Phenomena-based validation matrix for ex-vessel (containment) models and codes (Containment-CVM)

The Committee on the Safety of Nuclear Installations (CSNI) formed the CCVM (Containment Code Validation Matrix) task group in 2002. The objective of this group was to define a basic set of available experiments for code validation, covering the range of containment (ex-vessel) phenomena expected in the course of light and heavy water reactor design basis accidents and beyond design basis accidents/severe accidents. It considered phenomena relevant to pressurised heavy water reactor (PHWR), pressurised water reactor (PWR) and boiling water reactor (BWR) designs of western origin as well as Vodo-Vodyanoi Energetichesky Reaktor (VVER) types of Eastern European origin. This work complemented the two existing CSNI validation matrices for thermal hydraulic code validation [NEA/CSNI/R(1993)14] and in-vessel core degradation [NEA/CSNI/R(2000)21].

Products

The outcome of this task was the "Containment Code Validation Matrix" [NEA/CSNI/R(2014)3, May 2014]. It was a supplement to the existing reports "Separate Effects Test Matrix for Thermal Hydraulic Code Validation" [OECD/GD(94)82, September 1993) and "In-Vessel Core Degradation Code Validation Matrix" [OECD/GD (96)14, 1996].

The report begins with a brief overview of the main features of PWR, BWR, Canadian deuterium uranium (CANDU) and VVER reactors. It also presents an analysis of the ex-vessel corium retention (core catcher). It then gives a general outline of the accident progression for light water (LWR) and heavy water reactors. The main focus captures most of the phenomena and safety systems employed in these reactor types and to highlight the differences.

The CCVM report contains a description of 127 phenomena, broken down into 6 categories:

• containment thermal-hydraulic phenomena;
• hydrogen behaviour (combustion, mitigation and generation) phenomena;
• aerosol and fission-product behaviour phenomena;
• iodine chemistry phenomena;
• core-melt distribution and behaviour in containment phenomena;
• systems phenomena.

A synopsis is provided for each phenomenon, including a description, references for further information, significance for a design-basis accident (DBA) and a severe accident/beyond-design-basis accident (SA/BDBA) and a list of experiments that may be used for code validation.

The report identified 213 experiments, broken down into the same 6 categories (as done for the phenomena). An experiment synopsis is provided for each test. Along with a test description and references, the synopsis also identifies the availability of the report and data, phenomena covered by the test, type of test (separate effect, combined effect or integral test), whether it covers DBA and/or SA/BDBA conditions, the range of key experimental parameters and past code validation/benchmarks.

This CCVM identified experiments for 93% of the phenomena requiring validation. However, if only experiments suitable for CFD validation are considered, then only about half of the phenomena are covered by the CCVM.

Safety significance

This activity aims at a significant improvement of the reliability of severe accident containment models/codes including the influence of accident management procedures. The result is an increased understanding of the related phenomena and demonstrates where experimental gaps still exist.

Use and users of the results

The validation matrix is of use to code developers in listing key phenomena/processes and experiments in which these phenomena/processes are explored, and to code users who wish to be trained in the performance of a code using a selected set of experiments independent of those used in its development. The emphasis of the report is towards the latter application. Integral experiments are preferred, where interactions of phenomena as well as the phenomena themselves are considered as being more representative of these conditions expected during plant transients. Separate effect tests are also of value, in order to ensure that the full set of phenomena/ processes and their associated parameter ranges are adequately covered.

Lead organisation

The lead organisations for the group were:

• 2002-2009 – Gesellschaft für Anlagen- und Reaktorsicherheit mbH (GRS)
• 2009-2014 – Atomic Energy of Canada Limited (AECL)

Participants

Last updated: 4 May 2015