Natural ventilation potential for gymnasia – Case study of ventilation and comfort in a multisport facility in northeastern United States

Building and Environment 108 (2016) 85e98

Contents lists available at ScienceDirect

Building and Environment journal homepage: www.elsevier.com/locate/buildenv

Natural ventilation potential for gymnasia e Case study of ventilation and comfort in a multisport facility in northeastern United States Zheng Cheng a, *, Lingling Li a, William P. Bahnfleth b a b

School of Architecture, Harbin Institute of Technology, 66 Xidazhi Road, Office130, Harbin 150060, China Department of Architectural Engineering, Pennsylvania State University, 104 Engineering Unit A, University Park, PA 16802, United States

a r t i c l e i n f o

a b s t r a c t

Article history: Received 1 May 2016 Received in revised form 14 August 2016 Accepted 18 August 2016 Available online 21 August 2016

The natural ventilation potential to maintain acceptable indoor air quality (IAQ) and thermal comfort in gymnasia was investigated using a university multisport facility in northeastern United States as a case study building. A parametric modeling study was conducted considering the effects of opening configurations and control strategies during the summer months. The thermal accuracy of the model was verified using field measurements during August 2015. Performance metrics for IAQ and thermal comfort were the percentages of occupied hours during which ventilation rate met or exceeded ASHRAE Standard 62.1e2013 and temperature met adaptive thermal comfort criteria of ASHRAE Standard 55e2013, respectively. Wind direction was found having a major effect on cross ventilation rate. Wind and buoyancy driven forces could complement or oppose each other depending on the wind direction and opening position. Relative to the base case, larger net openings that were more evenly distributed performed better. Rooftop vents improved ventilation performance, particularly under unfavorable wind conditions. With improved opening configurations, the acceptable ventilation hours increased from 21.5% to 99.5% of occupied time for the maximum occupancy. The strictest temperature-controlled strategy had the best thermal performance. Thermal comfort conditions could be maintained during 85.3% of the occupied hours. However, the temperature rule largely shortened the opening operation time, and consequently decreased the acceptable ventilation hours to only 47.1%. Continuously natural ventilation during occupied time gave the longest combined IAQ-thermal acceptable hours, 73.9% of the occupied time, although it moderately decreased the thermal comfort hours to 74.2%. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Natural ventilation Gymnasia Ventilation rate Thermal comfort Opening configuration Control strategy

Contents 1. 2. 3. 4.

5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Literature review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 Case study information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Research method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 4.1. Simulation approach of IES-VE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 4.2. Accuracy verification of IES-VE model based on site-measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 4.3. Design of the simulated base model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 4.4. Performance indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4.4.1. Minimum ventilation rate according to ASHRAE standard 62.1e2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4.4.2. Temperature excess method based on adaptive thermal comfort model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4.5. Parametric study design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.1. Influence of opening configuration on ventilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.2. Ventilation and thermal performance of opening-improved model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

* Corresponding author. E-mail addresses: [email protected] (Z. Cheng), [email protected] (L. Li), wbahnfl[email protected] (W.P. Bahnfleth). http://dx.doi.org/10.1016/j.buildenv.2016.08.019 0360-1323/© 2016 Elsevier Ltd. All rights reserved.

86

6. 7.

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

5.3. Thermal influence of natural ventilation control strategies for gymnasium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 5.4. Combined IAQ-thermal performance of opening control strategies for gymnasium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

1. Introduction Where feasible, natural ventilation is considered to be an attractive solution to ventilate and cool buildings. Well-designed natural ventilation can passively maintain a comfortable and healthy indoor environment and consequently decrease the amount of energy consumed by ventilation and cooling systems [1]. The increasing use of natural ventilation has been widely noted. However, in most large spaces such as gymnasia, hybrid ventilation and air conditioning systems still dominate [2e4]. The research on the natural ventilation performance in gymnasia is deficient. In this research, the feasibility of using natural ventilation for gymnasia was investigated using a multisport facility at a university campus in the northeastern United States as a case study building. The ventilation rate for achieving an acceptable indoor air quality and the thermal comfort were the two key performance metrics considered. A parametric modeling methodology was developed that considered the effects of the opening configurations and opening control strategies. 2. Literature review Natural ventilation is the process of exchanging air between indoor and outdoor environments. The mechanism of natural ventilation mainly depends on the wind effect, thermal buoyancy, or their combination. It is considered more difficult to introduce sufficient winddriven flows in long spaces, primarily because of the smaller pressure difference between the windward and leeward facades and larger indoor resistance compared to shorter spaces [5,6]. It has been suggested that a building's length should be less than five times its ceiling height [7]. Schulze et al. [8] found that the cross ventilation rate largely depended on both the opening arrangement and effective opening area based on the results of airflow network methods. Heiselberg et al. [9] used an experiment method and found that the characteristics of the air movement through different opening types differed greatly. Kang and Lee [10] tested a louver combination that aligned the angle of the outer louver blades with the oncoming wind and elongated the inner louver blades to guide the entrained air down to the ground surface and found that it was effective at pushing the stagnant flow inside a long factory building. Leea et al. [11] investigated horizontal and vertical shading louvers with a 0
angle and found that they contributed to a greater vertical flow convection than those with 30
, 60
, and 90
angles. Using computational fluid dynamics (CFD) simulation [12,13] and wind tunnel methods [14], the position of the inlet opening was found to predominantly affect the ventilation rate of cross-ventilated buildings, whereas the impact of the vertical position of the outlet opening was relatively small. Cui et al. [15] suggested that larger window should be placed in the dominant wind direction. Similarly, Tantasavasdi et al. [16] found that a larger inlet was more helpful than a larger outlet for cross

