ICONE11-36212 NUMERICAL STUDY ON THE CHARACTERISTICS OF AIR BUBBLE OSCILLATION IN WATER

Abstract

In both a boiling water reactor and an advanced type of pressurized water reactor under construction in Korea named APR1400,when a pressure relieving system is in operation, water, air and steam discharge successively into a sub-cooled water pool through spargers. Among the phenomena occurring during the discharging processes, the air bubble clouds with a low-frequency and high-amplitude oscillation may result in significant damage to the submerged structures if the resonance between the bubble clouds and structures occur. The phenomena involved are so complicated that most predictions of frequency and pressure loads have resorted to experimental work and computational approach has been precluded. This paper deals with a numerical simulation on the behavior of air bubble discharging into a sub-cooled water pool through a sparger without LRR (Load Reduction Ring) by using a commercial CFD code, FLUENT. The simulation includes both the transient flows of water, air and steam in the pipe and the oscillation of the air bubble subsequently formed in the pool. The VOF (Volume Of Fluid) model is selected to simulate the multi-flows of water, air and steam. It precludes a modeling of steam condensation in the pool since the resultant loads of steam condensation are relatively lower than those from the air bubble oscillation. The simulation shows that one large air bubble which is formed at the sparger head oscillates and moves upward to the water surface subsequently. Comparison of the cases with and without LRR (Note that the authors had subsequently. Comparison of the cases with and without LRR (Note that the authors had previously performed a simulation on the behavior of air bubble with LRR.) shows that the maximum dynamic pressures at the wall without LRR are much larger than those with LRR, which implies that LRR has a great impact on reducing the dynamic pressures. The frequency of the case with LRR is slightly lower due to the larger size of air bubble cloud formed only at the sparger head. The pressure and temperature distributions obtained in the computation seem to be reasonable from a physical viewpoint. It seems that the change of initial air mass in the computation range does not have an impact on the dynamic pressures at the wall. Whereas, the frequency increases as the air mass decreases, which is consistent with the well-known trend that the frequency of a large air bubble is inversely proportional to the air bubble radius. The dynamic pressures and the frequency at the wall increase as the inlet boundary pressure increases, which is consistent with the previous investigations.