Predicting wind driven cross ventilation in buildings with small openings
MetadataShow full item record
Designing wind driven cross ventilation for a building is challenging due to the dynamic characteristics of wind. While numerous studies have studied various aspects of cross ventilation, few have had an opportunity to examine the topic with a holistic approach utilizing multiple research techniques. Thus, this dissertation combined three different investigation methods: wind tunnel analysis, full scale experiments and computational fluid dynamics (CFD) to examine the physics of wind driven cross ventilation. Following the systematic approaches of the three methods, this study first conducted full scale measurements of wind properties, façade pressures, air flow rates through small window openings, and tracer gas concentrations in a multi-zone test house. Secondly, a scaled model of the test house was studied in a boundary layer wind tunnel (BLWT) for its façade pressures and ventilation rate under various wind incident angles. Finally, a CFD model of the test house was simulated under various constraints to determine the factors which affect indoor air distribution during wind driven cross ventilation events. The full scale experimental results showed a strong correlation between the cross ventilation rate and the wind velocity component normal to the inlet openings. This correlation suggested that the cross ventilation flow rate could be estimated from wind conditions alone. A closer examination of the wind characteristics also revealed that the cyclical pattern of changing wind direction could be impacted by obstructions which are kilometers upwind, suggesting that distant landscapes could have an impact on cross ventilation flows. The combination of CFD and full scale measurements also showed that local heat sources can generate significant buoyancy driven flow and affect indoor mixing during wind-driven cross ventilation scenarios. Experimentally validated parametric CFD analyses demonstrated the effect of interior heat loads in driving internal airflow, and suggest that a small source (35W/m2) can increase the indoor mixing from less than 1 ACH to 8 ACH between indoor spaces. Finally, the wind tunnel and CFD coupled analysis was found to predict the cross ventilation flow which was also validated against the full scaled measurements. The prediction, which may only be applicable to similar building types with small openings, showed significant agreement that such method has potential as an innovative design tool for natural ventilation in buildings.