Elaboration of a Phenomena Identification Ranking Table (PIRT) for the modelling of In-Vessel Retention. (October 2020)

• Record Type: Journal Article
• Authors: Fichot, F.; Carénini, L.; Bakouta, N.; Esmaili, H.; Humphries, L.; Laato, T.; Le Tellier, R.; Saas, L.; Melnikov, I.; Pandazis, P.; Weber, S.; Park, R.J.; Filippov, A.; Strizhov, V.
• Is Part Of: Annals of nuclear energy. Volume 146(2020)
• Publication Date: 2020-10
• Subjects: IVR strategy -- PIRT -- Molten pool -- Reactor pressure vessel -- Stratification; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 621.4805
• DOI: 10.1016/j.anucene.2020.107617
• Languages: English
• ISSNs: 0306-4549

Abstract

Highlights: Identification of important phenomena for IVR modelling. All major severe accident codes included. Uncertainty evaluation from sensitivity studies. Importance of transient stratification and heat transfer. Abstract: The strategy for In-Vessel Retention (IVR) of corium is currently considered for several new reactor designs in various countries. One of the issues for the demonstration of the success of this strategy is that there are significant uncertainties in physical modelling of corium in the lower plenum and its transient chemical and thermal interactions with the vessel leading to its significant ablation. Since the initial approaches developed in the nineties for the AP600 and Loviisa VVER-440 plant, knowledge about corium pools (thermochemistry and heat transfer characteristics) and mechanical behaviour of the vessel has improved and it is now possible to model more accurately this phenomenon. The H2020 European project IVMR (In-Vessel Melt Retention) addresses the issue of selecting and improving IVR models for safety evaluation. As a first step, a Phenomena Identification Ranking Table (PIRT) involving the relevant physical processes was developed. The methodology followed to build this PIRT results from the consideration of a few principles: (a) Identifying and separating the risks with respect to which the importance of a physical process is evaluated. (b) Defining physical processes or parameters which can be considered as independent of the other ones. (c) Avoiding expert judgement, as much as possible, and using, instead, the results of previous sensitivity studies to estimate the impact of each physical process or parameter. In order to obtain results representing the currently "shared" state of knowledge, code developers of the most widely used codes were consulted. It includes ASTEC, MELCOR, MAAP (EDF version), SOCRAT, ATHLET-CD integral codes and PROCOR, SIMPLE, HEFEST_URAN, and IVRSYS dedicated codes (specific to IVR calculations). One of the positive outcomes of this PIRT is that the results show significant tendencies, with distinct groups of phenomena or parameters. This allows the identification of the uncertainties and of the phenomena/variables with the highest (or lowest) importance. Some dispersion in the results could also be noticed but it can be understood either because of the variability due to reactor design or the impact of models chosen by experts in the codes they have developed. When the dispersion cannot be explained, it indicates that the phenomenon is poorly understood and may deserve further research. Among the phenomena with highest importance, the heat transfers in the top metal layer and the chemical/thermal interaction with the oxide crust have been identified, as well as the transient formation and stratification of metal and oxide layers, including thermochemical peculiarities of the (U, Zr, Fe, O) system. Another phenomenon of highest importance is the mechanical behaviour of the thin ablated vessel wall, where elasticity, plasticity and creep all play a role. It is concluded that codes should include improved models for those phenomena in order to be able to provide a reasonable assessment of potential for IVR.