STATISTICAL ANALYSIS OF AGR GRAPHITE CORE REACTOR - COUPLING BETWEEN A STATISTICAL ANALYSIS TOOL (OPENTURNS) AND A NONLINEAR DYNAMICS MODEL (GCOEUR)

摘要

Risk assessment in Advanced Gas-cooled Reactor (AGR) core in case of seismic loadings is of importance regarding AGR plants life extension. A statistical analysis methodology is conceived based on the coupling of a statistical analysis tool and a mechanical model taking into account the inner and physical state of the core, made out of an array of different kinds of graphite bricks. The model developed by Atkins in LS-Dyna (GCore) and implemented in Code_Aster (GCoeur) is a 3-dimensional nonlinear finite element computational dynamics model. The aim of this model is to assess the seismic response of AGR reactor core by taking into account the interactions between each brick through the use of discrete elements with impact behaviour. The purpose of this paper is to present this project, and more specifically the numerical model along with a first methodology for the statistical analysis of a reduced core seismic behaviour. This includes identifying the necessary input parameters to provide to OpenTurns and the output parameters needed by OpenTurns to determine the failure probability coupled then to a metamodel based on polynomial chaos to emulate the response.