Idea Transcript
Computational analysis of natural ventilation for a high-rise apartment building with a core well Yen-Yi Li 2, Hsu-Cheng Chiang 1, Hsiao-Chi Hsu 1, Kuo-Shu Hung 1, Hsi-Sheng Wu 1 Industrial Technology Research Institute (ITRI), Hsinchu, Taiwan Department of Interior Design, Shu Te University, Kaohsiung, Taiwan
SUMMARY High-rise apartment building is designed with a vertical hollow core, or void, to improve its ventilation and thermal comfort. Outdoor air flows through a series of passageways in this apartment and finally is discharged to the void at the centre of building, and thus creates a natural ventilation flow. Statistical comfort models and a full-scale CFD simulation of building air flow were used to analyze the feasibility of natural ventilation for a 24-storey building in Hsinchu city. As being always concerned of the realization of natural ventilation for a building in the sub-tropical climate areas, this paper has demonstrated some procedures to reveal the expectable performance of natural ventilation system. KEYWORDS Architecture and design, Natural ventilations, Residential, Thermal comfort 1 INTRODUCTION Due to scarce landscape resources in urban cities of Taiwan, modern buildings are constructed in high-rise type no matter for office or residential purpose. Construction companies always want to build high-rise buildings to increase economic values. In the meantime, they are seeking ways to differentiate their architecture designs in the market. In this case, fully glazed glass façade was considered for a 24-storey high-rise apartment building because it could provide a broad view over the city. The glass façade has been known to be more likely energy-wasting. In order to bring the unfavourable circumstance toward a positive direction, the concept of natural ventilation was introduced to its design. Natural ventilation system has been often proved to improve human health and comfort. However, very few studies for highrise apartment buildings can be referenced especially in the hot and humid climate zones. In a natural ventilation building, air flow is created by buoyancy force, wind force or both. Natural ventilation relying solely on buoyancy force can not be easily achieved in hot climate areas, where the temperature difference between indoor and outdoor is relatively small. Some designs considered adding a solar chimney at the top of a double-skin façade to ensure a reliable and stable ventilation performance for buildings even without the assistance of wind force. But for high-rise buildings, more feasible practice is a wind-induced ventilation system. Choi et al. (2007), Wang and Wong (2007) and Karava et al. (2007) studied the cross ventilation which wind force contributes to the air flow by creating a pressure difference between the windward and leeward of building openings. In contrast, Kotani et al. (2003), Priyadarsini et al. (2004) and Prajongsan and Sharples (2012) diverted the cross flow into a vertical flow by enclosing a shaft in the buildings. Ventilation shafts located at proper positions in room can establish a better oriented ventilation air flow. It can also resolve the deficiency for rooms with single-side ventilation in which the openings are in just one external wall. In this study, natural ventilation of this 24-storey apartment building with a vertical light well or void was investigated. Such a building structure was commonly used in modern office buildings in Japan (Nikken Sekkei 2006). Meanwhile, Kotani et al. (2003) had done a serial of
experimental studies focused on the air quality in the light well which was likely to be contaminated by the exhaust air from apartments. In the current study, the owner hopes that the special building design can improve the ventilation performance and lessen the nighttime loads by avoiding heat accumulation during the daytime. 2 METHODS Before conducting a complete numerical simulation for the modeled high-rise apartment building, feasibility of natural ventilation in Hsinchu city was examined based on two comfort models which require only the inputs of weather data. Hsinchu city is located in the northern part of Taiwan with latitude of 24.49 and longitude of 121.04. The weather is a typical subtropical climate. The average temperature in summer is around 26~30°C but the highest temperature can get to 33~34°C in daytime and occasionally last for weeks. The average temperature in winter is about 15~20°C. The mean relative humidity is about 75~80% throughout the year. The city is known for its windy weather. Surface wind accelerates its speed across the city because of several geographical features, such as the city is at the narrowest throat of Taiwan Strait and has mountainous terrain. Therefore, our goal is to use the encouraged prevailing wind to design a natural ventilation system for this high-rise apartment building. Evaluating the feasibility of natural ventilation according to statistical models According to the study of de Dear and Brager (1998), people in natural ventilated buildings wear appropriate clothes and open or close windows to adjust the air flow. So there is a difference between the predicted comfort level and the observed response. Their studies suggested that psychological adaptation is the most likely explanation and lead to an alternative prediction of thermal comfort ranges in ASHRAE Standard 55 (2004). The optimal comfort temperature linked only to the monthly outdoor temperature can be expressed as the regression line as following,
Tcomf = 0.31 × Ta ,out + 17.8
(1)
Where Tcomf (°C) is optimum comfort temperature which is characterized in terms of average outdoor dry bulb temperature, Ta,out. The indoor comfort temperatures at which 80% and 90% of occupants feel comfortable respectively are shown in Figure 1 for Hsinchu city. Based on this new thermal comfort model, buildings in this city although can be naturally ventilated but most of the time in summer is close to or above the upper 80% acceptable limits, and likely ceiling fans or other air moving devices are needed to enhance indoor air movement for proper comfort. Wang and Wong (2007) did derive a modified PMV model which takes into account of the effect of indoor air speed for naturally ventilated residential buildings from field survey data in Singapore. The PMV thermal comfort model has the regression equation as following,
PMV = −11.7853 + 0.4232 × Temp − 0.57889 × V
(2)
