Numerical and Experimental Analysis of Passive Cooling through Night Ventilation (Numerieke en experimentele analyse van passieve koeling door nachtventilatie)

Nowadays, the application of active cooling in office buildings is hardly questioned in building practices. Also in temperate climates, active cooling is often assumed to be a necessary condition for a good thermal comfort. However, an optimal building design can go a long way towards reducing the cooling load. For example, night cooling can maintain thermal comfort during the day, by pre-cooling the building at night using cool exterior air at high air change rates. By day, the thermal mass buffers the heat released in the building, thereby reducing and delaying air temperature peaks.To address this, a thorough study is made of the transient behaviour of a room with night ventilation. The methodology can be subdivided into three main parts, a first part being numerical, a second part experimental and a third part focussing on implementation in BES models. The numerical CFD study focusses on obtaining a better insight in the transient surface heat transfer during night ventilation. In particular, the study takes into account the influence of thermal mass by including concrete slabs as floor and ceiling in the model. Furthermore, the transient behaviour of the air-mass system is simulated for different air change rates and initial indoor-outdoor temperature differences. The influence of excluding radiation from the model is investigated, as well as the impact of geometrical variations. Interesting conclusions could be drawn regarding the unsteady flow pattern. Three flow patterns were observed: buoyant flow, forced flow and transition flow, which is the intermediate regime between the first two flow regimes. The flow pattern strongly influenced the corresponding local surface convection. However, as this numerical study is limited to 2D with simplified geometry and constant boundary conditions, only qualitative conclusions can be drawn. An experimental study is required to confirm the observed phenomena.The second part of the study will therefore discuss the measurements conducted in a full scale test room under realistic climatic boundary conditions. As was the case in the numerical model, the thermal mass is also provided in the experimental set-up, by including two 15 cm thick concrete slabs as floor and ceiling. The room has dimensions close to a one-cell office, and is exposed to the climate through a SW-oriented façade with automated hopper window. The room is provided with more than 200 sensors for a detailed monitoring of air, surface and mass temperatures, along with surface heat flux, air velocity and pressure difference. Apart from the mechanical ventilation, the room is free floating. The measurements provide a detailed insight into the phenomena occurring during night ventilation under realistic circumstances. They allow confronting the assumptions made in BES models with reality and confirm the tendencies that were observed in the numerical study, such as the three flow regimes. The determination of the air change rate proved challenging and was investigated with tracer gas experiments. Furthermore, a detailed analysis was made of the thermal response of the room. Though determination of convective heat transfer coefficients was not possible with sufficient accuracy, the assumptions used in BES models could be confronted with the measurement results. Smoke visualisation experiments were conducted to determine a correlation between thermal parameters and flow behaviour. These findings are confronted with literature. Finally, a number of selected nights are simulated with CFD as well, illustrating that CFD must be used with extreme care.The final part of the study focusses on the implementation in BES models. The room is simulated in TRNSYS 17, using the results of a four week measurement campaign as boundary conditions and verification data. The influence of included multiple air nodes in one thermal zone is investigated, illustrating the difficulty of an accurate modelling of the intra-zonal

Jaar van publicatie:2013

Toegankelijkheid:Closed