• CFD
• DNS
• Pressurized Thermal Shock

Pressurized Thermal Shock (PTS) is considered as an important issue that challenges the integrity of the Reactor Pressure Vessel (RPV). A PTS consists of a rapid cooling of the RPV wall in a high pressure environment that may induce the propagation of flaws inside the vessel wall. The most severe PTS event has been identified in the Emergency Core Cooling (ECC) injection during a Loss-of-Coolant Accident (LOCA). The injected cold water mixes with the hot primary water present in the cold leg and the mixture flows towards the downcomer, where further mixing takes place, leading to large temperature gradients at the vessel surface.
Thermal Hydraulic system codes, traditionally used for Nuclear Reactor Safety (NRS) applications, fail to reliably predict the complex three-dimensional thermal mixing phenomena occurring during the ECC injection. Hence, CFD can bring real benefits in terms of more predictive capabilities. However, to gain trust in the application of CFD to PTS, a comprehensive validation programme is necessary, by comparing CFD results with the available experimental data. In the absence of detailed experimental data for the RPV cooling during ECC injection, high fidelity Direct Numerical Simulation (DNS) databases constitute a valid alternative and could serve as a reference.
The aim of this work is to design a numerical experiment to generate a high quality reference DNS database for a simplified PTS configuration. This takes into account the turbulent mixing in the downcomer and the evolution of temperature distribution for both structures and fluid during a single-phase ECC injection scenario. The spectral element solver NEK5000 has been selected to perform the DNS calculations. Such a DNS analysis represents a very demanding application. On one hand, it has to represent a physically meaningful test for the validation of CFD models to PTS. On the other hand, it has to be simple enough to isolate the phenomena of interest from the overall real scenario, and to be performed by DNS. As a result, a calibration study consisting of a wide range of RANS calculations has been performed in order to obtain a feasible and well representing PTS computational domain for DNS. Therefore, the resulting numerical experiment has been optimized in terms of geometry, boundary conditions, and properties in order to meet a feasible computational demand.
Furthermore, preliminary DNS calculations for turbulent flow in a square duct have been carried out with the solver NEK5000, and results are compared with previously obtained RANS results and with the available reference databases.