Thermal and ventilation performance of a naturally ventilated sports hall within an aquatic centre

Energy and Buildings 58 (2013) 111–122

Contents lists available at SciVerse ScienceDirect

Energy and Buildings journal homepage: www.elsevier.com/locate/enbuild

Thermal and ventilation performance of a naturally ventilated sports hall within an aquatic centre Priyadarsini Rajagopalan ∗ , Mark B. Luther 1 School of Architecture and Building, Deakin University, 1 Gheringhap Street, Geelong, Australia

a r t i c l e

i n f o

Article history: Received 31 January 2012 Received in revised form 3 November 2012 Accepted 15 November 2012 Keywords: Sports hall Thermal comfort Ventilation CO2 levels Air change rate

a b s t r a c t There has been an increasing demand for sports facilities in urban areas recently. As a result of this, more attention is drawn towards not only the energy performance of these building typologies, but also creating a healthy indoor environment for its users. This study investigates the thermal and ventilation performance of a naturally and hybrid (assisted by exhaust fans) ventilated sports hall within an aquatic centre situated in the temperate climate of Victoria, Australia. Its evaluation predominantly considers continuous on-site measurements of air temperature stratification, thermal comfort, CO2 levels, thermal images and tracer gas ventilation studies. Further ventilation analysis is accompanied by CFD simulations towards the development of optimised conditioning strategies. A high level of thermal discomfort was observed for this space during a late summer period when over-heating is a concern. A number of energy efficient strategies are considered to improve the thermal comfort condition without adopting refrigerant conditioning and not sacrificing indoor air quality. A better understanding of how to improve and control such spaces primarily under a naturally ventilated condition is the outcome of this study. © 2012 Elsevier B.V. All rights reserved.

1. Introduction There has been a growing demand for top quality recreational facilities all over the world recently. The indoor sports hall has become one of the favourite venues for sport activities. Traditionally, sport in Australia is an outdoor pastime. However, in recent times, an increased awareness of skin cancers and other health risks, coupled with a broadening of the range of sports being played, has seen an upsurge in indoor recreational pursuits and a trend towards multi-purpose facilities. The growing desire for better indoor environmental quality in the indoor sports centres has resulted in a marked increase in energy consumption in this building sector. While current research on indoor environmental quality mainly focuses on residences and offices, few studies are concerned with the indoor environment of large enclosures. Trianti Stourna et al. investigated the technical, functional and administrative obstacles in energy conservation in sports centres for improving energy efficiency as well as thermal and visual performance [1]. A numerical study using computational fluid dynamics revealed that significant thermal stratification occurs in the gymnasium and the annual cooling load can be overestimated by 45.4%

∗ Corresponding author. Tel.: +61 3 52278391; fax: +61 3 52278303. E-mail addresses: [email protected] (P. Rajagopalan), [email protected] (M.B. Luther). 1 Tel.: +61 3 52278385; fax: +61 3 52278303. 0378-7788/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.enbuild.2012.11.022

for the best exhaust position when the effect of thermal stratification is not considered [2]. Studies regarding sports centres mainly deal with ventilation issues in terms of energy cost and saving, while indoor air quality is not directly addressed. Stathopoulou et al. investigated the air quality in two large athletic halls with different ventilation characteristics and found that outdoor pollution significantly affected indoor air quality of both halls [3]. In a study conducted in Switzerland, indoor air pollution profiles were assessed in three different halls for recreational activities. Sharp increases in pollution concentrations were observed when visitors entered the halls and an exponential decrease in concentration after the event ended [4]. Chow et al. proposed new method for assessing the ventilation in large spaces based on tracer gas step down or decay method inside a control volume [5]. The amount of fresh air supplied to the space was observed to be sufficient, but the distribution of air was not considered carefully. For this particular case, it was difficult to provide a comfortable thermal environment with good indoor air quality, thus wasting energy on operating the environmental control systems. In Australia, aquatic centres are multipurpose indoor facilities. They provide the public with activities that enhance the community’s health, play sport and improve fitness. According to the Aquatic and Recreation Institute (ARV), these recreational facilities attract around 263 million visits a year. Due to the age of the existing building stock, a large number of the aquatic centres are about to be refurbished. In this paper, the indoor environmental conditions of a sports hall were measured using the Mobile Architecture and

112

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

Fig. 1. Floor plan of the aquatic centre.

