On the suitability of steady RANS CFD for forced mixing ventilation at transitional slot Reynolds numbers.

UNLABELLED Accurate prediction of ventilation flow is of primary importance for designing a healthy, comfortable, and energy-efficient indoor environment. Since the 1970s, the use of computational fluid dynamics (CFD) has increased tremendously, and nowadays, it is one of the primary methods to assess ventilation flow in buildings. The most commonly used numerical approach consists of solving the steady Reynolds-averaged Navier-Stokes (RANS) equations with a turbulence model to provide closure. This article presents a detailed validation study of steady RANS for isothermal forced mixing ventilation of a cubical enclosure driven by a transitional wall jet. The validation is performed using particle image velocimetry (PIV) measurements for slot Reynolds numbers of 1000 and 2500. Results obtained with the renormalization group (RNG) k-ε model, a low-Reynolds k-ε model, the shear stress transport (SST) k-ω model, and a Reynolds stress model (RSM) are compared with detailed experimental data. In general, the RNG k-ε model shows the weakest performance, whereas the low-Re k-ε model shows the best agreement with the measurements. In addition, the influence of the turbulence model on the predicted air exchange efficiency in the cubical enclosure is analyzed, indicating differences up to 44% for this particular case. PRACTICAL IMPLICATIONS This article presents a detailed numerical study of isothermal forced mixing ventilation driven by a low-velocity (transitional) wall jet using steady computational fluid dynamics (CFD) simulations. It is shown that the numerically obtained room airflow patterns are highly dependent on the chosen turbulence model and large differences with experimentally obtained velocity fields can be present. The renormalization group (RNG) k-ε model, which is commonly used for room airflow modeling, shows the largest deviations from the measured velocities, indicating the care that must be taken when selecting a turbulence model for room airflow prediction. As a result of the different predictions of the flow pattern in the room, large differences are present between the predicted air exchange efficiency obtained with the four tested turbulence models, which can be as high as 44%.