Toshio Yamanaka, Eunsu Lim, Tomohiro Kobayashi, Toshihiko Sajima, Kanji Fukuyama
Year:
2019
Languages: English | Pages: 10 pp
Bibliographic info:
40th AIVC - 8th TightVent - 6th venticool Conference - Ghent, Belgium - 15-16 October 2019

The ventilative cooling by natural ventilation is important technology for the buildings in urban area for the sake of energy saving and BCP (Business Continuity Plan). In fact, a large number of high-rise buildings in urban area in Japanese metropolises are equipped with natural ventilation apparatus such as openings and chimneys or shafts. In design stage of the building, generally, building engineers have to decide the size and number of ventilation devices, but it is not easy because the data of wind pressure coefficient are not ready for engineers to use easily especially for the buildings in high density block of metropolis. In designing the natural ventilation devices, a wind guide or a wind vane sticking out from walls of the building is useful device to introduce natural wind from outside into the rooms. The wind velocity around the building, however, is essential to predict the performance of these devices. Here, the wind velocity around building depends on the distance between buildings, height above ground and wind speed of approach flow and so on. It can be said that the data of wind pressure on building wall and wind velocity around buildings are necessary for natural ventilation design of the buildings in high density blocks of metropolis. 
In this study, in order to provide the data for natural ventilation design of buildings in metropolis, the distribution of wind pressure coefficient on the building walls and wind velocity distribution between the buildings were measured by wind tunnel test with 1:1000 scaled models and CFD analysis by LES. In CFD analysis, the dependency of wind pressure and velocity on Reynolds number between buildings, that is so-called Reynolds Number Effect, was examined. 
As a result of CFD by LES, it was turned out that Reynolds number effect is inevitable to some extent in the wind tunnel test with 1:1000 scaled models under 10m/s of approaching wind, but the basic data of wind pressure coefficient and wind velocity were obtained by the wind tunnel test. Additionally, a simple method to predict the pressure loss and airflow rate based on the pressure boundary of the block without the gap between buildings were presented.  
In the presentation of AIVC conference, some movies of airflows around buildings simulated by LES will be presented with many important data of wind pressure or wind velocity.