Experimental study on two-dimensional film flow with local measurement methods

• An experimental study on the two-dimensional film flow with lateral air injection was performed.
• The ultrasonic thickness gauge was used to measure the local liquid film thickness.
• The depth-averaged PIV (Particle Image Velocimetry) method was applied to measure the local liquid film velocity.
• The uncertainty of the depth-averaged PIV was quantified with a validation experiment.
• Characteristics of two-dimensional film flow were classified following the four different flow patterns.

In an accident condition of a nuclear reactor, multidimensional two-phase flows may occur in the reactor vessel downcomer and reactor core. Therefore, those have been regarded as important issues for an advanced thermal-hydraulic safety analysis. In particular, the multi-dimensional two-phase flow in the upper downcomer during the reflood phase of large break loss of coolant accident appears with an interaction between a downward liquid and a transverse gas flow, which determines the bypass flow rate of the emergency core coolant and subsequently, the reflood coolant flow rate. At present, some thermal-hydraulic analysis codes incorporate multidimensional modules for the nuclear reactor safety analysis. However, their prediction capability for the two-phase cross flow in the upper downcomer has not been validated sufficiently against experimental data based on local measurements. For this reason, an experimental study was carried out for the two-phase cross flow to clarify the hydraulic phenomenon and provide local measurement data for the validation of the computational tools. The experiment was performed in a 1/10 scale unfolded downcomer of Advanced Power Reactor 1400 (APR1400). Pitot tubes, a depth-averaged PIV method and ultrasonic thickness gauge were applied for local measurement of the air velocity, the liquid film velocity and the liquid film thickness, respectively. The uncertainty of the depth-averaged PIV method for the averaged velocity of the liquid film was discussed. Also the local velocity profiles of air and liquid film and the liquid film thickness distribution were presented. These local experimental data of two-dimensional film flow which simulated the two-phase cross flow can be used to validate the multidimensional models in the system analysis codes and CFD codes.