Hot Rod Statistical Method for Large Break Loss of Coolant Accident With CATHARE

For nearly 30 years, Areva NP has contributed, with CEA, EDF and IRSN, to the development of CATHARE, one of the most advanced computer codes in the field of thermal-hydraulics calculations on the system-wide scale, with the objective to have at its disposal a high-performance tool, as the basis for the development of best-estimate methods for design and safety studies. At the same time, Areva NP has conceived with EDF a generic safety analysis method called Deterministic Realistic Method (DRM), which has been applied to elaborate and license ECCS evaluation models on the basis of the current CATHARE 2 V1.3L code version. DRM has then been extended by Areva NP with the Hot Rod Statistical Method (HRSM) which adds a statistical treatment of the hot rod-related uncertainties to the deterministic reactor system thermal hydraulic calculation and that can be used with other LB LOCA evaluation models. With the HRSM, the value of the peak cladding temperature envelope with a 95% probability level is determined based on direct Monte Carlo calculations statistically combining three key hot rod parameters randomly varied according to their probability distribution. The final value of the peak cladding temperature is then calculated taking into account an uncertainty calculated using the Bootstrap method. The HRSM is a method, based on CATHARE that is easy to implement as a complement to LB LOCA evaluation model. The large number of direct calculations performed ensures a very high degree of precision when evaluating the peak cladding temperature and the equivalent cladding reacted. The HRSM is currently used and licensed in France for the LB LOCA transients, and is now extended to SB LOCA transients using CATHARE 2 v2.5. An illustrative comparison of HRSM versus DRM typical results for Areva NP 3-loop and 4-loop plants then underlines the significant margins provided by this method.