Statistical approach of performance-based uncertainty quantification of prestressed concrete containment structures for internal pressure capacity

Abstract

With several experiences on economic losses and environmental impacts caused by serious accidents, the ultimate internal pressure and failure modes of Prestressed Concrete Containment Structures (PCCSs) during severe accidents have gained greater attention. Regulatory Guide allows the use of a 3D Finite Element (FE) model to determine the containment internal pressure capacity, because the experiments of PCCSs are a very cost and time-consuming process. In this study, an FE model of PCCS tested at Sandia National Laboratories is developed to identify an effect of parameters in the Concrete Damage Plasticity (CDP) model on the response of the PCCS and is validated from the experimental data. In addition, a closed-form equation is developed to predict the responses of the PCCS through a regression analysis in conjunction with the validated FE model and performance-based uncertainty quantification of the PCCS is estimated by numerical analysis and the closed-form equation. The developed closed-form equation achieves high accuracy of 98% and the probabilistic characteristic of the PCCS obtained from the equation well represents the one obtained from the validated FE model. The closed-form equation rarely requires computational cost to investigate the probabilistic responses of the PCCS.