ventilated buildings. However, Peren et al. [17] found that the ventilation flow rates were significantly higher with a lower inletoutlet opening ratio by evaluating the volume flow rates in doublespan long spaces with leeward sawtooth roof. It might be possible to use the buoyancy force to complement an insufficient wind force. With a high ceiling, the buoyancy force in a gymnasium could be more significant than in buildings with lower ceilings. ASHRAE Standard 62.1e2013 specifies that in across naturally ventilated building, the area of the openings that connect directly to the outdoors should be a minimum of 4% of the net occupiable floor area [7]. However, Lin and Chuah [18] indicated that the thermal buoyancy in a large space with a ceiling higher than 6 m and an opening-to-floor ratio greater than 0.9% can introduce adequate fresh air to satisfy the indoor air quality (IAQ) requirement. Bartzanas et al. [19] indicated that the combination of roof and side openings provides better air exchange and cooling performance than any split cases. Hunt and Linden [20] highlighted that the vertical relationship between leeward and windward openings was a major factor determining the combined effect of ventilation driven forces. Stavridou and Prinos [21] observed that the combined ventilation efficiency increased with the vertical distance between the midpoints of a low inlet and high outlet. Natural ventilation can be used for cooling a building's interior whenever the outdoor temperature is lower than the indoor temperature. Occupants can adapt to a broader temperature range in human-controlled naturally ventilated spaces [22]. The cooling potential of natural ventilation has been evaluated under multiple climate conditions [23e26]. Nighttime ventilation provides a significant pre-cooling benefit by storing cooling energy in a building's thermal mass at night and preventing temperature climb during occupied time [27e30]. In addition, the local climate, thermal properties of a building's construction, and internal heat gain have obvious influences on the cooling performance of natural ventilation [31], along with the opening configurations [32,33]. Although shading devices are helpful to prevent excess solar gain in a room [34], they are also obstacles to airflow movement [11]. The occupants' behavior has been found to have stronger effects on both the ventilation and thermal performance than factors related to the building [35e37]. Fabi et al. [38] showed that the physical environmental variables that impact a window opening behavior include the indoor and outdoor temperatures, solar radiation, wind speed, and CO2 concentration. Yin et al. [39] investigated the natural ventilation potential in China using a temperature-control opening strategy, while Schulze et al. [8] compared the influences of the thermal and IAQ priority opening behaviors and found that temperature rules could effectively prevent overcooling. G. van Moeseke et al. [40] designed three control modes for daytime natural ventilation for the Belgian weather and found flow rate modulation on external temperature was an efficient way to manage both overheating and overcooling problem.

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

3. Case study information The case study gymnasium was a university indoor multisport facility located at State College, PA, which is in the northeastern part of the US mainland at 40
47’N 77
51’W. It consists of two main parts: a large sports space and auxiliary rooms at the northwest corner. Only the large sports space was considered in this study, which includes a 200 m running track, multi-function turf field, and 800-seat permanent stand (Fig. 1a and b). This large space is 63.2 m in depth and 147.8 m in length, with a maximum height of 14.3 m. It has the typical construction features found in the northeastern United States, as presented in Table 1. This building is located on a gently sloped lawn at the edge of the university campus. There are no significant wind obstructions or external shading sources. The sports space in the case study building has no cooling system in summer and depends exclusively on a hybrid ventilation system to adjust the indoor thermal environment and IAQ. Air enters the space through 17 openings on the western façade (Fig. 1c), which include 4 openings in the corridor room on the first floor, 11 openings in the back of the spectator stand on the second floor, 2 openings in the exercise room on the second floor. All the openings on the west façade are covered by combined stationaryoperable louvers and perforated plates. The exhaust exits through 19 fans which are evenly located at the upper level of the eastern façade (Fig. 1d). The ventilation rating of each fan is 25,500 m3/h (15,000 cfm). The fans and operable louvers are manually controlled in an asynchronous manner. The case study building hosts a variety of activities, and its occupancy schedule is highly diverse. The typical occupancy is summarized in Fig. 2 based on the schedule records for 2014 and 2015. During weekdays, the building is open from 8:00 through 17:00 mainly for public exercise and track team training activities. Typical occupancy during weekdays varies from month to month, with 50, 150, 200 occupants in June, July, August respectively. On weekends, the sports hall is occupied by competition events. The occupancy varies from 500 to 1700 depending on the scale of activities. In order to simplify the model, the extreme situation in the building with full occupancy is chosen to be the typical condition during weekend. The thermal influence of occupancy was evaluated. The

87

indoor air temperature in the sports hall with 1700 occupants is averagely 0.41
C higher than that with 500 occupants. In addition, the heat gains from the occupants vary with different activity intensities ranging from sedentary spectators to competing athletes. Therefore, three typical activity intensities were used: active athletes, spectators, and others, including resting athletes and coaches. The humidity impact on the indoor environment was not investigated in this research so the latent heat gain from the human body is outside the scope of consideration. Table 2 only lists the indoor sensible heat gains [41]. State College, the location of the case study building, is in climate zone 5 according to the IECC climate zone definitions [42]. Fig. 3 shows the average outdoor air temperature range at State College based on the TMY3 data [42]. Historically, the coldest month is January, with an average temperature of 2.3
C, and the hottest month is July, with an average temperature of 22
C. In the winter and transition seasons, the average outdoor air temperature is well below the comfortable temperature range. The cooling load in this period is very low. Therefore, this research focused on the natural ventilation potential of a gymnasium in the summer months. The dominant wind in the summer is from the west, with an average wind speed of approximately 6.1 km/h (2.2 m/s), as shown in Fig. 4. 4. Research method 4.1. Simulation approach of IES-VE The commercial simulation software IES-VE was used for the thermal and airflow modeling of the case study building. MacroFlo is the airflow simulation program in IES-VE. The flow through each opening was calculated as a function of the characteristics of the openings and the imposed pressure difference. In MacroFlo, openings are considered to be sharp-edged orifices through which the airflow rate qn (m3/s) can be calculated using Eq. (1) [43]:

qn ¼ Cd Aop

qffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2DP=r

Fig. 1. Architectural drawings and façade appearance of case study building.

(1)

88

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

Table 1 Characteristics of case study space. 1999

Dimensions Orientation External wall Internal wall Higher roof Lower roof Glazing

63.2 m in depth, 147.8 m in length, and maximum height of 14.3 m North-south with 17
rotation to the west 101 mm brick facing, 50 mm of insulation, 50 mm air cavity, 203 mm of lightweight concrete block, U value: 0.44 W/m2$K 203 mm lightweight concrete block plastered on both sides, U value: 0.97 W/m2$K Manufactured metal roof, 150 mm of insulation, metal deck, U value: 0.93 W/m2$K Membrane roofing, 150 mm of insulation, metal deck, U value: 0.91 W/m2$K Transparent glass, U value: 1.95 W/m2$K, visible light transmittance: 0.76

Number of occupants

Construction time

1800 1500 1200 900 600 300 0

Weekday-June

Weekday-July

Weekday- August

Weekend

Fig. 2. Typical occupancy in summer.