Where Temp (°C) indicates indoor air temperature and V (m/s) refers to indoor air velocity measured at 1.2m above the floor. As relative humidity is always in the high level (above 60%) in Singapore and its variation is also highly correlated with dry bulb temperature, it does not explicitly appear in the equation. The equation is suitable for the Clo of 0.34~0.5 and the Met of 1.0 which are the regular conditions for residential buildings. According to their suggestion, the upper limit of indoor discomfort should be set to be PMV=1.1. Assumingly if indoor air
temperature equals to outdoor air temperature, apparently in Table 1, the maximum outdoor temperatures from July to September will make extra indoor air moving required to keep the modified PMV smaller than 1.1, however, higher air speeds might create a nuisance problem. Therefore, in summer, when high outdoor temperature persists, air conditioners will be needed to maintain a comfortable living, especially during the daytime for buildings in Hsinchu city. Adaptive Comfort for Space With Operable Windows in Hsinchu City 35.00
80% Acceptability
33.00
Temp (degree C)
31.00
90% Acceptability
29.00 Maximum average outdoor temperature Average outdoor temperature
27.00 25.00 23.00 21.00
90% Acceptability
19.00 17.00
80% Acceptability
15.00 Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Figure 1. Indoor comfort ranges in naturally ventilated buildings in Hsinchu city Table 1. Comfort conditions for naturally ventilated buildings in Hsinchu city Month Maximum average outdoor temperature (°C) Room air speed (m/s) Modified PMV
May
Jun
Jul
Aug
Sep
Oct
28.70
29.80
33.60
34.00
32.30
27.50
-0.36
0.55 0.50
2.3 1.10
2.6 1.10
1.35 1.10
--0.15
Numerical simulation for natural ventilation of the modeled building Based on the results of the comfort models, it is not possible to have a building in Hsinchu city to be naturally ventilated for a whole year. Therefore, the best alternative design is to have this 24-storey building be most benefitted from natural ventilation flow to prevent heat from accumulating inside this residential building in daytime and to lessen the pull-down loads for air conditioners when rooms are occupied. The perspective view of whole building and the single floor plan are shown in Figure 2. Although this 24-storey building is located at the center of city, there is no adjacent high-rise building, which might disturb the coming wind, because this building is right next to the City Park. From the figure of floor plan, it can be seen that the building is designed to be ventilated through a series of air flow passageways. Outdoor air enters a suite of rooms through the openings of windows, leaves the rooms from the grilles mounted on the walls of each room, then gathers at the main corridor and finally is exhausted to outside from two sound-attenuated channels connecting the apartment to the void which composes the hollow core of the building.
The flow resistance components, grilles and sound-attenuated channels, were experimentally tested by using a Fig. 15 test rig of AMAC 210 standard to acquire the flow correlations to assure the accuracy of simulation. The flow resistance with respective to the flow rate can be expressed as the regression equation of ΔP = c • Q a , where ΔP(mmAq) is the flow resistance and Q(cmm) is the flow rate. The coefficients of c and a are shown in Table 2 for these two flow resistance components. The flow resistance of sound attenuated channel is almost in square proportional to the flow rate which is very similar to the performance of orifices, while the flow resistance of grille is linearly proportional to the flow rate.
The solar radiation loads on the building façade can be approximated by building energy simulation code and iteratively coupled with CFD program (Wang and Wong, 2007) but the processes could be very time consuming. In this study, after carefully examining the built form, orientation, façade and climate data, the solar radiation loads on the south-east and the south-west façades are assumed to be 150 W/m2 and 400 W/m2 respectively; the two north facing façades are assumed to have a 50 W/m2 heat load considering the radiation reflecting from the roofs of surrounding low-rise buildings. Internal heat sources such as lighting and equipments have been ignored. The building façade is constructed by 8mm Low-E glass with thermal conductance of 4.1 W/(m2K). Local wind direction in the summer is coming from the south and south-west directions in the average speed of 1.8~2.2 m/s. The wind profile is calculated by the following equation, v = vo ( z / zo )α , in which vo is the average wind speed of the prevailing wind at the height of 10 m, α is the coefficient of surface roughness and in this case 0.33 is used. 400W/m2 50W/m2 Wind c d c b Solar load 150W/m2
50W/m2 a
a- opening of window,
b- grille on the wall,
c- ventilation channel,
d- void
(a) Perspective (b) Floor Plan Figure 2. Geometric model of the high-rise apartment building in simulation Table 2. Regression correlations of two flow resistance components c
a
R2
Grille on the wall
0.0272
1.197
0.937
Sound-attenuated channel
0.0048
2.0443
0.996
Flow component
Picture of component
The CFD simulation of air flow was first conducted for a single floor geometric model in a very fine mesh with 515,646 grids. The single floor simulation was then followed by a fullscale simulation for this 24-storey building in a rather coarse mesh with similar grid number like the single floor due to the constraint of our computing power. The simplification led to a certain deviation of air flow rate, but it can still provide valuable information to reveal the natural ventilation flow driven by both wind and thermal buoyancy forces. 3 RESULTS The vertical room temperature distributions on the section cutting from the windward façade, through the two ventilation channels and the void, and to the leeward façade are shown in Figure 3. The temperature distributions for the rooms with and without ventilation channels have shown totally different patterns at the same outdoor temperature, wind speed and 10 cm openings of windows. It is obvious that the additional ventilation grilles and channels in this building can establish smooth flow paths to discharge the hot air from the rooms to the void, while the traditional apartment without these ventilation components ended up with higher room air temperatures for those rooms exposed to direct solar radiation (i.e. windward rooms).
