American Nuclear Society

Home / Publications / Journals / Nuclear Technology / Volume 167 / Number 1

Improvement of Core Modeling in ICARE/CATHARE: Application to the Calculation of a Six-Inch-Break LOCA Leading to a Severely Degraded Situation

Patrick Drai, Olivier Marchand, Patrick Chatelard, Florian Fichot, Joëlle Fleurot

Nuclear Technology / Volume 167 / Number 1 / July 2009 / Pages 235-246

Technical Paper / NURETH-12 / Thermal Hydraulics /

In order to analyze the course of a hypothetical severe accident, the French "Institut de Radioprotection et de Sûreté Nucléaire" in the last decade has developed computer codes that have been extensively used for supporting the Level 2 Probabilistic Safety Assessment (PSA2) and, in general, for the safety analysis of French pressurized water reactors (PWRs).

In particular, the computer code ICARE/CATHARE V1 is a tool that has been widely validated and intensively used within the framework of the PSA2 of the 900-MW(electric) French PWR. This code has been tested on many accident scenarios, and the results obtained have been considered to be satisfactory and reliable up to the end of the early degradation phase. But, severe accidents in PWRs are characterized by a continuous evolution of the core geometry due to chemical reactions, melting, and mechanical failure of the rods and other structures. These local variations of the porosity and other parameters lead to multidimensional flows and heat transfers. So, the lack of a multidimensional two-phase thermal-hydraulic model appeared to be prejudicial to achieve best-estimate reactor studies with ICARE/CATHARE V1 in the case of large core blockages and/or in the case of large cavity appearance. In accordance, a full multidimensional modeling (covering both the fluid flow and the corium behavior) was developed and introduced in a new ICARE/CATHARE version referenced as V2, which includes two options for the thermal-hydraulic modeling: either one-dimensional (1D) or two-dimensional (2D).

The first part of this paper demonstrates that without activating the new V2 models, ICARE/CATHARE V2(1D) is able to reproduce the results obtained with ICARE/CATHARE V1 on the basis of a 6-in.-break loss-of-coolant accident. Then, in order to illustrate some of the new V2 modeling improvements, the last part is focused on the results obtained with ICARE/CATHARE V2(2D), and a preliminary comparison is made with ICARE/CATHARE V2(1D).

This 1D-2D comparison points out in particular the important role that could be played in the course of a severe accident by the multidimensional flow pattern.

Questions or comments about the site? Contact the ANS Webmaster.