Built Environment Laboratory (MABEL) facility [6]. Measurements were taken during summer period where high level of discomfort was observed. A number of strategies are proposed and analysed in order to recommend some guidelines for practical retrofitting of these building types. 2. Sports hall building description The construction of the aquatic centre built 25–30 years ago is typical of many commercial buildings built in the 1980s and exhibits facades that are generally light metal or brick veneer. The sports hall is 62 m long, 36 m wide and 10 m high with the total volume of 22,320 m3 . There are 19 continuous vents of size 950 mm × 950 mm along the lower west wall. There is no cooling system in the hall and three exhaust fans installed in the ceiling assist the ventilation. The fans remain off for most of the user period. Fig. 1 illustrates the floor plan showing the location of monitoring equipment and type of measurements undertaken at each of the selected areas of the building with the MABEL facility. The ‘measurement survey mode’ consisted of measurements taken in 15 min intervals at each location at 10:00, 13:00, and 16:00 and after hours (22:00). Figs. 2 and 3 show the view of the sports hall and the vents at the western fac¸ade respectively.

Fig. 2. View of the sports hall.

3. Methodology Measurements were conducted during a moderately hot week in March. Air temperature stratification (1 m, 2 m, 3 m, 4 m, 5 m, 6 m), thermal comfort and CO2 concentrations were measured internally. In addition, on-site external weather conditions and solar radiation were collected at 15 min intervals. This data provides an indication of the thermal conditions and air quality of this space during occupied and non-occupied hours. A survey was

Fig. 3. Vents along the west wall.

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

113

Table 1 Summary of measured parameters. Parameters measured

Standards

Measurement description

Comfort Weather, solar and light (external on site data)

Essential environmental data

Thermal comfort levels

ASHRAE 55, ISO 7730,

Defines the external conditions under which internal measured parameters are taken. Predicted mean vote & percentage dissatisfied (PMV/PPD). Also the ‘adaptive model’ of comfort.

Ventilation and IAQ Air change rates (effective air change)

AS 1668.2-2002 ASHRAE 129-1997

Indoor air quality (CO2 )

ISO 6242 ASHRAE 62, IAQ

Air temperature Stratification

N/A

conducted among the staff on the various areas of the aquatic centre. This provided further evidence in regards to the perceived state of thermal conditions in the sports hall. The measured parameters were analysed to develop various potential strategies for natural and low energy conditioning. In order to assess the proposed strategies, IES thermal and CFD simulation program was employed [7,8]. Table 1 summarises the measured parameters, relevant standard and deliverables for the evaluation. The findings are typical of other naturally ventilated sport hall facilities of similar construction located in a temperate climate. The overheating problems encountered in such facilities could be remedied within this facility using the proposed solution strategies. 3.1. Instrumentation The instrumentation applied here is a comfort cart which has been designed according to the ASHRAE (American Society of Heating, Refrigerating and Air-Conditioning Engineers) standard. The carts measure the dry-bulb (DB) temperature, the globe temperature (Tg) and the air velocity at 0.1 m, 0.6 m and 1.1 m heights. An additional relative humidity sensor is also included at a central location on the cart. Recently an air-velocity and dry-bulb temperature sensor at a 1.7 m height as well as a CO2 meter have been added. The calibration and margin of error for the various instruments are is provided in Table 2. Tracer gas testing can determine the actual air exchange rates under different operating conditions. This can be compared to theoretical design air change rates and building operational modes to determine compliance. The instrumentation applied is the Bruel & Kaejer multi doser and gas analyser

Room ventilation (air change) rates under various operational conditions. In conjunction with CSIRO measurements. NABERS and Greenstar descriptors quantifies IAQ at various locations: CO2 These are air temperature measurement on a vertical strand taken at progressive 1.0 m intervals.