Table 2 Sensible heat gains in case study of gymnasium [41]. Items

Sensible heat gains

Spectators Active athletes Other occupants Lighting

65 W/person 210 W/person 75 W/person 15 W/m2

where Cd is the discharge coefficient. In IES-VE, discharge coefficient is preset to be 0.62 for openings that are small in relation to the adjacent spaces; DP (Pa) is the pressure difference across the opening, which includes the wind and buoyancy pressure for natural ventilation; r (kg/m3) is the density of the air entering the opening; and Aop(m2) is the net open area of the orifice, which is determined using the equivalent area ratio and opening percentage. The opening percentage is specified by the control profiles in IES-VE, which could be set based on time or physical environmental variables such as the temperature and CO2 concentration. The equivalent area is defined as a measure of the aerodynamic performance of a ventilator. It is the area of a sharp-edged circular orifice through which air would pass at the same volume flow rate,

under an identical applied pressure difference, as the opening under consideration [42]. It can be calculated using Eq. (2) [44]:

Aeq ¼ Cd Af

. Cdo

(2)

where Cdo is the discharge coefficient of a standard circular sharpedged orifice, which is equal to 0.61; Cd¼0.62; and Af is the free area, which is defined as the geometric open area of a ventilator [41]. Based on Eq. (2), the equivalent area ratios of several typical opening types are presented in Table 3 [45,46]. With the same gross opening size, the equivalent area of an opening is crucially determined by operable constructions. In addition, for the infiltration ventilation rate through cracks qc (m3/s), Eq. (3) is used in IES-VE:

. 0:5  qc ¼ CL rref r Dp0:6

(3)

where C is the crack flow coefficient (m3s1m1Pa1), for which a moderate value of 0.15 was used; L is the length of the crack (m); r is the density of the air entering the crack (kg/m3); rref is a reference

Fig. 3. Outdoor air temperature analysis for State College, PA.

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

89

Fig. 4. Summer wind rose for State College, PA.

Table 3 Free area ratios and equivalent area ratios of typical openings. Opening types

Fixed windows

Awning window

Sliding window

Casement window

Combined louver window

Adjustable louver window

Free area ratio [46] Equivalent area ratio

0% 0%

1% 1%

27% 27%

35% 35%

47% [47] 48%

82% 83%

air density of 1.21 kg/m3; and DP is the pressure difference across the crack (Pa). In this research, the infiltration ventilation rate was not taken into account when calculating the intentional ventilation rate, but the thermal influence of the infiltration was considered. ApacheSim is the dynamic thermal simulation program in IESVE, which is based on the first-principles mathematical modeling of the heat transfer processes occurring within and around a building [47]. It uses a stirred tank model of the air in a room. The calculations are based on the concepts of the bulk air temperature and humidity, which are assumed to be uniform within a zone. Within ApacheSim, the conduction, convection, and radiation heat transfer processes for each element of the building fabric are individually modeled. The air exchange data are dynamically exchanged between ApacheSim and MacroFlo to achieve the simultaneous solutions of the inter-dependent thermal and air flow balances. Weather data can be imported as the boundary condition by a program called APLocate, which is the weather and site location editor for IES-VE. In this research, TMY3 file was used to analyze the natural ventilation potential for the gymnasium under the typical annual weather conditions of the case study location. 4.2. Accuracy verification of IES-VE model based on sitemeasurement In order to verify the accuracy of the IES model for the gymnasium, the simulation results for the actual hybrid ventilated

gymnasium with and without fan operation were compared with field measurement data. The weather data during the measurement period were recorded at weather station KPASTATE17 located approximately 1700 m southwest of the facility, with the exception of the data for the cloud cover/visibility, which were taken from the nearest ISD station at University Park airport, 5000 m north of the building. The schedule for the occupancy and indoor equipment operation, i.e., fans and lighting, was recorded when the building was open during the measurement period. At night and on weekends, the building was unoccupied, and all of the equipment was supposed to be turned off. The building schedule during the measurement period is listed in Table 4. Field measurements were conducted from the afternoon of August 7, 2015, through the morning of August 14, 2015, at nine locations (Figs. 5 and 6). A vertical array of sensors (L2.1-L2.4) was located at the plane center of the case study building with heights from 2.1 m (7 ft) to 13.1 m (43 ft) above the sports field. The heights of L1, L3, and L6 were all 5.8 m (19 ft) above the sports field to minimize the risk of damage as a result of sports activities. Measurement location L4 was above the last row of spectator seats at the same height as the center point of an adjacent louvered opening. L5 was located at the opening that connected the corridor and sports field. It was an important pathway for air movement between the outdoors, corridor, and sports space. Placed at a height of 2.1 m (7 ft), the sensors at L5 and L2.1 were both well protected by guard bars. HOBO data loggers were used to record the indoor air

Table 4 People and equipment schedules during measurement period. Date

08/08

08/09

08/10

08/11

08/12

08/13

People Fans and louvers Lighting

0 OFF OFF

0 OFF OFF

4 7:00e14:30 8:00e16:30

4 7:00e15:30 8:00e16:30

4 8:50e16:20 8:00e16:30

10 OFF 8:00e16:30

90

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

Fig. 5. Plan view of measurement locations.

Fig. 6. Section view of measurement locations.

temperatures. Table 5 lists the characteristics of these devices. Fig. 7 shows the measured and simulated air temperatures over the six-day measurement period. The measured temperatures were the average values recorded at the nine positions. When fans were continuously closed on Aug.8, Aug.9 and Aug.13, the thermal insulation properties of the building's construction materials were help to provide more stable indoor air temperatures than the outdoor temperatures. Heat storage by the thermal mass during the daytime was capable of preventing an extreme temperature drop at night. Both the simulated and measured data managed to verify that. During this period, the simulated and measured temperatures were generally close with a maximum discrepancy of 1.07
C occurring on the early morning of Aug. 9. When the fans and louver windows were open during August 10e12, the indoor and outdoor temperatures tended to be close as a result of air convection. A constant volume of 7650 m3/h for each fan (30% of the rated value) was used in the model to account for possible performance degradation because they were installed and put into use since 1999. The average discrepancy during the fans-on period was 0.41
C. On the night of Aug. 11, the measured temperature dipped twice following a decrease in the outdoor air temperature, which led to severe disagreements. A possible cause of this discrepancy was related to the unscheduled operation of the building equipment during that period. Both the administrative staff and tracking team have the access of the building. The staff closed the building at 17:00 of Aug.11. Public and researchers couldn't stay in the building after that. But it was possible that the tracking team was back late

that night and opened the fans or doors. Having doors open or turning fans on would bring in cooler outdoor air, which could result in a decrease in the indoor air temperature. A comparison of the predicted and measured temperatures using the commonly used thermal comfort criterion of “temperature excess hours” [48] showed that the percentage of error was only 3.5%. In general, the IES model's temperature prediction was sufficiently accurate on both a real-time scale and when considering thermal comfort criteria for a parametric study which mainly focused on the performance differences between simulated scenarios. 4.3. Design of the simulated base model In order to investigate the natural ventilation potential in the gymnasium, a simulation model was constructed by converting the existing hybrid ventilation system to natural ventilation. This was accomplished by changing the vents in which the exhaust fans are actually mounted into openable windows. An equivalent area ratio of 20% was used to define the ventilation efficiency of openings in the base model. The ventilation system comparison between the existing building and the simulated model is presented in Table 6. Due to the opposite openings with different heights, natural ventilation in the simulated model would be driven by the combined effect of wind and buoyancy force. In addition, the object building space was divided into three zones in the IES model: 1) the sports field and permanent spectator

