Analysis of the FLUOLE experiment for the APOLLO2 validation of PWR core reflectors

The recent release of the JEFF3.1/CEA2005 nuclear data library and the latest improvements of the APOLLO2.8 code (281-group SHEM energy mesh, mixture self-shielding treatment and performing 2D solver using characteristics method) have allowed the development of a new powerful deterministic depletion calculation scheme applied to BWR and PWR, called REL2005. For three years, CEA has been carrying out an intensive program of numerical and experimental validation of this new neutronics package APOLLO2.8-MOC/CEA2005-SHEM/REL2005 for LWR calculations. This paper is devoted to the reflector effect in the specific case of the 1300 MWe French PWRs. This heterogeneous 2 cm-thick SS baffle, with water reflector beyond, generates a challenging problem in GEN-II calculation. In a first step, an extensive numerical validation of APOLLO2.8 has been carried out against TRIPOLI4 continuous energy Monte Carlo calculations. Then, the experimental validation was performed against FLUOLE measurements (FLUOLE is a PWR mock-up in the EOLE facility at Cadarache Centre). Calculation- Experiment agreement on core residual reactivity is better than 200 pcm. However, pin-by-pin power at the core/baffle interface is slightly overestimated by about 2 to 3 %. The present work demonstrates that the APOLLO2.8-MOC/CEA2005-SHEM/REL2005 calculation can be considered as a reference deterministic calculation. This conclusion means that Monte Carlo calculation is no longer required to obtain reliable PWR reflector parameters for industrial calculations. (authors)