Stochastic Finite Element Model Updating Of Bridges By Using The Meta-model

Although massive monitoring data could be obtained as the running of structuralhealth monitoring system of bridges, it is still a huge challenge to build the baselinefinite element model (FEM) of bridges by using these data. The reasons could beattributed to two aspects. On one hand, the dynamic characteristics of bridges alwaysvary with the change of environmental factors. On the other hand, there exists noise thatinevitably interference the modal parameter identification, leading to the uncertainty ofthis process. Therefore, to solve the issue of building the accurate baseline FEM ofbridges by using the massive monitoring data, an algorithm about stochastic FEMupdating technique of bridge structure based on Meta-model is proposed in this paper,and the contents are outlined below.Firstly, a comparative study of two kinds of Meta-model (Kriging Methodologyand Response Surface Methodology) was investigated so as to provide the principlesand recommendations of Meta-model applying to update the FEM of bridges.Secondly, stochastic FEM updating technique based on Meta-model was proposed.The Meta-model was applied to obtain the deterministic function relationship betweenmodal parameters and updating parameters, and the estimated mean value andcovariance of updating parameters were implemented to acquire the statisticalcharacteristics of monitoring modal parameters of bridges.Finally, the Meta-model of Dongying Yellow River Bridge was completed. Theproposed method was utilized to build the baseline FEM of this bridge.