Abstract Supercritical carbon dioxide (S-CO2) Brayton cycles have emerged as a promising solution for power generation systems, owing to their high thermal efficiency across a wide range of heat source temperatures. These systems are particularly well-suited for small and micro-reactors due to their robust load-following capabilities. Radial turbines remain an optimal choice for power outputs in the range of several megawatts, and numerous research groups have conducted computational fluid dynamics (CFD) studies to evaluate their performance. However, the validation of these CFD analyses has predominantly relied on one-dimensional design codes rather than experimental data, which limits their reliability. At the Korea Advanced Institute of Science and Technology (KAIST), a radial inflow turbine is being operated in the Autonomous Brayton Cycle (ABC) Test Loop to produce a detailed performance map, including metrics such as pressure ratio, power output, and pressure difference. This study aims to validate previously proposed turbulence models by comparing CFD simulations with experimental data obtained from the test loop. The simulations are conducted using Ansys CFX 2024R2, with S-CO2 properties directly imported from the NIST REFPROP database to ensure reasonable accuracy. The CFD analysis employed a CAD model of the radial turbine used in recent experiments. The results showed a discrepancy of approximately 10% in pressure when compared with the experimental data.
Lee et al. (Mon,) studied this question.