Windward room without ventilation channel
Leeward room without ventilation channel
Windward room with ventilation channel
Leeward room with ventilation channel
Figure 3. Temperature distributions for the rooms with and without ventilation channels (at the outdoor temperature of 28 °C and the wind speed of 2 m/s) Table 3 shows the vertical room temperature distribution as the wind speed goes up from 0.5 to 4 m/s. Higher wind speed can create a larger ventilation air flow and eventually bring the windward room temperature close to the outdoor. But the leeward rooms are less affected by the wind speed and always stay at a comfortable range. In particular, thermal buoyancy driven flow is weak and results in high room temperature as in the case of wind speed of 0.5 m/s. It agrees with the finding of Priyadarini et al. (2004) that passive stack effect is not reliable in hot climate areas, where the temperature difference between indoor and outdoor is small. Table 3. Temperature distributions of rooms at the windward and the leeward sides of building at different outdoor wind speeds and the outdoor temperature of 28 °C Wind speed
0.5 m/s
1.0 m/s
2.0 m/s
4.0 m/s
Temp.
Windward room Leeward room
The flow simulation results for the whole building are depicted in Figure 4 in terms of speed and temperature distributions at the low wind speed (1 m/s) and the high wind speed (4 m/s) conditions, respectively. In here, the wind speed is calculated at the building height by using the specified wind profile. At the low wind speed condition, natural ventilation is mainly invoked by buoyancy force and unfortunately ventilation flow rate is not enough to keep the windward side rooms, which receive high solar radiation loads, in a comfortable indoor temperature. A clear thermal plume is formed and adhered to the left wall of the void. It indicates that the discharged air speed from the ventilation channels on each floor is too small to form a jet flow. In contrast, at the high wind speed condition, the discharged air speed is very high so that there are jet flows on every floor. Under the circumstance, the whole apartment building is very well ventilated and cooled by the wind-driven ventilation flow.
Wind at the building height = 1 m/s
Wind at the building height = 4 m/s
Figure 4. Speed and temperature distributions of modeled building at different wind speeds 4 DISCUSSION Numerical simulation of a whole building has been done in a rather coarse mesh compared to the one used for a single floor in this study which will inevitably result in large deviation. Seeking for a greater computing power certainly can conquer the problem. Another way is to represent the flow from the openings of windows to the void with flow correlations to avoid heavy use of computing resources. It will be the future research direction of our interests. 5 CONCLUSIONS Building designers have often been concerned about the applications of natural ventilation for buildings in a hot and humid climate. This paper has illustrated some procedures integrating the use of comfort models and numerical flow simulation to estimate the performance of natural ventilation building. The results seem to be able to provide valuable information. ACKNOWLEDGEMENT The authors would like to express their gratitude for Energy R&D funding from Bureau of Energy, Ministry of Economic Affairs, R.O.C. (Taiwan) to support this research work. 6 REFERENCES Choi T. et al. 2007. Improving ventilation performance in high-rise residential building by natural ventilation system. Proceedings of the International Conference on Sustainable Building Asia, Seoul, Korea, 363–368. Wang L. and Wong N.H. 2007. The impacts of ventilation strategies and facade on indoor thermal environment for naturally ventilated residential buildings in Singapore. Building and Environment, 42, 4006–4015. Karava P. et al. 2007. Wind-induced natural ventilation analysis. Solar Energy, 81, 20–30. Kotani H. et al. 2003. Natural ventilation of light well in high-rise apartment building. Energy and Buildings, 35, 427-434. Priyadarisini R. et al. 2004. Enhancement of natural ventilation in high-rise residential building using stack system. Energy and Buildings, 36, 61-71. Prajongsan P. and Sharples S. 2012. Enhancing natural ventilation, thermal comfort and energy saving in high-rise residential buildings in Bangkok through the use of ventilation shafts. Building and Environment, 50, 104-113. Nikken Sekkei Ltd. 2006. The vertical void is ready for the future changes. Tokyo, Japan. de Dear, R.J. and Brager G.S. 1998. Towards an adaptive model of thermal comfort and preference. ASHRAE Transactions, Vol 104 (1), 145-167. ASHRAE. 2004. ANSI/ASHRAE Standard 55-2004, Thermal Environmental Conditions for Human Occupancy. Atlanta: American Society of Heating, Refrigerating, and AirConditioning Engineers, Inc.