An earthquake is one of the most serious phenomena for the safety of a nuclear reactor in Japan. Therefore, structural safety of nuclear reactors has been studied and nuclear reactors ware contracted with structural safety for a big earthquake. However, it is not enough for safety operation of nuclear reactors because thermal-fluid safety is not confirmed under the earthquake. For instance, behavior of gas-liquid two-phase flow is unknown under the earthquake conditions. Especially, fluctuation of void fraction is an important factor for the safety operation of the nuclear reactor. In the previous work, fluctuation of void faction in bubbly flow was studied experimentally and theoretically investigate the stability of the bubbly flow. In such studies, flow rate or void fraction fluctuations were given to the steady bubbly flow. In case of the earthquake, the fluctuation is not only the flow rate, but also a body force on the two-phase flow and shear force through the pipe wall. Interactions of gas and liquid through their interface also act on the behavior of the two-phase flow. The fluctuation of the void fraction is not clear for such complicated situation under the earthquake.
Therefore, the behavior of gas-liquid two-phase flow is investigated experimentally and numerically in the series of study. In this study, to develop the predictive technology of two-phase flow dynamics under earthquake acceleration, a detailed two-phase flow simulation code with an advanced interface tracking method TPFIT was expanded to two-phase flow simulation under earthquake conditions. In this paper, the bubbly flow in a horizontal pipe excited by oscillation acceleration and under the fluctuation of the liquid flow was simulated by using the expanded TPFIT. Predicted time series of velocity profiles around the bubbles and shapes of bubbles were compared with measured results under flow rate fluctuation and structural vibration. Predicted results were almost same as measured results qualitatively. And it was concluded that the expanded TPFIT can be applied to qualitative analysis of bubbly flow under accelerating conditions.