Performance-based probabilistic capacity models for reinforced concrete and prestressed concrete protective structures subjected to missile impact

Abstract

A performance-based design (PBD) framework for nuclear containment structure (NCS) subject to missile impact is developed. These are, in turn, used to estimate probabilistic energy-based capacity models centred on three performance levels associated with four damage states of NCS. The present study considers multi-level PBD as opposed to single-level conventional collapse based design. Bayesian inference is used to evaluate the unknown model parameters based on available numerical data. The data of reinforced concrete (RC) and prestressed concrete (PC) panels subjected to hard missile impact is obtained from finite element (FE) numerical modelling (LS-DYNA). Formulations for local damages such as penetration depth, perforation limit and residual velocity of the missile are also developed. The established models account for the multi-modal response of the structure, the transition of failure modes and their interaction, inherent aleatoric and epistemic uncertainties associated with the modelling, configuration and material properties. A comparative study is performed with these predicted models and experimentation achieves the desired prediction level. These probabilistic models can be used for designing the NCS when subject to ballistic missiles. The work can also be extended to scenarios like ship collisions, vehicles falling on slabs due to tornadoes by following the developed framework. Not alone NCS the obtained results accommodate other protective structures like bunkers, storage tanks, residential slabs, and many similar configurations considered in the study.