Sensitivity Analysis by the Use of a Surrogate Model in LB-LOCA: LOFT L2-5 With CATHARE-2 V2.5 Code

Titre du congrès :NURETH-12 (Nuclear Reactor Thermal Hydraulics) Ville du congrès :Pittsburgh Date du congrès :30/09/2007

The revision of the 10 CFR50.46 in 1988 has made possible the use of best-estimate codes may be used in safety demonstration and licensing, provided that uncertainties are added to the relevant output parameters before comparing them with the acceptance criteria. The uncertainty of output parameters comes principally from the lack of knowledge of the input parameters. In the safety analysis of the large break loss of coolant accident (LB-LOCA), for a best-estimate analysis, it was agreed that the 95th percentile estimated with a high degree of confidence should be lower than the acceptance criteria. It appeared necessary to IRSN, technical support of the French Safety Authority, to get more insight into these strategies. Furthermore Best-Estimate Plus Uncertainty (BEPU) analysis will be developed not only in thermal-hydraulics but in other fields such as in neutronics. The application of the BEPU analysis can be made in three steps in a statistical evaluation: 1.The definition and the evaluation of uncertainties of the input parameters: determination of the statistical characteristics. 2.The modelling and understanding of the output parameters. 3.The evaluation of the 95th percentile with a high degree of confidence. As a rule the sensitivity analysis is done with linear correlation coefficients. This paper presents a new approach to perform a more accurate sensitivity analysis. Toward the objective of assessing uncertainty, it is useful to analyse the peak cladding temperature (PCT) further, one of the quantitative criteria of the LB-LOCA for instance. Indeed we would like to determine and to classify the main input parameters or to trace the sensitivity of the PCT to each input parameter. Such information is difficult to obtain directly by Monte-Carlo methodologies using the thermal-hydraulic code, because it is rather time-consuming. Therefore, it is natural to replace it with a simpler model, estimated with few runs. This approach is called response surface methodology. We suggest Kriging for its construction. Finally, we make a sensitivity analysis with the SOBOL methodology. The paper presents the application of the previously described methodology on the LOFT (Loss-of-Fluid Test) loss of coolant experiment L2-5 which simulated a double-ended offset shear of a cold leg in the primary coolant system. LOFT L2-5 was the subject of the former ISP 13 and the ongoing BEMUSE international problem. The output is the second maximum Peak Cladding Temperature of the fuel. The best estimate code used is CATHARE2 V2.5.