AN EMPIRICAL METHOD FOR INVERSE UNCERTAINTY QUANTIFICATION IN CRITICAL FLOW SIMULATION

Uncertainty quantification in thermal-hydraulic and severe accident analysis is often performed by the probabilistic propagation of input uncertainty through a simulation code. The probability density functions of the uncertain inputs are usually characterized by expert judgement. Inverse Uncertainty Quantification (IUQ) aims at providing a way to go beyond the reliance on expert judgment by making use of available experimental data. In this paper, we developed an empirical IUQ method within the ATRIUM project of the OECD/NEA/CSNI/WGAMA. The method is applied to critical flow simulation results obtained by TRACE best estimate thermal-hydraulic system code developed by the U.S. Nuclear Regulatory Commission and the data from two experimental facilities (Sozzi-Sutherland and Super Moby Dick). The uncertainties are characterized and used to perform a forward uncertainty propagation on a generic three loop PWR-900, considering an accident scenario with the guillotine break of the pressurizer surge line.