|Institution:||University of Cincinnati|
|Department:||Engineering : Electrical Engineering|
|Keywords:||Bridge FE Model; Model Calibration; Linking|
|Full text PDF:||http://rave.ohiolink.edu/etdc/view?acc_num=ucin1092674830|
3D finite element analysis (FEA) is an extremely efficient tool in the bridge condition evaluation process. The process typically consists of generating an Finite Element (FE) model, performing dynamic field testing, calibrating the model to field test data, and using the results from the calibrated model to rate the condition of the bridge or investigate unique loadings or retrofit schemes. This thesis will focus on the calibration strategy for FE bridge models. Model calibration starts from a nominal bridge model. Experimental data which was processed from a bridge field test is used for calibration reference. The nominal bridge was built according to the bridge plan. Many of the differences between experimental and analytic results are due to modeling limitations and experimental error. In the model calibration, during the iterative calibration process, some FE parameters are adjusted to make the analytical bridge static and dynamic responses match the corresponding bridge’s modal and truckload performance acquired from the field tests. An automatic, steepest descent algorithm based calibration strategy, combined with the bisection method, is presented. Two bridge models will be calibrated as examples and calibration results will be shown in each automatic calibration step.