Evita in-vessel aspects and ASTEC perspectives in SARNET

QUENCH workshop, 13-15 Octobre 2003, Karlsruhe
J.P. Van Dorsselaere (IRSN), F. Jacq, S.Pignet (IRSN), H.J. Allelein (GRS), M.Buck, M.Buerger (IKE), M. Bendiab, M.K.Koch (RUB/LEE), J.M. Veteau (CEA).

The aim of the EVITA project (EC 5th Framework Programme) was the validation of the IRSN-GRS Severe Accident integral code by independent users, and a final users’ experience feedback to the development team. It gathered 18 European partners (utilities, designers, safety authorities, R&D) from 8 countries. Two successive versions were released: V0.3 in 2000 and V1.0 mid-2002. The latter simulates the complete scenario, starting from the front-end thermalhydraulics phase, and includes improved core degradation models. Two types of work were performed: first, validation on experiments mostly selected form OECD ISPs (LOFT-LP-FP2, CORA-13, VANAM…), and secondly, plant applications on different reactors (French PWR 900, German PWR 1300, VVER-440 and –1000). Results were compared with detailed codes for validation part and with integral codes (MELCOR, MAAP4) for plant applications part in order to position ASTEC with respect to the State of the Art.

Two new modules were developed in this version ASTEC V1.0: CESAR for RCS two-phase thermalhydraulics (5-equation approach with a drift flux correlation for water velocity), and DIVA for core degradation, based on most models from the IRSN detailed code ICARE2 (only some specific simplified models were specifically developed: core thermalhydraulics and corium in lower plenum).

Some examples of DIVA validation are presented on the following experiments: PERICLES boil-off (work by CEA), CORA-13 (work by IKE), LOFT-LP-FP2 (work by RUB), and QUENCH-05 (work by IRSN). See also the FZK presentation at the QUENCH workshop on a preliminary DIVA application to the ISP45.

These first applications showed that the level of ASTEC models is near the State of the Art in thermalhydraulics and early core degradation. Some needs of improvements have already been identified (droplet field in the dry region…) but the main ones concern reflooding of intact and degraded cores. The code portability on different computers was checked to be good. An important EVITA outcome was also the needs of improvement of robustness of the CESAR-DIVA coupling and of post-processing tools.
An important step will be in 2004 the new EC 6th FWP project Severe Accident Network of Excellence SARNET (gathering 52 organizations) where ASTEC will be the reference European integral code. Successive versions will be released to 30 organizations, and a first user’s club will take place in early 2005. Similar work as in EVITA will continue on validation and on plant applications benchmarks. But other objectives, in the frame of a future version ASTEC V2, will be: on one hand, extension of models to BWR, CANDU, RBMK and to future NPP designs; on the other hand, progressive integration into ASTEC of models which will be developed in the course of R&D in the SARNET thematic topics Corium, Containment and Source term.

In the next 2 years, the IRSN-GRS efforts will focus on code consolidation (especially CESAR-DIVA) through an extension of validation (TMI-2, LOFT…), and through plant calculations and benchmarks with other codes. Another objective (again in the frame of ASTEC V2 version) will be the merging of DIVA module with the detailed code ICARE2.