Table 5 Measurement device information. Measurement devices Onset HOBO Temp/Rh/2 Ext Channels Data Logger

Recorded data Air temperature

Range


20 to 0 C 0 to 50
C

Accuracy

Measurement interval

e ±0.35
C

1 min

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

91

Fig. 7. Comparison of simulated indoor air temperature and measured results (August 8e13, 2015).

seats, 2) the exercise room on the second floor, and 3) the corridor underneath the spectator seats. Both the exercise room and corridor had openings in the west sidewall and large holes directly connected to the sports field. Therefore, the ventilation rate reported in this research was the sum of the three zones, while thermal comfort was assessed only for the sports field zone. 4.4. Performance indicators 4.4.1. Minimum ventilation rate according to ASHRAE standard 62.1e2013 In ASHRAE Standard 62.1e2013, the ventilation rate procedure specifies minimum ventilation rates, which are intended to provide acceptable indoor air quality to human occupants and minimize adverse health effects [7]. Using this procedure, the outdoor air intake rates were determined based on the space type, occupancy level, and floor area. The minimum ventilation rates for the case study building in accordance with the standard are presented in Table 7. For the base model, the minimum ventilation rate varied with the occupancy, from a maximum of 14,813 L/s to a minimum of 8233 L/s. The IAQ performance was evaluated by calculating the hours exceeding the minimum ventilation requirements. 4.4.2. Temperature excess method based on adaptive thermal comfort model The thermal comfort was assessed on the basis of the adaptive thermal comfort model defined in ASHRAE Standard 55e2015 [49]. This model is applied to occupant-controlled naturally conditioned spaces in which there is no mechanical cooling system and occupants are free to adapt their clothing to the indoor thermal conditions. In this research, spectators were chosen to be representative occupants because they are the main subpopulation in a typical gymnasium during an event. The metabolic rates of spectators are moderate, at 1.1e1.3 met. The performance relative to the standard was assessed by totaling the hours within and outside the comfort range.

The operative temperature is the key variable of the adaptive thermal comfort model. It is defined as the uniform temperature of an imaginary black enclosure and the air within it in which an occupant would exchange the same amount of heat by radiation plus convection as in the actual non-uniform environment [50]. In this research, the operative temperature was approximated by the average of the air temperature (Ta) and the mean radiant temperature (Tr) (see Eq. (4)). This common assumption is of acceptable accuracy for representative occupants engaged in near-sedentary physical activity (with metabolic rates of 1.0e1.4 met), that are not exposed to direct sunlight, where the air velocity is no greater than 0.2 m/s, and the difference between the mean radiant temperature and average air temperature is < 4
C.

To ¼ ðTa þ Tr Þ=2

(4)

In the adaptive thermal comfort model, the allowable operative room temperature range could be calculated using Eq. (5) and Eq. (6):

Tmax;80% ð
CÞ ¼ 0:31TpmaðoutÞ þ 21:3

(5)

Tmin;80% ð
CÞ ¼ 0:31TpmaðoutÞ þ 14:3

(6)

where TpmaðoutÞ is the prevailing mean outdoor air temperature, which could be obtained from the typical meteorological year (TMY) weather file; and Tmax,80% and Tmin,80% are the upper and lower 80% thermally acceptability limits, respectively. According to the TMY3 data at the case study location, the temperature range of the 80% thermal acceptability in the summer months is listed in Table 8. 4.5. Parametric study design This research evaluated the IAQ and thermal performance of the naturally ventilated gymnasium in three stages.

Table 6 Ventilation system comparison between the existing building and the simulated base model.

Ventilation type Ventilation devices Opening distribution

Opening size

East West East West East West

Actual building

Simulated base model

Hybrid ventilation Exhaust fans Combined stationary-operable louvers and perforated plates 19 high openings close to the ceiling, height of midpoints: 11 m 4 openings on the first floor, height of midpoints: 1 m 13 openings on the second floor, height of midpoints: 5 m Total 72.5 m2 and about 3.8 m2 each Total 97.6 m2 and about 5.5 m2 each

Natural cross ventilation Openings with equivalent area ratio of 20%

92

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

Table 7 Minimum breathing zone ventilation rate according to ASHRAE Standard 62.1e2013 [7]. Building space

Gymnasium (play area) Gymnasium (spectator area)

Ra (L/s$m2)

Rp (L/s$per)

10 3.8

0.9 0.3

In the first stage, the influence of the opening configuration on the air exchange rate was investigated. The base gymnasium model served as the reference case, with the previously noted modifications. Parametric cases were created by varying three factors: the equivalent area ratio, opening-to-floor ratio, and opening distribution, as listed in Table 9. The equivalent area ratios investigated in this research varied from 20% to 80%, related to the ventilation efficiency of the different opening types. The opening-to-floor ratio defines the gross opening size of the entire building. Based on the requirement of ASHRAE 62.1e2013 for a cross-ventilated building [7], the maximum area ratio in this research was set to be 4%, compared with 2% for the base case. Then, the opening distribution was changed by modifying the opening sizes in different positions: with the same total opening area, the sidewall openings were enlarged by extending the width or new rooftop vents were considered to enhance the buoyancy-driven natural ventilation. The opening configurations with the best ventilation performances in the three groups were combined together. The ventilation and thermal performances of the opening-improved model was investigated at a later stage. Finally, five opening control strategies, as listed in Table 10, were applied, with constant natural ventilation as the reference. The control strategies were based on different time and temperature criteria. Control strategy 1 was IAQ oriented: once the building was occupied (8:00e17:00), the windows would be open to maintain good indoor air quality. In order to maintain the indoor temperature within the temperature comfort zone, two different temperature criteria were designed. The first criterion, used for control strategies 2, 3, and 4, allowed outdoor air to enter the space if its temperature was in the comfort zone, even when it was above the current room temperature. It was applied to balance the IAQ and thermal comfort demands, which followed different nighttime schedules. The second temperature criterion, used for strategy 5, was completely thermally oriented: outdoor air was admitted only when it had the potential to cool the building. 5. Results 5.1. Influence of opening configuration on ventilation Increasing both the equivalent area ratio and opening-to-floor ratio improved the ventilation by increasing the total amount of the net opening area. The net opening area was defined as the product of the gross opening size and equivalent area ratio in this research. Fig. 8a and b shows that with a 60% increase in the equivalent area ratio, the acceptable full-occupancy (1700 occupants) ventilated hours rose from around 20% to more than 90%.

