Modeling the consequences of fuel assembly bowing on PWR core neutronics using a Monte-Carlo code

Stanislas de Lambert (1) , Guillaume Campioni (2) , Vincent Faucher (3, 4) , Bertrand Leturcq (5) , Jérome Cardolaccia (1)

1 SEMT - Service d'Etudes Mécaniques et Thermiques
2 SERMA - Service d’Études des Réacteurs et de Mathématiques Appliquées
3 CEA-DES (ex-DEN) - CEA-Direction des Energies (ex-Direction de l'Energie Nucléaire)
4 DTN - Département Technologie Nucléaire
5 LM2S - Laboratoire de Mécanique Systèmes et Simulation

The effect of assemblies bowing in PWR nuclear reactors onto core neutronics is an observed phenomenon still poorly understood which can lead to Power Ratio Tilt. Studies of the consequences of rod/assembly bowing involve many different fields addressed by nuclear power plant, such as neutronics, thermohydraulics, mechanics.. . in a complex combination of multi-physical interactions. For the neu-tronic part, the modeling of bowed assemblies in Monte Carlo codes must allow to correctly describe the shape of fuel rods. In this article, two discrete ways to model bowed geometries are tested: the first one consists in a stacking of vertical small cylinders following the shape of the fuel rod by small shifts between neighboring cylinders; the second one, newly introduced in the present research, consists in a sequence of rotated cylindrical segments arranged to as to follow the shape of the fuel rod more closely. Both models are used to reproduce two specific bowing patterns, namely C-shape and S-shape, for which a reference modeling involving an analytical toroidal volume cut by planes is available for use with CEA's Monte Carlo code Tripoli-4 Ò. Results of comparisons between both models and analytical reference show that, even if the segment modeling requires a specific effort to handle implementation constraints, it appears preferable compared to stacking modeling. It provides accuracy with fewer discrete entities and is therefore computationally affordable and it is far more robust when increasing bowing deflection. This approach is thus only considered ready for its application to any kind of bowing patterns.

Domaines: Sciences de l'ingénieur [physics]

Stanislas de Lambert, Guillaume Campioni, Vincent Faucher, Bertrand Leturcq, Jérome Cardolaccia. Modeling the consequences of fuel assembly bowing on PWR core neutronics using a Monte-Carlo code. Annals of Nuclear Energy, 2019, 134, pp.330-341. ⟨10.1016/j.anucene.2019.06.017⟩. ⟨cea-02479080⟩