(1302 and 1303 type). The dosing utilised huge stand fans placed at the corners and centre of the hall to provide a uniform mixing of the tracer gas (SF6 ). 4. Measurement and analysis 4.1. Weather data A weather station was positioned on the rooftop of the building in order to record the on-site weather data. The weather conditions are representative of an autumn season where moderate to warm temperatures are experienced and overheating periods are possible. Fig. 4 shows the weather conditions including dry-bulb temperature, humidity as well as global solar radiation. Ambient temperature ranged from 20–30 ◦ C between night and day. Fig. 5 illustrates the external air properties together with the internal air temperatures on a psychrometric chart. Several points of exterior temperature are compared with those of internal air temperature. The dark shade shows the summer comfort zone and the lighter shade represents the winter comfort zone. Simultaneous analysis of indoor–outdoor points helps to learn where the points lie on the chart relative to the comfort zone. Also this information helps in determining the energy content of the air, when it could be used for, and what type of conditioning (energy) is required to bring this air into the comfort zone. In particular, the external air in conjunction with the internal air conditions could provide useful information. It is noticed, for example, that temperatures exceeding a 26 ◦ C external temperature also exhibit a low humidity content suggesting that low energy evaporative cooling methods could be proposed.

Table 2 Specification and Calibration details of instruments. Sensor

Specification

Calibration

3 TSI omnidirectional anemometers

• Factory calibrated at TSI’s wind tunnel • Reference anemometers are traceable to NIST standards • Calibration procedures compliant with ISO 9001 and ISO 10012

3 Omega 44032 linear thermistor composite for air temperatures

Time constant adjustable to 0.2–2 s with default setting 0.2 s Range = 0.05–2.5 m/s Accuracy = 3% of reading Interchangeability ±0.1 ◦ C Time constant 1 s

3 Omega 44032 linear thermistor composite for globe temperatures

Interchangeability ± 0.1 ◦ C Time constant circa 10 min

1 HyCal integrated circuit humidity sensor

Repeatability ±0.5% RH at 25 ◦ C Total accuracy ±2% RH at 25 ◦ C Hysteresis ±0.8% of span max Time constant 15 sec at 25 ◦ C

• 6 point weather bath calibration between 10 and 35 ◦ C • Reference temperature from Cassella Assmann Aspirated psychrometer mercury-in-glass thermometer • 6 point weather bath calibration between 10 and 35 ◦ C • Reference temperature from Cassella Assmann Aspirated psychrometer mecury-in-glass thermometer The linear calibration function used in the dataloggers was derived from equilibration of sensor in the sealed atmosphere above 3 different saturated salt solutions using the HyCal portable calibration cells (HC-60 series, prepared in compliance with ASTM standard E 104-85): • Lithium chloride solution for 11.3 RH at 25 ◦ C • Potassium carbonate solution for 43.16% RH at 25 ◦ C Sodium chloride solution for 75.29% RH at 25 ◦ C

114

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

Fig. 4. Weather conditions.

4.2. Thermal comfort parameters Both ISO 7730 thermal comfort model and adaptive model are applied here. ISO 7730 model has the capability to alter occupant clothing and metabolic rate for sports activity. The prediction is an indication of how comfortable occupants conducting a sport activity in the hall might be for the indicated period. Note that the calculation of this standard takes into account, the dry-bulb temperature, mean radiant temperature, humidity, air velocity as well as occupant clothing level and their metabolic rate. This provides the calculation for the predicted percentage dissatisfied (PPD) to be made along the lines of the ISO 7730 standard. Figs. 6 and 7 show the calculated thermal comfort indices for the indoor, for a clothing level of 0.4 CLO (light clothing) and a metabolic rate of 4 activity [9]. Figs. 6 and 7 indicate periods where a high degree of discomfort may be experienced. It is to be noted that the periods of greater discomfort are concurrent with high indoor air temperatures. These are also synonymous with higher outdoor air temperatures. An adaptive model of comfort according to de Dear and Brager [10] is plotted in Fig. 8, where the mean daily external temperature

is compared against the 15 min average interior air temperature taken at 1 m height of the air temperature stratification strand. This graph illustrates that there are few hours where discomfort can be expected, given that the mean outdoor air temperature is 25.7 ◦ C. However, it should be realised that the adaptive model was developed using office workers. Perhaps this is not too far off from spectators in a sports hall. Fig. 9 shows the staff response related to the temperature in the stadium during the day. People experience discomfort due to high temperature and it is to be noted that none of them are engaged in active sports at the time of survey. Spaces like this should consider two comfort levels; one for spectators the other for active sports participants. 4.3. Thermal imaging A thermal imaging camera was used internally during the daytime periods to analyse the surface areas of the building envelope that heat up excessively. These problematic areas of the building envelope should be considered with respect to the resulting

Fig. 5. Psychrometric chart with internal and external conditions.

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

Fig. 6. ISO 7730 thermal comfort calculation.