Floor area (m2)

8343.4 746.7

Occupancy range

50e100 0-1600

Minimum ventilation rate (L/s) Min

Max

8233

14,813

With double-width openings, maintaining the base equivalent area ratio at 20%, the acceptable ventilated hours increased to about 60%. Very similar ventilation rates were obtained for cases with the same net opening-to-floor ratio (0.006) achieved by different combinations of equivalent area ratios and overall opening sizes. This suggested that increasing the net opening area was the truly crucial reason for the improvement in the ventilation performance. However, the feasible range of the two factors must be considered. The maximum equivalent area ratio of openings could reach approximately 80% with an advanced louver design. However, the practical range of the opening-to-floor ratio is smaller. For example, the maximum opening-to-floor ratio of the case study building is around 50% based on the ratio of the walls to floor area. The four cases shown in Fig. 8c increase the net opening-to-floor ratio to 0.006, from 0.004 of the base case value, by enlarging the opening areas in diverse positions. A comparison of the three cases with only sidewall openings shows that the case that simultaneously enlarges the east and west openings works the best while the ventilation improvement of the case which enlarges only west sidewall openings is poorest. Furthermore, Fig. 9a shows that enlarging only the east openings always provides a higher ventilation rate than enlarging the west openings, regardless of the wind direction. Table 11 presents a sensitivity analysis to explain this phenomenon. With the same total net opening size, when the total areas of the east and west openings size are closer, the ventilation performance is better. The uniformity of the sidewall opening distribution has a considerable impact on the cross-ventilation performance. In addition, Fig. 9a shows that the ventilation rates of the four cases are all higher when the wind comes from west or east relative to that when wind comes from south or north. Rooftop vents are helpful to maintain a more stable ventilation performance under different wind directions. But the fluctuation is still considerable. The wind direction has a major effect on the ventilation rate for the building with openings in opposite sidewalls. Furthermore, although the total areas of east openings are all larger than that of west in the four cases, Fig. 9a shows west wind is always more helpful to increase the ventilation rate than the east when wind speed is the same. According to the reviewed researches [15,16], larger inlets are more effective to improve the cross ventilation performance than larger outlets. But in this research, the performance of combined buoyancy and wind driven ventilation is more complicated. When the wind comes from the east, the windward openings locate higher than the leeward ones. The buoyancy effect offsets the wind driven force, which results in the poor performance under east wind condition. Rooftop vents can ease the conflict between the wind-driven force and buoyancy effect, the

Table 8 Comfortable operative temperature range at State College (
C). Month Prevailing mean outdoor air temperature (Tpma(out))

80% acceptable operative temperature upper limit (Tmax,80%)

80% acceptable operative temperature lower limit (Tmin,80%)

June July Aug

27.8 28.1 27.4

20.8 21.2 20.5

20.9 22.0 19.5

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

93

Table 9 Opening configuration cases. Cases

Equivalent area ratio

Opening-floor ratio

Opening distribution

Reference case Case group 1

20% 30% 80% 20% 20% 20% 20% 20% 20%

2% 2% 2% 3% 4% 3% 3% 3% 3%

Original sizes and positions of west and east openings Original sizes and positions of west and east openings Original sizes and positions of west and east openings Extend width of all east and west sidewall openings Extend width of all east and west sidewall openings Extend width of only west sidewall openings Extend width of only east sidewall openings Extend width of all east and west sidewall openings Add rooftop vents

Case group 2 Case group 3

Table 10 Opening control strategies. Control strategies

Time

Control criteria

Reference Strategy 1

24 h 8:00e17:00 Night 8:00e17:00 Night 8:00e17:00 Night 24 h 24 h

Windows always open Windows always open Windows always closed When Tout < Tmax,80% & Tin Windows always closed When Tout < Tmax,80% & Tin Windows always open When Tout < Tmax,80% & Tin When Tout < Tmax,80% & Tin

Strategy 2 Strategy 3 Strategy 4 Strategy 5

> Tmin,80%, open windows > Tmin,80%, open windows > Tmin,80%, open windows > Tmin,80%& Tout
Fig. 8. Comparison of acceptable ventilation time with different opening configurations: a) effect of equivalent area ratio with opening-floor ratio of 2%, b) effect of opening-floor ratio with equivalent area ratio of 20%, c) effect of opening distributions with net opening-floor ratio of 0.006.

ventilation performance of which is more uniform under different wind directions. Moreover, in order to investigate the magnitude of buoyancy effect, Fig. 9c evaluates the airflow movement under a windless condition. It shows that the configuration with rooftop vents consistently has the best ventilation rate, along with an increase in the indoor-outdoor air temperature difference. Rooftop vents also perform better than larger sidewall windows under low wind speeds, as shown in Fig. 9b. Rooftop vents are more efficient at promoting a buoyancy effect for this 14.3-m-high gymnasium than only sidewall opening with height differences.

5.2. Ventilation and thermal performance of opening-improved model Based on the preceding analysis, the opening architecture of the base model was improved in three aspects, as listed in Table 12. Fig. 10 compares the ventilation performance of the base case with that of the improved opening configurations relative to the ASHRAE Standard 62.1 minimum ventilation requirements for low and high occupancy. The percentage of hours that exceeded the 1700-occupant minimum ventilation requirement increased from 21.5% to 99.5%, while the hours at or above

94

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

Fig. 9. Effects of wind and buoyancy effect on ventilation performance: a) effect of wind direction with V ¼ 3 m/s and 2
C
Table 11 Sensitivity analysis of cross ventilation performance to opening distribution. Cases

