How good is “good agreement” Considerations when comparing experiments with simulations for rod bundles with wire spacers

In validation exercises, numerical simulations are compared directly with experimental data. If the agreement is sufficiently good, then the model is considered validated (with the reported accuracy), and it can be used with a high level of confidence for the investigation of closely related scenarios that have not been or cannot be studied experimentally. A key element in this comparison is the justification of the modeling assumptions and simplifications, demonstrating that the dominant physical phenomena are well represented, and thus simulations of similar scenarios are reliable. Rod bundles with wire spacers are used as fuel assemblies in many liquid-metal cooled fast reactor designs. The thermal-hydraulic scenario is complex due to the intricate geometry and the low Prandtl number of the coolant. While several experimental campaigns are reported in the open literature, some important aspects must be taken into account for comparing numerical and experimental results. This work discusses the impact of geometry simplification (e.g. wire shape) and uncertainty in the location of thermo-couples, as well the influence of material properties and boundary conditions. Moreover, relative errors must be defined with respect to the correct reference value in order for them to strongly support conclusions regarding the accuracy. Two reference cases are selected for detailed analysis. They cover the study of the velocity profile in an isothermal case, and the temperature profile in a heated case with local blockages.