Impact of shading louvers on wind-driven single-sided ventilation in a multi-storey building

. As a passive building energy-saving technology, shading systems are usually used to reduce the solar radiation into the unit and reduce the energy consumption of air conditioning system. For multi-storey buildings, shading devices can not only change the indoor thermal environment, but also change the indoor wind environment and air quality because their impact on airflow patterns of natural ventilation. In this paper, the numerical simulation method was used to investigate external shading louvers on the single-sided ventilation performance and indoor air quality of units at different positions in a multi-storey building. Three approaching airflow directions and three rotation angles of louvers were considered and analysed. The results show that the evaluation of ventilation performance of shading louvers in multi-storey buildings is important for shaded buildings, which provide a new perspective for the geometry design of shading louvers and ventilation design of buildings.


Introduction
With the increasingly serious environmental and energy problems, energy conservation and emission reduction have become important measures to maintain a lowcarbon economy and sustainable development in China. Building energy consumption accounts for a large proportion of the total energy consumption and is one of the important links in promoting energy conservation and emission reduction, which has attracted great attention. As a passive building technology for energysaving, shading devices are usually used to reduce solar radiation entering units to reduce energy consumption of air-conditioning systems [1,2]. In addition, shading devices have an influence on natural ventilation during the transitional seasons and can change the indoor environment and air quality while satisfying indoor light and thermal conditions. The impact of architectural shading on indoor air quality is based on its change of the thermal and wind environments of buildings. As building components, the presence of shading devices interferes with airflow of natural ventilation, which changes the wind characteristics around the building, such as convective heat transfer coefficient, wind pressure coefficient, and airflow patterns [3][4][5][6].
In this paper, numerical simulations of the coupled indoor and outdoor airflow field in a multi-storey building with and without shading louvers were performed. The indoor space was divided by floors and columns, and the single-sided natural ventilation capacities in different units were investigated under different approaching airflow conditions and different shaded conditions. The influences of rotation angle of louvers and wind direction on air change rate and age of * Corresponding author: qhtao@jmu.edu.cn air were analysed. The results show that the impact of shading louvers on natural ventilation capacity is related to the unit location, rotation angle, and wind direction. The findings of this study can be used as a reference for shading design and operation settings based on singlesided natural ventilation.

Geometry model
In this study, a 1:30 reduced-scale model was selected to measure the wind velocity field around a cubic building. The simplified geometric model of the shaded building is shown in Fig. 1. The building height h 0 was 0.5m. As shown in Fig. 1(a), 26 rows of single-sided shading louvers were installed near the facade, and the W of all the shading louvers was 0.025m, and B was 0.0175m. In the shaded cases, rotation angle θ of shading louvers was taken as 0°, 45° and 75°, respectively. The incident angle β of approaching airflow was taken as 0°, 90° and 180°, respectively. The non-shaded cases were also performed. The building units are shown in Fig. 1(b).

Numerical settings
The realizable k-ε model was selected for the simulation study. The dimensions of computational domain were selected according to the reference guides [7,8]. with a distance of 5h 0 from the building roof to the top boundary of the computational domain, 5h 0 from the windward facade to the inlet boundary of the computational domain, 10h 0 from the leeward facade to the outlet boundary of the computational domain, and 5h 0 from the lateral facades to the lateral boundaries of the computational domain. The inlet boundary was defined as velocity inlet. The outlet boundary was defined as outflow. The top and lateral surfaces of the computational domain were defined as symmetric boundary. The bottom of the computational domain, the building surfaces and the surfaces of shading louvers were defined as non-slip wall. The velocity profile of atmosphere boundary layer was described by Eq. (1) : where U(z) is the velocity of approaching airflow at height z, m/s. U ref is the reference velocity (equal to 3.41m/s) at building height. α is the parameter (specified as 0.27) to describe the surface roughness caused by the ground features.
On the inlet boundary of the computational domain, the turbulent intensity I, turbulent kinetic energy k, and turbulent dissipation rate ε were defined by Eq. (2)-(4), respectively. The constant C μ is taken as 0.09.
The ventilation capacity is evaluated by air change rate, as shown in Eq. (5) : (5) where ACH is air change rate per hour, h -1 . Q is the ventilation rate, m 3 /s. Vol is the unit volume, m 3 .