Enlarge west openings Enlarge east openings Enlarge both west and east openings

Individual opening area

Total opening area

Percentage of occupied hours

West: East

West: East

Exceed 50-occupant min ventilation rate limit

Exceed 1700-occupant min ventilation rate limit

2.68:1 0.66:1 1.43:1

1:1.62 1:1.38 1:1.25

81.5% 81.9% 87.2%

24.8% 42.5% 47.6%

the 50-occupant requirement reached 99.9% for the openingimproved case. The influence of the opening modification on the indoor thermal comfort was investigated with no ventilation (i.e., all ventilation openings closed) and continuous ventilation (all openings opened 24 h a day). Three models were simulated: the base model, opening-improved model, and improved model with 0.6-m-deep external shading boards, which were installed along the top edges of the side openings. For the opening-improved gymnasium model, the increased solar heat gain by the larger glazing openings led to 21 additional over-heated hours (out of 828 in total) compared to the base case, as shown in Fig. 11. The operation of the shading devices was helpful to shorten the over-heated time by 14 h, but the cooling potential of natural ventilation was more significant. With constant natural ventilation, the over-heated hours for both the base and improved model were significantly decreased by 44 h and 85 h, respectively. The cooling efficiency of the opening-improved case was much higher than that of the base case despite more solar heat being obtained from the larger openings. However, the over-cooling problem of natural ventilation was crucial, especially for the opening-improved model which could introduce more ambient air than the base one when it was cold outside. The overcooled time of the improved case was dramatically lengthened, and

consequently led to 68 h shorter thermally comfortable time than the base model. Active control strategies to address the overcooling problem were further investigated. 5.3. Thermal influence of natural ventilation control strategies for gymnasium The five control strategies listed in Table 10 were simulated to determine their effect on the indoor thermal condition of the opening-improved model. Fig. 12 shows that all of the cases with controlled natural ventilation had longer acceptable thermal comfort times (by 32e124 h) than the reference case. The thermal improvement of the temperature control strategies was superior to the scheduled strategy, which only considered the indoor occupancy. By applying continuous nighttime ventilation, the overheated time of case 3 was half that of case 2. The cooling potential of nighttime ventilation is very effective. However, it also results in serious over-cooling, which is detrimental to comfort. Temperature-based control rules can largely overcome the overcooling problem. The over-cooled time of case 4 with controlled nighttime ventilation was 69 h shorter than that of case 3, and that of case 2 was also 42 h shorter than that of case 1 because of controlled daytime ventilation. Case 4 and case 5, which used

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

95

Table 12 Opening strategy of improved ventilation case.

Base model Opening-improved model

Equivalent area ratio

Opening-floor ratio

Opening distribution

20% 80%

2% 4%

Original east and west sidewall windows Add rooftop vent and extend the width of west and east openings in meantime

Fig. 10. The contrast of ventilation rate for base case and opening-improved case.

Fig. 11. Thermal comfort hours of base model and opening-improved model in summer.

temperature-control rules all day, had better thermal performances than those cases with part-time control. Openings would stay close when ambient air temperature was out of 80% acceptable temperature range. Either too hot or too cold ambient air couldn't enter the room. Both the over-heating and over-cooling problems were well addressed. Case 5 which utilized the strictest temperaturebased rule had the best thermal performance among the five cases. Not only the desired temperature range but also the difference between the indoor and outdoor temperatures was considered. A total of 85.3% of the occupied time in the case study gymnasium could be thermally comfortable relative to only 70.3% for the reference case. 5.4. Combined IAQ-thermal performance of opening control strategies for gymnasium The opening control strategies significantly affect the ventilation rate of a naturally ventilated building, as well as the indoor thermal comfort. The combined IAQ-thermal acceptable hours, during which both the minimum acceptable ventilation rate and

the 80% acceptable thermal requirements could be satisfied, were calculated. Based on the previous analysis, this investigation was conducted on control strategies 1, 4, and 5, which have different orientations: acceptable indoor air quality or thermal comfort. Fig. 13 clearly shows that these strategies involve trade-offs. Although case 5 has the best thermal performance, its acceptable ventilation time is strongly limited by the strictest temperaturecontrol rule. As a result, case 5 has the shortest combined acceptable time: only 46.6% of the occupied hours assuming full occupancy. The temperature-control rule of case 4 is slightly looser than that of case 5, which allows comfortable but hotter ambient air to enter the indoor environment. However, the cooling performance of case 4 is still significant, with an over-heated time only 2 h longer than case 5, while the acceptable ventilating time was 204 h longer. The combined 1700-occupant acceptable hours of case 4 were eventually 71.3% of the occupied time, 204 h longer than case 5. In contrast, while moderately reducing the thermal comfort hours, case 1, which had continuously natural ventilation during occupied time, gave the best combined IAQ-thermal performance, with 73.9% of the occupied hours assuming full occupancy. Thus, although

96

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

Fig. 12. Thermal comfort hours for opening-improved case with controlled natural ventilation in summer.

temperature-based control strategies are effective to improve the indoor thermal performance, their restriction on the ventilation performance is also considerable. This conflict needs more effort to balance. 6. Discussion In this research, the natural ventilation potential to maintain acceptable air quality and thermal comfort in gymnasia has been investigated using a university multisport facility in northeastern United States as a case study building. First, the ventilation influence of three opening parameters has been evaluated: equivalent area ratio, gross opening size and opening position. According to the evaluation results, an opening-improved model has been built up for the case study gymnasium. Then, the thermal influence of opening improvements has been studied in two aspects: heat gain and natural ventilation cooling potential. Five different control strategies have been designed and utilized to address the overcooling problem of natural ventilation. Finally, the combined impacts of control strategies have been analyzed. It is important to mention the limitations of the current study, which should be addressed in future research: 1) In this research, only the indoor air temperature prediction of the model (not the air flow rates or humidity) was validated using site measurements. The accuracy of the ventilation predictions was not verified, but the match with the thermal results implied that the modeling of the ventilation rates was acceptably accurate. As noted, the simulated indoor air temperatures were commonly higher than the measured ones. This might

have caused the results to be biased toward fewer overcooled hours and more overheated hours. 2) In the gymnasium case study, there were height differences between the sidewall openings in the west and east facades, which could introduce a buoyancy effect. Therefore, the ventilation rates under the same wind condition would vary because of the indoor-outdoor air temperature differences, as shown in Fig. 9a and b. Although the average R-value of the regressed curves in Fig. 9a is relatively low, the qualitative differences in the results between four different opening distributions are clear and informative. 3) This research was a preliminary effort to evaluate the natural ventilation performance in a gymnasium by assuming the space has a uniform indoor environment. However, the detailed distribution of the indoor air temperature and air flow needs more precise study for large spaces. The CFD simulation method is necessary to ensure an appropriate fresh air distribution. 4) This research focused on the feasibility of maintaining acceptable indoor air quality and thermal comfort in naturally ventilated gymnasium. However, the energy saving potential of natural ventilation utilized in gymnasia is also a significant issue. It could be the evaluation criteria to analyze the trade-offs of opening control strategies. It is very necessary to conduct further studies in order to advance our understanding of that.

