Integral loop test and assessment of modified RELAP5/MOD3.3 for RCS coolant inventory during a LBLOCA

This study investigates experimentally and analytically the thermal hydraulic phenomena during the transition from design basis accident (DBA) to beyond-DBA, particularly, the depletion of core coolant inventory. To investigate the overall thermal hydraulic behavior after safety injection (SI) failure during a large-break loss-of-coolant accident (LBLOCA) in a cold leg, an integral loop test was performed at the Seoul National University Integral Test Facility (SNUF), which was scaled down to 1/6.4 in length and 1/178 in area from the advanced power reactor 1400 MWe (APR1400) according to the three-level scaling method. The plant condition at 200.0 s as the base case and those at 625.0 and 1950.0 s as test cases after the initiation of LBLOCA were applied as initial conditions in the experiments, respectively. The experimental results showed that the sweepout increased the coolant flow discharged to the break depending on the steam flow rate in intact cold legs and the initial downcomer coolant level and expedited the depletion of the core coolant inventory.In the meantime, since RELAP5/MOD3.3 uses the average properties of donor volume as those of its connected junction, this scheme causes the mass and the energy flux in a junction to be calculated incorrectly if significant phase separation occurs in the donor volume such as in the downcomer during the LBLOCA. The sweepout model was developed and implemented in RELAP5/MOD3.3 to improve its calculation of coolant inventory during the LBLOCA. To assess the applicability of the modified RELAP5/MOD3.3 to the actual system, the experiments in SNUF were simulated by both the original and the modified RELAP5/MOD3.3. The original one predicted the discharge flow rate at the break larger than that of the experiment. On the other hand, the modified one calculated the discharge flow rate more similar to that of the experiment than the original one did. As a result, the modified RELAP5/MOD3.3 reduced the coolant flow discharged to the break to delay the initiation time of heater heat-up after SI failure during LBLOCA in a cold leg. This improved RELAP5/MOD3.3 will support a more realistic thermal hydraulic analysis in an integrated analysis system.