Methodology for Thermal Analysis of Spent Fuel Dry Storage Cask Design using MARS Code

The casks have been designed to meet the safety acceptance criteria provided by US NRC in terms of structural, thermal, confinement, shielding, criticality, and material against various off-normal conditions and postulated accident scenarios. The passive decay heat removal capacities of the designed spent fuel dry storage casks have been quantified by experiments with scale-down mock-ups and numerical simulations with computational fluid dynamic (CFD) codes. Many research also analyzed the thermal behavior inside the cask regarding various conditions by commercial CFD codes and experiments. However, the thermal analysis during the transient condition through CFD codes require much time, and the obtained results are sensitively dependent on many parameters such as mesh cell number, boundary conditions, and applied models. Besides, realization of the transient conditions in the experiments with scale-down mock-up or full-scale product is difficult. Thus, one-dimensional transient analysis code regarding thermal-hydraulics behavior in a large size system, MARS (multi-dimensional analysis for reactor safety) code, was selected as simulation tool for spent fuel dry storage casks. In this study, experimental work and MARS simulation were peformed with comparison of their results to find the best-estimate analysis method regarding the thermal behavior of the spent fuel dry storage cask. For the analysis of thermal behavior inside the spent fuel dry storage cask, MARS code, which is one-dimensional system analysis code was used in this study. The analysis methodology considering the effects of cylinder geometry on natural convective heat transfer, and radiation heat transfer showed better prediction capability than the simulation results based on default models in terms of air temperature distribution at steady state. It is expected that the best-estimate methodology, suggested by this study, will be more efficient safety analysis method, compared to the previous analysis tools, because it reduces calculation time required to the simulation of transient conditions such as off-normal and accident with good prediction accuracy.