Abstract
The use of turbulence models leads to uncertainties in the computed results because the models are not universal. Therefore, it is essential to validate the CFD program by experimental data. A computational fluid dynamics (CFD) program with the Re-Normalization Group (RNG) k-ϵ model was used for prediction. Validation was conducted by comparing the flow patterns, vertical profiles of temperature, concentration, velocity, and turbulence intensity between measurement and computation for an individual office, a cubicle office, and a quarter of a classroom. The computed air temperature and velocity agree well with the measured data. The agreement is less satisfactory between the computed and measured temperatures in the lower part of the office. The CFD model used therefore should be able to predict the indoor environments characterized by strong pressure and buoyancy driven flows.
UDC: