Prolonged Available Time and Safe States: Final Report from the 2019 NKS-R PROSAFE activity

Definitions for accident states and safe states are decisive for both deterministic and probabilistic safety assessments (DSA & PSA) of nuclear facilities. For instance, the IAEA’s guides on the performance of deterministic and probabilistic safety assessments state that determination of mission times should take into account the time it takes to reach a safe, stable shutdown state. Fundamentally, it is a matter of finding an appropriate balance between the level of realism of models and practicality of the modelling approach. One cross-cutting modelling issue in this respect is the choice of mission time and related success criteria for systems, and the possibility to realistically include recovery and repair for long time windows. In DSA, it is often adopted from the previous praxis justifying what is sufficient. In PSA, the modelling approach itself forces to simplify treatment of mission time, and repairs are mostly not considered. Use of single time window simplifies modelling, but in the light of occurred events (Fukushima Daichii), implementation of new technology in the nuclear power plants (e.g. independent core cooling), consideration of non-reactor nuclear facilities (e.g. spent fuel pools) and decommissioning phase reactors, such a simplified approach may need justification and/or to be reconsidered. In any case, the definition of a mission time is dependent on the definition of safe and stable state. Since selection of mission time has an impact on many modelling aspects, and hence on the PSA results, it is important to study possibilities to treat mission times more realistically. For longer time windows, it becomes evident to consider e.g. time-dependent success criteria and possibilities for recovery and repair. However, for these issues there is not yet a consensus on how they should be addressed. The PROSAFE project started 2019 with financial support from NKS, NPSAG and SAFIR, with the objective to improve the quality of safety assessment methods with respect to safe and stable state definition and assessment of long time windows, including human reliability analysis in long time window scenarios, use of dynamic success criteria, crediting repairs and modelling of different time windows.