Verification of a TRACE EPR™ model on the basis of a scaling calculation of an SBLOCA ROSA test

摘要

In cooperation with the Finnish Radiation and Nuclear Safety Authority (STUK), a project has been launched at the Paul Scherrer Institute (PSI) aimed at performing safety evaluations of the Olkiluoto-3 nuclear power plant (NPP), the first EPR™, a generation III pressurizer water reactor (PWR); with particular emphasis on small-and large-break loss-of-coolant-accidents (SB/LB-LOCAs) and main steam-line breaks. As a first step of this work, the best estimate system code TRACE has been used to develop a model of Olkiluoto-3. In order to test the nodalization, a scaling calculation from the rig of safety assessment (ROSA) test facility has been performed. The ROSA large scale test facility (LSTF) was built to simulate Westinghouse design pressurized water reactors (PWR) with a four-loop configuration. Even though there are differences between the EPR™ and the Westinghouse designs, the number of similarities is large enough to carry out scaling calculations on SBLOCA and LOCA cases from the ROSA facility; as a matter of fact, the main differences are located in the secondary side. Test 6-1 of the ROSA 1 programme, an SBLOCA with the break situated in the upper head of the reactor pressure vessel (RPV), was of special interest since a very good agreement with the experiment was obtained with a TRACE input deck. In order to perform such scaling calculation, the set-points of the secondary relief and safety valves in the EPR™ nodalization had to be changed to those used in the ROSA facility, the break size and the core power had to be scaled by a factor of 60 (according to the core power and core volume) and the pumps coast down had to be adapted to the ones of the test. The calculation showed very similar results as the experiment and the ROSA-TRACE calculation. The only significant difference observed was a faster primary depressurization after the break flow turned to single-vapor flow. This difference could be explained on the basis of geometrical differences between the EPR™ and ROSA/Westinghouse RPVs designs.