7. Conclusions Based on the reported analysis, the following conclusions can be drawn:

Fig. 13. Combined IAQ-thermal performance of cases with control strategies 1, 4 and 5.

Z. Cheng et al. / Building and Environment 108 (2016) 85e98

1) The potential for natural ventilation to maintain an acceptable indoor air quality and thermal comfort in gymnasia may be significant. After improving the opening configurations, the fulloccupant acceptable ventilation time of the target gymnasium could increase from 21.5% to 99.5%. With the best temperaturecontrol rules, an 80% acceptable thermal comfort condition could be maintained during 85.3% of the occupied time, which was 11.7% longer than the base case. 2) For the ventilation performance of gymnasia, the net opening size is a crucial factor determining the air flow rate of natural ventilation. It can be modified by changing the opening construction or adjusting the gross size of openings. 3) The uniformity of the opening distribution positively affects the cross-ventilation efficiency when a building has multiple inlets and outlets in opposite sidewalls. 4) Buoyancy effect can complement or oppose the wind-driven force depending on the wind direction and opening position. The buoyancy effect introduced by rooftop vents in this 14.3-mhigh gymnasium was significant to promote air exchange under disadvantageous wind conditions, e.g., a low wind speed and an adverse wind direction. 5) For the thermal performance of gymnasia, shading devices are helpful to prevent excessive solar heat gain. The cooling performance of natural ventilation is more efficient in a gymnasium with an improved opening configuration compared to the base model. 6) The thermal influence of the opening control strategy is much more significant than changing the building architecture if an adequate amount of opening area is provided. Both the overheating and over-cooling problem of the gymnasium was largely relieved by continuously temperature-controlled natural ventilation in this research. 7) The strictest temperature-control strategy had the best thermal performance in this research. But the acceptable ventilation hours decreased to 47.1% and consequently led to the shortest combined IAQ-thermal acceptable time, only 46.6% of the occupied hours assuming full occupancy. The IAQ oriented control strategy, continuously natural ventilation during occupied time, gave the longest combined acceptable hours, 73.9% of the occupied time, although it moderately decreased the thermal comfort hours to 74.2%. More attention is needed to achieve the IAQ-thermal balance of controlled natural ventilation. Acknowledgements This work was conducted during a visiting doctoral scholar program at Pennsylvania State University supervised by Dr. William P. Bahnfleth. It was supported by the China Scholarship Council (grant number 201406120202). The assistance of Mr. Paul Kremer, Mr. Glenn Lelko, and Mrs. Dayna Wenger during the site measurements is greatly appreciated. Nomenclature qn Aop Cd DP

r

Aeq Af Cdo qc C

ventilation flow rate (m3/s) net open area of the opening (m2) discharge coefficient () pressure difference across the opening (Pa) air density (kg/m3) equivalent area of the opening (m2) free area of the opening (m2) discharge coefficient of a standard circular sharp-edged orifice () infiltration ventilation rate (m3/s) crack flow coefficient (m3s1m1Pa1)

L

rref Rp Ra To Ta Tr Tmax;80% Tmin;80% TpmaðoutÞ Tout Tin

97

crack length (m) reference air density (kg/m3) outdoor airflow rate required per person (L/s$per) outdoor airflow rate required per unit area (L/s$m2) operative temperature (
C) air temperature (
C) mean radiant temperature (
C) upper 80% acceptability limit (
C) lower 80% acceptability limit (
C) prevailing mean outdoor air temperature (
C) outdoor air temperature (
C) indoor air temperature (
C)

References [1] T. Yu, P. Heiselberg, B. Lei, M. Pomianowski, C. Zhang, A novel system solution for cooling and ventilation in office buildings: a review of applied technologies and a case study, Energy Build. 90 (2015) 142e155. [2] T. Nishioka, K. Ohtaka, N. Hashimoto, H. Onojima, Measurement and evaluation of the indoor thermal environment in a large domed stadium, Energy Build. 32 (2000) 217e223. [3] C. Huang, Z. Zou, M. Li, et al., Measurements of indoor thermal environment and energy analysis in a large space building in typical seasons, Build. Environ. 42 (2007) 1869e1877. [4] A.I. Stamou, I. Katsiris, A. Schaelin, Evaluation of thermal comfort in galatsi Arena of the olympics “Athens 2004” using a CFD model, Appl. Therm. Eng. 28 (2008) 1206e1215. [5] C. Chu, B. Chiang, Wind-driven cross ventilation in long buildings, Build. Environ. 80 (2014) 150e158. [6] C. Chu, B. Chiang, Wind-driven cross ventilation with internal obstacles, Energy Build. 67 (2013) 201e209. [7] American Society of Heating, Refrigerating and Air-conditioning Engineers, Inc, ANSI/ASHRAE Standard 62.1e2013: Ventilation for Acceptable Indoor Air Quality, Atlanta, USA, 2013. [8] T. Schulze, U. Eicker, Controlled natural ventilation for energy efficient buildings, Energy Build. 56 (2013) 221e232. [9] P. Heiselberg, K. Svidt, P.V. Nielsen, Characteristics of air$ow from open windows, Build. Environ. 36 (2001) 859e869. [10] J. Kang, S. Lee, Improvement of natural ventilation in a large factory building using a louver ventilator, Build. Environ. 43 (2008) 2132e2141. [11] J. Leea, M. Alshayeba, J.D. Chang, A study of shading device configuration on the natural ventilation efficiency and energy performance of a double skin façade, Procedia Eng. 118 (2015) 310e317. [12] H. Shetabivash, Investigation of opening position and shape on the natural cross ventilation, Energy Build. 93 (2015) 1e15. [13] J.I. Peren, T. van Hooff, B.C.C. Leite, B. Blocken, CFD analysis of cross-ventilation of a generic isolated building with asymmetric opening positions: impact of roof angle and opening location, Build. Environ. 85 (2015) 263e276. [14] Y. Tominaga, B. Blocken, Wind tunnel analysis of flow and dispersion in crossventilated isolated buildings: impact of opening positions, J. Wind Eng. Industrial Aerodynamics 155 (2016) 74e88. [15] S. Cui, P. Stabat, D. Marchio, Numerical simulation of wind-driven natural ventilation: effects of loggia and facade porosity on air change rate, Build. Environ. (2016), http://dx.doi.org/10.1016/j.buildenv.2016.03.021. [16] C. Tantasavasdi, J. Srebric, Q. Chen, Natural ventilation design for houses in Thailand, Energy Build. 33 (2001) 815e824. [17] J.I. Peren, T. van Hooff, B.C.C. Leite, B. Blocken, CFD simulation of wind-driven upward cross ventilation and its enhancement in long buildings: impact of single-span versus double-span leeward sawtooth roof and opening ratio, Build. Environ. 96 (2016) 142e156. [18] J. Lin, Y. Chuah, A study on the potential of natural ventilation and cooling for large spaces in subtropical climatic regions, Build. Environ. 46 (2011) 89e97. [19] T. Bartzanas, T. Boulard, C. Kittas, Effect of vent arrangement on windward ventilation of a tunnel greenhouse, Biosyst. Eng. 88 (4) (2004) 479e490. [20] G.R. Hunt, P.F. Linden, Steady-state flows in an enclosure ventilated by buoyancy forces assisted by wind, J. Fluid Mech. 426 (2001) 355e386. [21] A.D. Stavridou, P.E. Prinos, Natural ventilation of buildings due to buoyancy assisted by wind: investigating cross ventilation with computational and laboratory simulation, Build. Environ. 66 (2013) 104e119. [22] R.J. de Dear, G.S. Brager, Thermal comfort in naturally ventilated buildings: revisions to ASHRAE Standard 55, Energy Build. 34 (2002) 549e561. [23] M. Haase, A. Amato, An investigation of the potential for natural ventilation and building orientation to achieve thermal comfort in warm and humid climates, Sol. Energy 83 (2009) 389e399. [24] I. Oropeza-Perez, P. Alberg Østergaard, Potential of natural ventilation in temperate countries e a case study of Denmark, Appl. Energy 114 (2014) 520e530. [25] H.M. Taleb, Natural ventilation as energy efficient solution for achieving lowenergy houses in Dubai, Energy Build. 99 (2015) 284e291. [26] K. Mouslia, G. Semprini, Thermal performances of traditional houses in dry

