A 25% double-ended LOCA in the PSB-VVER facility

The paper presents the results of the experimental investigation in the PSB-VVER facility and post-test analysis with RELAP5/MOD3.3 of the thermal-hydraulic response of the PSB-VVER to a 25% double-ended Hot Leg Break (HLB). The test scenario included loss of off-site power concurrently with the scram-signal and the safety system operation as described in the reference VVER-1000 operational manual in the case of this type of accident assuming one diesel-generator failure. The key transient parameter trends as well as sequence of events and phenomena are given in the paper. RELAP5/MOD3.3 a post-test analysis has been performed using the experimental data gained as a base. The reasonable qualitative agreement between the key calculated and measured variables has been shown. The quantitative code accuracy evaluation has shown that the total average amplitude of the main parameters' deviations AAtot < 0.17 that means good calculation as qualified according to UMAE method for comparison 'test-post-test calculation' results. Being verified against the 25% HLB LOCA test data, RELAP5/MOD3.3 input model was used in pretest calculation of guillotine (2x100%) hot leg break transient in the PSB-VVER. Also the appropriate calculations were performed for VVER-1000 to evaluate capability of the PSB-VVER facility for modeling realistically in relation with the VVER-1000 reactor the double-ended coolant discharge during hot leg guillotine break LOCA. Quantitative comparison between PSB-VVER pre-test calculation and reference reactor calculation has given AAtot< 0.28 that corresponds to satisfactory quality of the VVER-1000 hot leg guillotine break LOCA modeling in the PSB-VVER. (author)