UNCERTAINTY AND SENSITIVITY ANALYSIS OF THE PHEBUS FPT1 TEST SIMULATION RESULTS

The PHEBUS program is an example of an international cooperative research program providing
valuable data for validating severe accident analysis computer codes. One of the main objectives of the
PHEBUS program is to study the process in the bundle of the light water reactors during severe accident
conditions as well as the release, transport, and retention of fission products.
In this article the PHEBUS - FPT1 test will be analyzed in more detail, focusing on the late phase of
bundle degradation. Best estimate simulations are provided using best estimate, severe accident code -
RELAP/SCDAPSIM. However, the simulation of processes in the nuclear installation during accidents
is attended with many uncertainties. There are three major sources of uncertainties: code or model
uncertainty, representation or simulation uncertainty, and plant (installation) uncertainty. This article is
focusing on the uncertainties which come from the initial parameters and boundary conditions. For the
uncertainty and sensitivity analysis of the RELAP/SCDAPSIM calculation results, the GRS
methodology is used together with SUSA statistical tool.
The results of uncertainty analysis show the accuracy of the calculation results. The upper and lower
limits of the calculation results (cladding temperature and total hydrogen generation) is bounding
experimental data. The results of the sensitivity analysis give a perception of the influence of the
uncertain initial parameters on the calculation results. Sensitivity analysis indicated thermal
conductivity of ZrO2 as the dominant initial parameter for the calculation results. Using experience
gained form the application of Uncertainty and Sensitivity analysis will be possible to update the model
in order to get more accurate simulation results of the PHEBUS - FPT1 test.