Assessment of predictions of the response of the Hualien containment model during forced vibration tests

Abstract

Blind predictions for the response of the 1/4-scale reinforced concrete Hualien (Taiwan) containment model during forced vibration tests are compared with the observed data. The predictions obtained by the CLASSI approach reflect the experimental conditions prior to and after backfill of the soil surrounding the embedded foundation. The experimental data show a strong and unexpected coupling between the response in the NS and EW directions which is not present in the results for the axisymmetric theoretical models. Also, significant differences can be seen between the experimental responses in the two orthogonal horizontal directions which minimize cross-axis coupling. Although these differences are not accounted for in the theoretical models, the discrepancies between predictions and observations are within the uncertainty of the structural and geotechnical data. The obtained differences between predictions and observations give an excellent measure of the prediction error that can be expected in this type of analysis from uncertainty in the data. A detailed assessment of the initial structural and geotechnical data based on extensive comparisons with the results of previous identification studies is also presented. Finally, comparisons between the observed response and calculations based on revised models for the structure and the soil show that current methods of analysis can account accurately for the observed response.