Modeling and Assessment for the Estimation of Heat Transfer Coefficients on Curved Downwards-Facing Heated Walls in Application to ERVC

External Reactor Vessel Cooling (ERVC) is a proposed technique for severe accident mitigation in nuclear power plants. ERVC is aimed at containing the molten corium inside the Reactor Pressure Vessel (RPV) by transporting heat through the coolant (water) which occupies the intentionally flooded reactor cavity and is in direct contact with the outer surface of the lower head. An experimental study along with a computational fluid dynamics (CFD) model are used in the present study to investigate the local convection heat transfer coefficient on curved heated upper walls during the early stages of this type of severe accidents. The CFD study is realized using the open-source code OpenFOAM, in which proper formulation of the physics in terms of mass, momentum, and energy conservation is implemented. Results from the CFD model are compared to the experimental data obtained from a flow-boiling test facility, commissioned, and operated at Khalifa University, in the single-phase regime. The experimental findings highlight that the flat plate correlations are inaccurate for the estimation of the heat transfer coefficient on RPV lower head, especially at high Reynolds number values. Furthermore, the fully resolved turbulence model k-omega SST is the most accurate in predicting the actual thermal boundary layer and turbulent mixing. Quantification of the uncertainty in the experimental measurements as well as the discretization error in the numerical results is provided. In addition, a sensitivity analysis is provided for the inlet turbulence conditions to highlight their importance for accurate solutions. This study provides a framework for the assessment and validation of CFD codes dealing with heat transfer in scenarios where the flow is affected by curved geometries and fluid properties’ variations. In addition, it paves the way for more accurate heat transfer correlations suitable for heat flux partitioning models that can accurately capture the boiling curve in ERVC systems.