Uncertainty and sensitivity analysis of the Phébus FPT3 test

Abstract

Since the accident in the Fukushima Daichii Nuclear Power plant specific attention has been paid to the analyses and possible mitigation of severe accidents, as well as to design extension conditions. There is also a growing interest in applying not only the so-called best estimate codes to evaluate the postulated severe accident sequences, but to complement these studies with uncertainty analyses. The Phébus FP (Fission Product) research program, conducted by IRSN and CEA, aimed to study the physical phenomena that govern core degradation as well as, the release, transfer, and deposition of fission products and their behaviour in the containment. In this paper the degradation phase of the Phébus FPT3 experiment is modelled and simulated with the help of the best estimate system code ATHLET-CD 3.2, the core degradation module of AC2. The study is complemented by an uncertainty analysis as well, applying the so-called GRS method. The paper considers uncertain model (e.g. parameters concerning Zr oxidation) as well as experimental (e.g. material properties) parameters and determines the 95/95 tolerance limits for the main figures of merit. Furthermore, a sensitivity analysis is performed, and the Spearman’s rank correlation coefficient is determined for each uncertain input parameter.