In this paper we present recent results of the application of the thermal-hydraulic code BAGIRA to the analysis of complex two-phase flows in nuclear power plants primary loops. In particular, we performed benchmark numerical simulation of an integral LOCA experiment performed on a test facility modeling the primary circuit of VVER-1000. In addition, we have also analyzed the flow patterns in the VVER-1000 steam generator vessel for stationary and transient operation regimes. For both of these experiments we have compared the numerical results with measured data. Finally, we demonstrate the capabilities of BAGIRA by modeling a hypothetical severe accident for a VVER-1000 type nuclear reactor. The numerical analysis, which modeled all stages of the hypothetical severe accident up to the complete ablation of the reactor cavity bottom, shows the importance of multi-dimensional flow effects.

Word subject domains have been widely used to improve the perform-ance of word sense disambiguation al-gorithms. However, comparatively little effort has been devoted so far to the disambiguation of word subject do-mains. The few existing approaches have focused on the development of al-gorithms specific to word domain dis-ambiguation. In this paper we explore an alternative approach where word domain disambiguation is achieved via word sense disambiguation. Our study shows that this approach yields very strong results, suggesting that word domain disambiguation can be ad-dressed in terms of word sense disam-biguation with no need for special purpose algorithms.