Fig. 7. Interior (occupant level) air temperature charted against predicted percentage dissatisfied (PPD–ISO 7730).

Fig. 8. Adaptive comfort graph.

115

116

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

Thermal comfort response of the occupants

3

Table 3 Measured air change rate.

Thermal Comfort Vote

2

Fans

Sample points

ACH

ACH average

Off

Sample 1 Sample 2 Sample 3 Sample 1 Sample 2 Sample 3

0.4 0.4 0.4 2.8 2.7 2.5

0.4

1

On 0 0

2

4

6

8

10

12

14

16

18

-1

the calculated air exchange rate for a fan capacity of 27 m3 /s and volume of 22,320 m3 is 4.35/h. Other studies demonstrated that predicted air exchange rates are consistently higher (by about 34%) than those obtained from tracer measurements [11].

-2

-3

2.67

Number of Parcipants

4.6. CO2 levels Fig. 9. Subjective evaluation of thermal comfort.

interior comfort conditions especially in cases where excessive air temperatures occur. Fig. 10a and b illustrate one of the reasons for dramatic interior overheating inside the sports hall during midday. The clear storey (Fig. 10a) exhibits high internal temperatures around the glass and frame, similarly Fig. 10b indicates the vent frames overheating to above 50 ◦ C. 4.4. Air temperature stratification Fig. 11 indicates some stratification as well as overheating periods within the space. The exhaust fans are turned off during the beginning of this measurement and are activated later in the evening. Due to the ventilation assistance by the exhaust fans, it is apparent that the stratification level is diminished in the afternoon. Overheating particularly takes place at a level above the 3.0 m. The degree of stratification, approximately 4 ◦ C over a 5.0 m vertical distance is not as significant as expected. Nevertheless, a distinctive period during the occupied hours (from 11.30-5 pm) is experienced when the temperature is above 28 ◦ C. 4.5. Ventilation and air change rates Testing the air change rate helps in understanding the ventilation potential of the existing space. An exponential decay is observed (Fig. 12) and this is analysed for three different locations within this space at approximately 1.0 m level. Table 3 shows the results of the air change rates for two different conditions: first without any fan operating (only the vents open at the lower west wall) and second when the ceiling fans are turned on at their highest output level. The average air change rate is 0.4 when the fans are not operating and 2.7 when the fans are tuned on. However,

The CO2 levels and relative humidity were measured at the survey mode locations using the comfort cart as mentioned in the previous section. Fig. 13 shows the CO2 levels within the sports hall alongside the humidity levels. It is noticed that for a relatively occupied period (40–50 people), the CO2 levels are maintained surprisingly low. Generally, CO2 levels ranging below 800 ppm are considered fine as they are well below the critical threshold of 1000 ppm. The CO2 levels are therefore exceptionally good in the sports hall and are an indication of good ventilation and air quality. The relative humidity also does not appear to cause a problem. 5. Analysis and development of solutions In favorable climates and buildings types, natural ventilation can be used as an alternative to air-conditioning, saving 10–30% of total energy consumption. The above information when adequately studied could lead to several possible solutions to enhance the application of natural ventilation. A combination of strategies aimed at internal load reduction will assist to achieve substantial cooling effect leading to an acceptable thermal comfort range in the interior. These are presented below and are considered in further numerical analysis. 5.1. Revisiting the exterior and interior air conditions One of the first important control strategies is to recognise the differences in the conditions of air between the exterior and interior. It is anticipated that such will lead to better ventilation and potential conditioning of the external air. From the psychrometric chart plotted (Fig. 5), it can be seen that there are periods where external air exceeds 28 ◦ C and contain a low moisture (humidity) level. Such conditions would be well suited for the introduction of

Fig. 10. Thermal Imaging of the Sports Hall (14:35 on March 17th). (a) Clearstory, (b) vent on the west wall.

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

117

Fig. 11. Air temperature stratification.

Decay curve 1.4

1.2

0.8

L Location i 1 0.6

Locaon 2 Locaon 3

0.4

0.2

0

00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 24:00 26:00 28:00 30:00 32:00 34:00 36:00 38:00 40:00 42:00 44:00 46:00 48:00 50:00 52:00 54:00 56:00 58:00 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00

SF6(PPM)

1

Time

Fig. 12. Decay curve for SF6.

