SAFETY EVALUATION OF PROTOTYPE FAST-BREEDER REACTOR: ANALYSIS OF ULOF ACCIDENT TO DEMONSTRATE IN-VESSEL RETENTION

摘要

In the original licensing application for the prototype fast-breeder reactor, MONJU, the event progression during an unprotected loss-of-flow (ULOF), which is one of the technically inconceivable events postulated beyond design basis, was evaluated. Through this evaluation, it was confirmed that radiological consequences could be suitably limited even if mechanical energy was released. Following the Fukushima-Daiichi accident, a new nuclear safety regulation has become effective in Japan. The conformity of MONJU to this new regulation, hence, should be investigated. The objectives of the present study are to conduct a preliminary evaluation of ULOF for MONJU reflecting the knowledge newly obtained after the original licensing application, and to gain the prospect of In-Vessel Retention (IVR) for the conformity of MONJU to the new regulation. In the evaluation of event progressions during ULOF, the whole sequence was categorized into 1) initiating, 2) transition, and 3) post-accident-material-relocation/post-accident-heat-removal (PAMR/PAHR) phases. In the initiating phase, fuel pin disruption caused by coolant boiling would result in axial fuel dispersion in subassembly (SA). In the transition phase, molten-core pool would be formed due to the failure of SA walls, and the molten fuel would be discharged through the control-rod guide tubes (CRGTs). In the PAMR/PAHR phase, molten fuels discharged through CRGTs would be relocated and be stably cooled in the lower plenum by decay-heat removal. The methodology of the present study and its results can be summarized as below: 1) The initiating phase was evaluated by SAS4A code reflecting the models and parameters for fuel-pin disruption and fuel dispersions based on the CABRI experiments. Contrary to the original licensing evaluation showing 380 MJ in mechanical energy release under conservative conditions, the present evaluation showed that no significant energy release would take place. 2) The transition phase was evaluated by 3-dimensional SIMMER-Ⅳ code reflecting the models and parameters for CRGT failure and molten-fuel discharge based on the EAGLE experiments. Contrary to the past 2-dimensional evaluation showing 150 MJ in mechanical energy release under conservative conditions, the present evaluation showed that the released mechanical energy would be remarkably reduced because the non-physical axisymmetric/coherent fuel compaction peculiar to 2-dimensional evaluation was appropriately mitigated in 3-dimensional evaluation. 3) The PAMR/PAHR phase was evaluated by S-COPD, FLUENT codes and heat-balance calculations reflecting the present evaluation of the precedent phases. Contrary to the past evaluation involving the uncertainties in molten-fuel fragmentation and debris-bed formation, the present evaluation showed that stable cooling of discharged core materials could be achieved even if fragmentation was incomplete. The preliminary evaluation in the present study showed that no significant mechanical energy release would take place, and that thermal failure of the reactor vessel could be avoided by the stable cooling of disrupted-core materials. This result suggests that the prospect of IVR against ULOF, which lies within the bounds of the original licensing evaluation and conforms to the new nuclear safety regulation, will be gained.