Evaluation of ASSERT-PV V3R1 void fraction predictions against the OECD/NEA benchmark data

The subchannel code ASSERT, developed by Atomic Energy of Canada Limited, is assessed against the steady-state void fraction cases from the OECD/NEA BFBT benchmark. Preliminary void fraction predictions from 84 BWR bundle tests were compared to experimentally reported values, and 45.2% of the simulated data fell within the reported experimental error. Overall, the code was able to predict the void fraction to within ±0.10 of the experimental value about 95% of the time, although calculations of the corner subchannels were not as accurate, falling within ±0.10 of the experimental value only 69.3% of the time. An analysis of four isolated cases suggested that the code over predicts the void fraction in the corner and side channels for cases at low pressure (≃1.0 MPa). A sensitivity analysis on one case demonstrated that the boundary condition uncertainties could affect the void fraction by up to 0.026. Accounting for the boundary condition uncertainties in this case increased the number of subchannel predictions falling within experimental error from 34 to 46. Additionally it was determined that for the case examined the boundary condition uncertainty most significantly affected the central subchannel void fraction prediction.