Fig. 13. Carbon dioxide & relative humidity.

118

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

Fig. 14. CO2 generation and O2 consumption as a function of physical activity, source [13].

direct evaporative cooling. Evaporative cooling of this type would easily bring 100% fresh external air into the building within the comfort zone during these excessively hot periods. Evaporative cooling yields a very cost effective and environmental conditioning solution. The following calculation shows the evaporative cooling potential. Eq. (1) gives the airflow passing through the vents A × v = V × ACH

(1)

where A is the area of the opening (m2 ), v is the velocity of air passing through the opening (m/s), V is the volume of the room (m3 ) and ACH is the Air Change Rate For a volume of 22320 m3 and opening area of 17 m2 , when the ACH is 2.7, the velocity of air passing through the opening will be 0.98 m/s. For evaporative cooling, the heat removed from the supply air can be calculated using Eq. (2) [12]. qsen = Ga × Cpa × (t2 − t1 )

(2)

where Qsen is the sensible heat in kW, Cpa is the specific heat of dry air at constant pressure = 1.005 kJ, t2 is the outside air temperature, t1 is the resulting indoor air temperature, Ga is the air flow rate in kg/s = V × ACH × density of air. For an outside air temperature of 30 ◦ C and to achieve the indoor air temperature of 25 ◦ C, the sensible heat calculated is 82.4 kW. To achieve comfortable indoor conditions, 82.4 kW heat has to be removed from the room through evaporative cooling. Evaporative coolers cost less to buy, less to install and less to operate than refrigeration air conditioners. In addition to evaporative cooling, opportunities for night purging are also recognised on the chart and this can allow the building interior to re-charge for the forthcoming hot day. This strategy is further evaluated in the forthcoming numerical simulations. The intention of this strategy is to extract as much cooling potential from the external environment as possible. Motorised vents can be used to seal off the building and ventilate at the right time. 5.2. Shading the building The thermal images in Fig. 10a and b indicate that there is justifiable reason to screen, shade or insulate particular window components from heating up the interior of the building. It is recognised that a cooler interior fac¸ade will reduce discomfort on hot

days. A shading screen fabricated from a protruded 3-D microscreen, situated on an external frame 50 mm from the existing glass fac¸ade was shown to eliminate 85% of the solar radiation upon a glass fac¸ade [13]. By blocking a significant portion of the radiation from reaching the glazing, the glass temperature will be substantially reduced from that of the original glazing system. In addition, night purging under the suitable external conditions will provide cooling to the internal thermal mass and enhance the effect of shading. 5.3. Demand controlled ventilation and CO2 levels It would seem reasonable, given the good air quality within the space, that continuous ventilation may not be necessary during excessively hot periods. In fact, such a strategy can be proposed if the west side vents were constructed of a closing louvre type and the exhaust (ceiling) fans are left off. The following calculation is a very conservative approach to determining the internal CO2 levels within the space if it were shut down to external ventilation for a given period. Conservative in the sense that CO2 concentration within such a space will undergo a stratified effect and the present calculation is for a complete mixing. Eq. (3) [14] provides the calculation for CO2 levels. This calculation assumes 50 active people in the sports hall with a metabolic rate of 4. As shown in Fig. 14 [14], the CO2 generated per person is 0.018 l/s. Q = Qx +

G

n

V

· 106

 (3)

where Q is the CO2 level in the room (ppm), Qx s the existing CO2 level, G is the CO2 generated per person (l/s), n is the number of people in the room, V is the volume of the room (m3 ) for 50 persons, the CO2 can be calculated as 3.24 m3 /h. Considering the room volume of 23,200 m3 , this is equivalent to 140 ppm/h. With existing CO2 level of 500 ppm, the room can last another 3.5 h with the openings closed, before reaching the threshold value of 1000 ppm. From these results, it can be concluded that, 3.5 h without external ventilation for this space can be tolerated even under a high activity level. At levels of 1000 ppm CO2 above outdoor, there is an occupant dissatisfaction rate of about 25% as shown in Fig. 15 [14]. CO2 sensors would need to be introduced into a control monitoring for shutting the vents and fans.

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

119

Fig. 16. Position of exhaust fan, vents and outlets for different scenarios.

Fig. 15. PPD as a function of CO2 , source [13].

