Finite element model validation of bridge based on structural health monitoring——Part Ⅱ:Uncertainty propagation and model validation  

作　　者：Xiaosong Lin Zhouhong Zong Jie Niu 

机构地区：[1]aSchool of Civil Engineering, Fujian University of Technology [2]School of Civil Engineering, Southeast University

出　　处：《Journal of Traffic and Transportation Engineering(English Edition)》2015年第4期279-289,共11页交通运输工程学报（英文版）

基　　金：supported by the National Natural Science Foundation of China(No.51178101,51378112);National Scientific and Technological Supporting Plan(No.2011BAK02B03);Scientific Research and Development Foundation of Fujian University of Technology(No.GY-Z10085)

摘　　要：Because of uncertainties involved in modeling, construction, and measurement systems, the assessment of the FE model validation must be conducted based on stochastic mea- surements to provide designers with confidence for further applications. In this study, based on the updated model using response surface methodology, a practical model vali- dation methodology via uncertainty propagation is presented. Several criteria of testing/ analysis correlation are introduced, and the sources of model and testing uncertainties are also discussed. After that, Monte Carlo stochastic finite element （FE） method is employed to perform the uncertainty quantification and propagation. The proposed methodology is illustrated with the examination of the validity of a large-span prestressed concrete continuous rigid frame bridge monitored under operational conditions. It can be concluded that the calculated frequencies and vibration modes of the updated FE model of Xiabaishi Bridge are consistent with the measured ones. The relative errors of each frequency are all less than 3.7%. Meanwhile, the overlap ratio indexes of each frequency are all more than 75%; The MAC values of each calculated vibration frequency are all more than 90%. The model of Xiabaishi Bridge is valid in the whole operation space including experimental design space, and its confidence level is upper than 95%. The validated FE model of Xia- baishi Bridge can reflect the current condition of Xiabaishi Bridge, and also can be used as basis of bridge health monitoring, damage identification and safety assessment.Because of uncertainties involved in modeling, construction, and measurement systems, the assessment of the FE model validation must be conducted based on stochastic mea- surements to provide designers with confidence for further applications. In this study, based on the updated model using response surface methodology, a practical model vali- dation methodology via uncertainty propagation is presented. Several criteria of testing/ analysis correlation are introduced, and the sources of model and testing uncertainties are also discussed. After that, Monte Carlo stochastic finite element （FE） method is employed to perform the uncertainty quantification and propagation. The proposed methodology is illustrated with the examination of the validity of a large-span prestressed concrete continuous rigid frame bridge monitored under operational conditions. It can be concluded that the calculated frequencies and vibration modes of the updated FE model of Xiabaishi Bridge are consistent with the measured ones. The relative errors of each frequency are all less than 3.7%. Meanwhile, the overlap ratio indexes of each frequency are all more than 75%; The MAC values of each calculated vibration frequency are all more than 90%. The model of Xiabaishi Bridge is valid in the whole operation space including experimental design space, and its confidence level is upper than 95%. The validated FE model of Xia- baishi Bridge can reflect the current condition of Xiabaishi Bridge, and also can be used as basis of bridge health monitoring, damage identification and safety assessment.

关 键 词：Bridge engineeringModel validationCorrelation of testing/analysisUncertainty quantification andpropagationEvaluation of model precision 

分 类 号：U446[建筑科学—桥梁与隧道工程]