Node configuration uncertainty in nuclear safety analyses

Abstract

In nuclear safety analyses, a 1-D thermal hydraulic system analysis code with the best estimate plus uncertainty (BEPU) approach is generally used. There are sources of uncertainty such as nodalization, physical models, and numerical scheme. Among them, the authors specifically focused on the node configuration uncertainty related to spatial discretization that is regarded as less dominant sources of uncertainty in the code previously. Furthermore, it was not easy to quantify the uncertainty from the node configuration. The authors quantified and analyzed the node configuration uncertainty to understand how much it affects the solution. In this study, the effect of spatial discretization are studied by discussing simulation results for various spatially discretized node systems. The discussion on the node configuration uncertainty will be presented by examples of SUbcooled BOiling (SUBO) test simulation and Loss Of Fluid Test (LOFT) simulation using Multi-dimensional Analysis of Reactor Safety (MARS) code developed in South Korea. From SUBO test simulation, it was found that when the number of nodes is small than the node configuration uncertainty can overwhelm the uncertainty from the physical model. From LOFT simulation, it was revealed that the node configuration can cause bifurcation in the complex nuclear system analyses. Thus, it is concluded that the node configuration uncertainty can be quite significant and has to be considered in the nuclear system analyses.