In order to improve the safety of Boiling Water Reactor (BWR), it is required to know the behavior of the plant when an accident occurred as can be seen at Fukushima Daiichi nuclear power plant accident. Especially, it is important to estimate the behavior of molten core jet in the lower part of the containment vessel at severe accident.
In the BWR lower plenum, the flow characteristics of molten core jet are affected by many complicated structures, such as control rod guide tubes, instrument guide tubes and core support plate. However, it is difficult to evaluate these effects on molten core jet experimentally. Therefore, we considered that multi-phase computational fluid dynamics approach is the best way to estimate the effects on molten core jet by complicated structure.
The objective of this study is to develop the evaluation method for the flow characteristic of molten core jet including the effects of the complicated structures in the lower plenum. So we are developing a simulation method to estimate the behavior of molten core jet falling down through the core support plate to the lower plenum of the BWR. The method has been developed based on interface tracking method code TPFIT (Two Phase Flow simulation code with Interface Tracking). To verify and validate the applicability of the developed method in detail, it is necessary to obtain the experimental data that can be compared with detailed numerical results by the TPFIT. Thus, in this study, we are carrying out experimental works by use of multi-phase flow visualization technique. In the experiments, time series of interface shapes are observed by high speed camera and velocity profiles in/out of the jet will be measured by the PIV method.
In this paper, the outline of the developing method based on the TPFIT was explained. And, the developing method was applied to preliminary experiment with/without modeled complicated structures. As the results, predicted interface shapes were almost agreed with measured data. However, predicted falling down velocity of the jet was lower than measured data. We considered causes of this underestimation and improved the method and simulation conditions to resolve this problem.