Multiscale Thermal-Hydraulic Analysis of the ATHENA Core Simulator

In the framework of the ALFRED research and development program, the ATHENA facility will be constructed for thermal-hydraulic analysis of full-scale ALFRED components and systems. The source system of the facility is the core simulator, which aims to be representative of an ALFRED average fuel assembly. Computational fluid dynamics (CFD) codes are gaining attention for the analysis of complex systems in pool-type reactors since they are able to reproduce three-dimensional phenomena. In this paper, a multiscale approach based on porous media is proposed to reduce the computational cost of the core simulator CFD model. The multiscale approach starts with the detailed simulation of the infinite lattice domain of the fuel assembly to characterize the porous media hydraulic behavior. Then the porous media are applied in the system model. Three different approaches are investigated: (1) adopting a single porous media for the entire fuel assembly, (2) representing the bundle with two porous domains, and (3) adopting the so-called hybrid medium. The results have been compared with the reference detailed CFD simulation for performance evaluation. The first step of the analysis is the application of the multiscale approach on the CIRCE fuel pin simulator to carry out a turbulence model validation against experimental data and a comparison of the three approaches with a proven CFD model. Then the approach is applied on the ATHENA core simulator exploiting the CIRCE results. The results obtained with the porous media models are compared with a detailed CFD simulation of the core simulator to evaluate the performance of the three approaches. Eventually, the best solution is applied on a model of the entire ATHENA core simulator integrated with the feeding region. The model is tested also in transient conditions. The numerical experiment demonstrates the effectiveness of the multiscale approach in reducing the computational cost while maintaining high accuracy in representing the quantities of interest.