98

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

Z. Cheng et al. / Building and Environment 108 (2016) 85e98 hot arid climate and the effect of natural ventilation on thermal comfort: a case study in Damascus, Energy Procedia 78 (2015) 2893e2898. N. Artmann, H. Manz, P. Heiselberg, Climatic potential for passive cooling of buildings by night-time ventilation in Europe, Appl. Energy 84 (2007) 187e201. G. Carrilho da Graça, Q. Chen, L.R. Glicksman, L.K. Norford, Simulation of wind driven ventilative cooling systems for an apartment building in Beijing and Shanghai, Energy Build. 34 (2002) 1e11. T. Kubota, D. Chyee, S. Ahmad, The effects of night ventilation technique on indoor thermal environment for residential buildings in hot-humid climate of Malaysia, Energy Build. 41 (2009) 829e839. R. Ramponi, A. Angelotti, B. Blocken, Energy saving potential of night ventilation: sensitivity to pressure coefficients for different European climates, Appl. Energy 123 (2014) 185e195. R. Yao, B. Li, K. Steemers, A. Short, Assessing the natural ventilation cooling potential of office buildings in different climate zones in China, Renew. Energy 34 (2009) 2697e2705. L. Wang, N.H. Wong, The impacts of ventilation strategies and facade on indoor thermal environment for naturally ventilated residential buildings in Singapore, Build. Environ. 42 (2007) 4006e4015. D. Prakash, P. Ravikumar, Analysis of thermal comfort and indoor air flow characteristics for a residential building room under generalized window opening position at the adjacent walls, Int. J. Sustain. Built Environ. 4 (2015) 42e57. A. Mavrogianni, M. Davies, J. Taylor, Z. Chalabi, P. Biddulph, E. Oikonomou, P. Das, B. Jones, The impact of occupancy patterns, occupant-controlled ventilation and shading on indoor overheating risk in domestic environments, Build. Environ. 78 (2014) 183e198. I. Oropeza-Perez, P. Alberg Østergaard, A. Remmen, Model of natural ventilation by using a coupled thermal-airflow simulation program, Energy Build. 49 (2012) 388e393. A. Ioannou, L.C.M. Itard, Energy performance and comfort in residential buildings: sensitivity for building parameters and occupancy, Energy Build. 92 (2015) 216e233.

€, J. Toftum, G. Clausen, Modeling ventilation rates in bedrooms based [37] G. Beko on building characteristics and occupant behavior, Build. Environ. 46 (2011) 2230e2237. [38] V. Fabi, R.V. Andersen, S. Corgnati, B.W. Olesen, Occupants' window opening behaviour: a literature review of factors influencing occupant behaviour and models, Build. Environ. 58 (2012) 188e198. [39] W. Yin, G. Zhang, W.Yang, X.Wang, Natural ventilation potential model considering solution multiplicity, window opening percentage, air velocity and humidity in China, Build. Environ. 45 (2010) 338e344. re, A. De Herde, Impact of control rules on the effi[40] G. van Moeseke, I. Bruye ciency of shading devices and free cooling for office buildings, Build. Environ. 42 (2007) 784e793. [41] American Society of Heating, Refrigerating and Air-conditioning Engineers, Inc, ASHRAE 2009 Fundamentals Handbook, Atlanta, USA, 2009. [42] S. Wilcox, W. Marion, Technical Report NREL/TP-581-43156: Users Manual for TMY3 Data Sets, 2008. http://www.nrel.gov/docs/fy08osti/43156.pdf (accessed 15.08.21). [43] IES Virtual Environment, MacroFlo Calculation Methods, 2016. https://www. iesve.com/support/userguides?page¼8 (accessed 16.06.12). [44] B.M. Jones, Ma J. Cook, S.D. Fitzgerald, C.R. Iddon, A review of ventilation opening area terminology, Energy Build. 118 (2016) 249e258. [45] Breezway Technical Bulletin, Comparative Ventilation Openings of Various Window Types, 2012. http://www.breezway.com.au/downloads/TB_ Ventilation_Openings.pdf (accessed 15.05.08). [46] Reliable Architectural Louvers, 2010. http://www.reliablelouvers.com/Portals/ 12/Files/Reliable_Bro_2_16A.pdf (accessed 15.05.11). [47] IES Virtual Environment, ApacheSim Calculation Methods, 2016. https:// www.iesve.com/support/userguides?page¼8 (accessed 16.06.12). [48] Z. Zhai, M. Johnson, M. Krarti, Assessment of natural and hybrid ventilation models in whole-building energy simulations, Energy Build. 43 (2011) 2251e2261. [49] American Society of Heating, Refrigerating and Air-conditioning Engineers, Inc, ANSI/ASHRAE Standard 55e2013: Thermal Environmental Conditions for Human Occupancy, Atlanta, USA, 2013.