Kim Goethals

Sensitivity of the predicted convective heat transfer in a cooled room to the computational fluid dynamics simulation approach

Computational fluid dynamics (CFD) plays an increasingly important role in the design, analysis and optimization of engineering systems. However, CFD does not necessarily provide reliable results. The most crucial numerical solution error is caused by inadequate grid resolution, and the key modelling error sources in CFD in ventilated indoor environments are turbulence modelling and diffuser modelling. Many researchers already proposed guidelines, but they based their analyses on local variables. In response, underlying study intended to verify the impact of the CFD simulation approach on the convective heat flux, an integral quantity. The authors tested several grids, Reynolds averaged Navier–Stokes turbulence models and diffuser models for three convection regimes in a cooled room. The diffuser modelling had a much larger impact than the grid and the turbulence modelling, as long as the jet dominated the airflow. So, CFD users, who want to model forced/mixed convection airflow indoors, certainly need to pay attention to the diffuser modelling.