The paper presents the investigation of a naturally cross ventilated building using both experimental and numerical methods with the parameters being the free-stream and the incidence angle of the wind to the openings of the building. The experimental methodology calculates the air change rate based either on measurements of the inlet velocity profile, the outlet velocity profile or the descending rate of the tracer gas concentration using the decay method. The numerical investigation is based on the solution of the governing Navier-Stokes equations in their full three dimensional expression, focusing on the time dependent character of the induced flow field. The numerical results are compared with corresponding experimental data for the three aforementioned experimental methodologies in the case of a full scale building inside a wind-tunnel. The numerical investigation reveals that for large incidence angle the flow is governed by an unsteady (in terms of time) character regarding especially the regions close to the inlet and outlet windows. Moreover, significant information and comments about the validity of the aforementioned experimental methodologies adopted in most of the recent works in this area of scientific research is provided, using the numerical predictions as a reference. It is concluded that velocity measurements in the inlet window are of high accuracy when the flow in this region has a steady character, whilst the accuracy of the tracer gas methodology depends significantly on the mixing of gas within the control volume of the building.

Keywords: Natural Ventilation, CFD, Trace Gas Methodology, Building