Validation of CFD Simulations for
Natural Ventilation
Yi Jiang1,
Camille Allocca1 and Qingyan Chen2
1Department
of Architecture, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
2School of Mechanical Engineering, Purdue University, West Lafayette,
IN 47907, USA
Order Paper
£3.50 (offer only available from this site)
Abstract
Natural ventilation, which may provide occupants with
good indoor air quality and a high level of thermal comfort, and reduce energy
costs, has become an important sustainable strategy in building designs. This
investigation used three computational fluid dynamics (CFD) models: steady
Reynolds averaged Navier-Stokes equation (RANS) modelling, unsteady RANS
modelling, and large eddy simulation (LES) to study both wind-driven and
buoyancy-driven natural ventilation. The validation of the CFD models used the
experimental data of wind-driven natural ventilation, obtained from a wind
tunnel with a scaled building model, and the data of buoyancy-driven
ventilation, obtained from a full-scale chamber. LES results seem to be more
accurate and informative than those obtained with the RANS modelling, but with
severe penalty in computing time. This investigation has also analysed
turbulence energy spectra of natural ventilation. The peak turbulence energy for
wind-driven natural ventilation is at frequencies higher than that for buoyancy
driven natural ventilation. Thus, the fluctuating flow field plays a more
important role in determining ventilation rate for wind-driven natural
ventilation than for buoyancy-driven natural ventilation.
Key words:
computational fluid dynamics (CFD), large eddy simulation (LES), Reynolds
averaged Navier-Stokes equation (RANS) modelling, experimental measurements,
natural ventilation.
|
