Copula-based cloud analysis for seismic fragility and its application to nuclear power plant structures

Abstract

This paper presents a copula-based cloud analysis method for seismic fragility assessment with practical applications for nuclear power plant structures. Copula theory enables the flexible and efficient modeling of joint probability density function (PDF) for critical parameters, specifically the intensity measures (IM) and the engineering demand parameters (EDP). It decomposes the joint PDF into the product of marginal PDFs and copula density function, which are then separately fitted to the data. The study showcases the versatility of parameterized joint distribution models in estimating optimal distribution parameters derived from structural seismic analysis data. By utilizing copula functions, the method calculates conditional EDP distributions given IM, facilitating the construction of seismic fragility curves. The research compares various copula models’ suitability for fitting fragility curves, employing methods like Akaike information criterion (AIC) for model selection. The study provides a comprehensive framework for seismic fragility assessment, from random sampling of structural parameters and seismic data to nonlinear finite element modeling and copula-based parametric cloud analysis. The application to prestressed concrete containment vessel (PCCV) structures exemplifies the method’s adaptability. This innovative approach not only highlights the flexibility of copula-based analysis in assessing structural performance under different dynamic scenarios but also offers invaluable insights for performance-based earthquake engineering (PBEE) and safety assessments.