Experimental and numerical evaluation of debris transport augmentation by turbulence during the recirculation-cooling phase of a LOCA

Abstract

In resolving the safety issue of sump clogging due to debris generated by the type of high-energy line break known as a GSI-191 event, determination of the debris transport fraction, which is likely to be augmented by fluid turbulence during the recirculation cooling phase of a pressurized water reactor (PWR), is very important in the sizing of the sump screen area. In the present study, the debris transport fraction during the recirculation cooling phase of the OPR1000 plant is evaluated with and without consideration of turbulence debris augmentation. To do this, first a computational fluid dynamics (CFD) analysis of the flooded containment floor during the recirculation cooling phase is performed to obtain mean flow fields and the turbulence kinetic energy field assuming a double-ended guillotine break of a hot leg. Then, experiments involving tumbling velocities measurements of the surrogate debris for the OPR1000 plant and supplementary CFD analyses are performed to verify the turbulence effect on debris transport. From these findings, the turbulence effect on the degree of debris floor tumbling augmentation was found to be represented by the algebraic sum of the mean horizontal velocity and the horizontal fluctuating velocity deduced from the turbulent kinetic energy (TKE). Based on this experimental finding and on the CFD analysis of the containment floor when flooded, the debris transport fraction is evaluated for typical fibrous types of debris, such as NUKON, with respect to two size classes. The result shows a considerable increase in the debris transport fraction when a turbulence effect is implemented compared to when it is not. Increases of 5.55 and 2.06 times are observed for large NUKON and small/fine NUKON, respectively. This result implies that the turbulence effect should be considered in the debris transport quantification for conservatism. It was also found that the debris transport fraction may change depending on which sump is active between the two sumps. For example, small/fine NUKON is much more transportable when a sump near the break operates compared to when a sump far from the break operates. This fact also implies that the location of the active sump should be considered as an important parameter in comprehensive debris transport evaluations given a maximum head loss at the sump screen.