Results and discussion
To investigate the single-sided ventilation capacity and air quality, ACH and age of air (AOA) were evaluated in simulation cases with different β and θ. The rotation angle θ was taken as 0°, 45° and 75°, respectively. The incident angle β was taken as 0°, 90° and 180°, respectively. The non-shaded cases were also performed for comparison. Fig. 2 shows the results of ACH in each unit of the multistorey building with the approaching airflow angle β of 0°, 90° and 180°, respectively. In the single-side ventilation with the wind angle of 0°, 90° and 180°, the values of ACH in all conditions range from 0.5h -1 to 19.8h -1 , 0.7h -1 to 15.0h -1 and 0.7h -1 to 10.6h -1 , respectively, and the minimum values of ACH in these conditions are 0.5 to 1.0 h -1 , 0.7h -1 to 2.1h -1 and 0.7h -1 to 1.2h -1 , respectively. The range of ACH for β=0° is the largest, while the range of ACH for β=180° is the smallest. In all shaded conditions, the proportion for units with ACH not less than 2h -1 is above 70%. In the cases for β=0°, the units with the risk of too low ACH are usually located in the leeward column of the fourth and fifth floors. In the cases for β=90°, the units with the risk of too low ACH are usually located in the windward column of the third and fourth floors. In the cases for β=180°, the units with the risk of too low ACH are usually located in the windward column of the fourth and fifth floors. As shown in Fig. 2, for a whole building with singleside ventilation, ACH in the shaded condition increases in most units compared to the non-shaded conditions and decreases only in some shaded conditions with β=90°. For β=0°, ACH of the building decreases as θ increases in the shaded conditions. ACH of the building increases by 58-67% at different θ compared to the nonshaded conditions. The percentage of the total number of units with decreasing number of ACH decreases and then increases with increasing θ, which is 23%, 13% and 30% for θ=0°, θ=45° and θ=75°, respectively. The results indicate that this percentage changes to a similar extent when the louvers are rotated from medium angle to lower or higher angle. For β=90°, ACH in the whole building increases with increasing θ in the shaded condition. Compared with the non-shaded condition, ACH in the whole building is reduced by 16% and 9% for θ=0° and θ=45°, respectively. The number of units with reduced ACH as a percentage of the total number of units decreases and then increases with increasing θ, which is 70%, 50% and 57% at θ=0°, θ=45° and θ=75°, respectively. The results indicate that this percentage changes slightly when the louvers are adjusted in the range of low θ to medium θ. For β=180°, the effect of shading louvers on ACH is less significant than those for β=0° and β=90°. ACH of the building decreases as θ increases in the shaded conditions. Compared with the non-shaded condition, ACH of the whole building increases by 2-10% when the θ increases from 0° to 75°. The percentage of units with a reduced ACH decreases and then increases with increasing θ, which is 57%, 53% and 73% at θ=0°, 45° and 75°, respectively. The results indicate that this percentage change is more obvious when the louvers is rotated from the medium θ to higher θ. and β=180° when the θ is taken as 45°. For β=0°, the inhomogeneity of local AOA at the pedestrian level in the middle unit of the W column is significantly higher than that in the lateral units. In the middle unit, the air is not fresh near the inner corridor. In the lateral units, the air is not fresh in the vortex near the corner of the facade. In the L columns, the distinction is that the inhomogeneity of local AOA at the pedestrian level in each unit is lower, but the shapes of AOA contours in each unit are similar to those of the corresponding units in the W columns. For β=90°, the inhomogeneity of local AOA at pedestrian level is significantly higher in the units of the W columns than those in the L columns. For the W columns, the inhomogeneity of local AOA at pedestrian level in Unit W1 is significantly higher than those in the other two units. The zones of fresh air in the units of the W columns are located along from the windowed facade to the indoor wall near the leeward facade, while the zones of fresh air in the units of the L columns are located along from the windowed facade to the indoor wall near the windward facade. The results indicates that the presence of shading louvers in this floor significantly changes the dispersion routes of fresh air. For β=180°, the contour shapes of the local AOA in the units of the W columns are similar to those of the L columns at θ=0°, but the contour density is higher. The shading louvers installed near the leeward facade of the higher floors make it more difficult for fresh air to enter the single-sided ventilated units. The air is the freshest for the L columns which do not have shading louvers. Since the middle L2 column is closer to the stagnant zone than the lateral L1 and L3 columns, the local AOA and the inhomogeneity are slightly higher.

Summary
In this paper, the single-sided ventilation capacity and air quality in the multi-story building with and without shading louvers were investigated by the numerical method. The simulation cases were performed for different shaded and wind conditions. The rotation angle of shading louvers ranged from 0° to 75°, and the incident angle of approaching airflow ranged from 0° to 180°. The results show that under the different conditions of single-side ventilation, air change rate per hour in the shaded conditions increase in most units compared to the non-shaded conditions. As the incident angle increases, the variations of age of air in some units have a difference in magnitude. The maximum increases in the age of air in some units are more than 1000s compared to the non-shaded conditions. This study is helpful to shading design and operation settings in naturally ventilated buildings with shading devices.