5.4. Optimising the position of exhaust fans and openings using CFD analysis Numerical simulations were conducted to perform computation fluid dynamics (CFD) analysis and to predict the airflow and thermal performance of the space. IES virtual environment-microflow software is used for the numerical simulation. k-ε turbulence model was adopted to enable the effect of turbulence to be predicted. The basic equations describing the indoor airflow consist of the conservation equations of continuity, momentum, and energy for a turbulent, buoyancy-affected incompressible fluid, together with two additional transport equations for the turbulent kinetic energy and its dispersion rate. The k-ε model assumes that the turbulence viscosity (t ) is linked to turbulence kinetic energy (k) and dissipation (ε) via the relation t = C (k2 /ε) where C is a constant and  is the fluid density. 5.4.1. Simulating existing scenario A simplified model of the sports hall was created in IES- modelIT. The total numbers of grid elements in the model were around 178,000. The thermal comfort parameters including PMV, PPD as well as the velocity and air change effectiveness were analysed. A number of different scenarios were tested. They include: changing the fan location to Position 2 and Position 3, incorporating another outlet at Position 2, 3, 4 and 5. Other inputs such as varying the inlet wind speed can predict the natural ventilation effectiveness for different wind conditions. Table 4 summarises the different scenarios tested. Fig. 16 shows the location of the openings as well as outlets. The location of exhaust fans as well as outlets are shown as “positions”. Sensitivity analysis was conducted for the CFD model and the model is validated by comparing with the measured air speed at a number of locations (see Fig. 17). The velocity vectors, PMV, PPD

and air change effectiveness for each of the scenarios at a height of 1.75 m were analysed in detail. PMV and PPD were calculated for two different types of activities and clothing levels: vigorous sports with sportswear and medium light work with light office wear. Fig. 18 shows the vector profile (plan view) for the basecase scenario. It can be seen that there is some air movement in front of the vents, but the rest of the room is found to be stagnant with effectively no air movement. The maximum air change effectiveness of 1.1 was found next to the inlet whereas around 0.5 was observed at most of the other areas. With rigorous sports activities, PMV was found to be very high which is in the range of 3. This value is very similar to the measured PMV and is expected for such high level of activity especially on a hot day. With light activity (for e.g. in the case of spectators), PMV ranged from 0.9 to 1.5 where it was 0.9 next to the inlet and 1.4 at most of the other locations. Very often during hot periods there is little wind, hence fan assistance is required. The existing system has fans located at the ceiling, which is not very efficient to provide thermal comfort at the occupied zone. The position and orientation of vents is therefore very important to optimise the available wind. 5.4.2. Simulation of different scenarios 5.4.2.1. Velocity profile. Comparing scenarios 1 and 2 (exhaust fans at higher and lower levels respectively), it was found that more air movement occurred for scenario 2 when the exhaust fans were at a lower level. Scenario 2 showed a clear improvement compared to the basecase, but there was still some stagnant area in the southwest side (see Fig. 19). The next scenario tested includes replacing the exhaust fans with opening as an outlet for natural ventilation. The air movement profile is much better even though the magnitude of velocity is lower. The low velocity is due to low external wind speed. When the outlet was shifted to Position 4, more air movement occurred in the south-west part of the room. A number of simulations were

Table 4 Different scenarios tested. Scenarios

Position of fans, inlet and outlet

Base case Scenario 1 Scenario 2 Scenario 3

3 exhaust fans at Position 1 Ceiling 3 exhaust fans at Position 3 3 exhaust fans at Position 2 No exhaust fans, outlet at Position 2, external wind speed 1 m/s No exhaust fans, outlet at Position 4, external wind speed 1 m/s No exhaust fans, outlet at Position 2 and 4, external wind speed 2 m/s No exhaust fans, outlet at Position 2 and 5 external wind speed 2 m/s

Scenario 4 Scenario 5 Scenario 6

Fig. 17. Comparison of measured and simulated air velocities.

120

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

Fig. 18. Velocity vectors for the basecase.

