Improving the prediction of zonal modeling for forced convection airflows in rooms

Abstract

Evaluating the efficiency of a ventilation system in providing a healthy indoor environment to building occupants requires the knowledge of the airflow inside the room prior to the calculation of the pollutant dispersal from their sources. Computational Fluid Dynamics (CFD) modeling is probably the most suitable method to achieve this as it provides detailed information about the airflow pattern but is still time-consuming although processor speed has tremendously increased during the past years. Consequently, parametric studies aimed at characterizing the influence of parameters such as the location of the air inlets and outlets or furniture on the air velocity, temperature and concentration distributions within the building are seldom achieved. Zonal, Coarse-grid CFD and Fast Fluid Dynamics (FFD) are intermediate models between CFD and single air node models; they can predict the airflow pattern in a room or group of rooms with less computational efforts but with a lower accuracy than CFD models. The present study aims at improving the zonal prediction for the case of forced convection in indoor spaces. Results show that the proposed alterations of zonal modeling greatly improve the prediction and that this model requires less computational efforts and returns more accurate results than Coarse-grid CFD and FFD.