Uncertainty analysis of transportable fluoride-salt-cooled high temperature reactor (TFHR) using dakota coupled with relap5-3D

Reliable systematic uncertainty analysis is of great significance for reactor safety and licensing. Methods have been developed and presented to quantify the uncertainty of computer code results in this paper. The uncertainty and sensitivity quantification investigation of MIT proposed 20MWth prismatic Transportable Fluoride-salt-cooled High Temperature Reactor (TFHR) is performed with the DAKOTA code coupled to the commercial nuclear best-estimate software RELAP5-3D. Eight model input parameters were selected with specified variations and probability density functions, a total of 200 steady-state calculations were then conducted in order to statistically determine the 5% and 95% peak fuel temperature tolerance intervals with 95% confidence level. It was found that uncertainties in the heat transfer coefficient and helium gap were the most significant contributors to the statistical results of the peak fuel temperature. About 0.5% of calculated fuel temperatures exceeded the fuel damage limit temperature 1573 K, if randomly propagated uncertainty has been taken into account. The findings indicate further design optimization from thermal-hydraulics prospective is needed.