Fig. 19. Velocity vectors for scenario 2.

conducted with varying the wind speeds to study the effect during windy conditions. It was found that when the external wind speed was higher, more air velocity was observed within the space. Scenarios 5 and 6 have two smaller outlets instead of one as shown in Fig. 16 and Table 4. External wind speed of 2 m/s was used as the boundary conditions to represent moderate windy condition. Figs. 20 and 21 show the velocity vectors for scenarios 5 and 6 respectively. Scenario 6 (1 outlet each at south and west walls)

shows better distribution of air and slightly higher wind velocity compared to scenario 5. 5.5. Air change effectiveness and thermal comfort parameters The air change effectiveness and comfort indices were analysed further. With the presence of exhaust vents, air change effectiveness increased to 0.75 for scenario 2 where the vents are at a lower

Fig. 20. Velocity vectors for scenario 5, inlet wind = 2 m/s.

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

121

Fig. 21. Velocity vectors for scenario 6, inlet wind speed 2 m/s.

position, whereas it was 0.6 when the vents were at a higher position on the wall. The air change effectiveness was more even for scenario 5 compared to scenario 6. PMV for medium light work was slightly improved for scenario 2. However, there was no significant change in PPD. With scenario 3, PMV was around 1.4–1.6 at most of the locations which indicates hot conditions. For scenario 4, when the outlet was located towards the south-west side, PMV was found to be around 0.9 which represents slightly hot but better conditions. Scenario 6 provided slightly better thermal comfort conditions compared to scenario 5. 5.6. Night purging With the existing high temperature outside during the day, it is expected that PMV will remain outside the comfort temperature levels. However, the external temperature went down to 20 ◦ C during the night. Simulations conducted to model this showed that temperature within the sports hall did not exceed 23 ◦ C when night purging was allowed. Fig. 22 shows the PMV contours when the external air temperature is 20 ◦ C. Night purging will allow the inside to become cool helping to prepare the room for the hot day ahead. The PMV range for light activity was in between 0 and 0.6, which is within the range of neutral temperatures and is considered acceptable. The PPD reduced from 46% to 20%. It was very hard to achieve comfort conditions with activity such as vigorous sports. Comfort can only be achieved with much lower external air temperature in such cases.

6. Discussion The Mobile Architecture and Built Environment Laboratory (MABEL) has measured several spaces of similar volume and construction type, such as swimming pools, ice arenas, and design studios using sophisticated equipment, built and suited towards the measurement of stratification, ventilation rates and comfort. Previous studies have shown that moderate stratification is very common in buildings with large open space. Calay et al. [15] states that stratification can be used positively to achieve improved ventilation especially in large manufacturing units where different work areas with different working temperature need to be sited together. Interestingly, the stratification under natural ventilation conditions is similar to that found in two cases of glazed skylight atrium spaces [16]. In these two cases it was discovered that there was a distinctive region at the upper portion of the atrium that exhibited a much greater rise in air temperature. Similar to the sports hall, two vertical regions (lower and upper) of thermal conditioning was observed. It is believed that this behaviour is caused by overheating of the interior construction as seen by the thermal imaging of the upper clerestory as shown in Fig 10. In other words, the air is not really moving from the lower to the upper regions of the space through thermal buoyancy, but rather, it is being heated from the top down. This finding further reinforces the proposed solutions of conditioning this space horizontally through the use of additional ventilation openings and fans at the lower occupied regions of the sports hall. Such solutions are also verified

Fig. 22. PMV for external temperature 20 ◦ C.

