bridge. The final test shows that the principal component analysis is used to extract 3 principal components affecting the bridge deck deformation, and the cumulative contribution value of the 3 principal components is 95.088%.The principal component-time series model is established, and the fitting precision is better than that of the multivariable time series method. The root-mean-square error of the three directions of the monitoring points dx, dy and dH are 2 mm, 2 mm , 8 mm, respectively. The abnormal change of 10 mm displacement caused by the structural damage can be identified by using the mean control chart. Therefore, such model has a certain effectiveness and applicability.

关 键 词：主成分分析 多变量时间序列模型 环境因素 桥梁监测 短期预警 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] U446[自动化与计算机技术—控制科学与工程]