122

P. Rajagopalan, M.B. Luther / Energy and Buildings 58 (2013) 111–122

through the CFD modelling. It should also be realised that the upper zone of the space could be independently cross ventilated for this region if overheating continues. In a previous study, Santamouris et al. [17] proposed and evaluated several scenarios for possible interventions to ‘practical retrofitting’. These included evaporative cooling and night purging. Lu et al. [18] suggested that energy control strategies for sports training arenas are certainly applicable to a broad range of building spaces such as classrooms, theatres, conference rooms and so on. These building spaces have a common feature that their opening hours are dominated by schedules, that is, occupied and unoccupied hours are well scheduled and known in advance. Most of the findings in this paper relate to occupant discomfort and anecdotally, are problems encountered in other naturally ventilated sport hall facilities of similar construction located in a temperate climate during hot weather. This paper provides simple, practical and cost effective solutions to the common problems encountered in such large enclosures such as: 1. During peak hours of the day, internal temperatures are higher than or equal to external temperatures. Also, external relative humidity is low. 2. Occupants are of two types: spectators and participants. Both need to achieve comfort in the same space by adjustment to different parameters. 3. Stratification occurs moderately within such high and voluminous space. Yet, the space exhibits temperatures above 28.0 ◦ C for a substantial portion of the day even when external temperatures is lower. 4. Solar radiation is heating various external building surfaces leading to significantly higher temperatures on the interior. 5. CO2 levels are surprisingly satisfactory and low for the facility even during heavily occupied hours suggesting that ventilation to the outdoors can be shut down for several hours of extreme external temperature. 6. Ventilation rates range from 0.4 (considering a ‘closed’ nonmechanically ventilated) to 2.7 (exhaust fan ventilated) ach/hr indicating potential hybrid solutions for ventilation. This is similar to findings from other studies that air change rates are usually over-estimated compared to what actually occurs in real life. 7. Conclusion The thermal and ventilation performance of a naturally ventilated (assisted by ceiling fans) sports hall within an aquatic centre was investigated using field measurements and numerical simulations. The measured results are analysed to develop various potential strategies for natural and low energy conditioning. Simultaneous analysis of indoor and outdoor points in a psychometric chart help to suggest what type of conditioning (and energy) is required to bring this air into the comfort zone. A high level of thermal discomfort during warm weather with high solar radiation was observed for this space. A number of energy efficient strategies are considered to improve the thermal comfort condition without refrigerant conditioning and sacrificing the indoor air quality. In addition, various scenarios were tested and analysed by changing the position of the exhaust fans as well as incorporating natural ventilation strategies. Analysis of the CO2 level revealed that 3–4 h without external ventilation for this space can be tolerated even under a high activity level. Based on this, a possible ventilation ‘shut-down’ method is proposed for excessively hot days. The air change effectiveness was found to be satisfactory even though a high level of thermal discomfort was experienced in the space. It was discovered that measured and simulated results asserted

each other quite satisfactorily. However, calculated air change rate indicated that the specified mechanical ventilation is not achieved. This is because uniformly exhausted result does not actually occur. Other literature has confirmed that calculated ventilation overestimates the measured results [11]. Therefore further research of measured and simulated ventilation behaviour is important and this needs to find its place into general practice. Lower level exhaust fans give better comfort conditions at the occupant level compared to exhaust fans located at the roof. Also natural ventilation performance was better when the openings were located at a lower level. Changing the location of openings towards the south-west side improved the air movement and comfort conditions. Uniform velocity profile and relatively better thermal comfort was obtained with two smaller openings each positioned at different walls (scenario 6). With the existing external air temperature, it was hard to achieve very good comfort conditions when the occupants were involved in vigorous sports activities. When the external air temperature is lower, night purging will allow the inside to become cool. The results of this study provide simple, practical and cost effective solutions to the common problems encountered in such large enclosures. This will assist in the refurbishment plans as well as design of similar facilities. References [1] E. Trianti-Stourna, K. Spiropoulou, C. Theofilaktos, K. Droutsa, C.A. Balaras, M. Santamouris, et al., Energy conservation strategies for sports centers. Part A sports halls, Energy and Buildings 27 (1998) 108–121. [2] C.J. Lam, L.S.A. Chan, CFD analysis and energy simulation of a gymnasium, Building and Environment 36 (2001) 351–358. [3] O.I. Stathopoulou, V.D. Assimakopoulos, H.A. Flocas, C.G. Helmis, An experimental study of air quality inside large athletic halls, Building and Environment 43 (2008) 834–848. [4] M. Junker, T. Koller, C. Monn, An assessment of indoor air contaminants in buildings with recreational activity, The Science of the Total Environment 246 (2000) 139–152. [5] W.K. Chow, W.Y. Fung, L.T. Wonga, Preliminary studies on a new method for assessing ventilation in large spaces, Building and Environment 37 (2002) 145–152. [6] M.B. Luther, Developing an “As performing” building assessment, Journal of Green Building 4 (summer edition (3)) (2009) 113–120, College Publishing, USA. [7] IES Microflow (CFD) user guide, Virtual Environment 6.0, Integrated Environmental Solutions Limited. [8] A. Raj, D. Hes, R. Padovani, C. Jensen, Building satisfaction – using thermal modelling to identify areas of building use focus for post occupancy evaluation, in: Proceedings of Building Simulation 2011: 12th Conference of International Building Performance Simulation Association, 14–16 November Sydney, Australia, 2011. [9] ASHRAE standard 55, Thermal Environmental Conditions for Human Occupancy, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., 2004. [10] R. de Dear, G.S. Brager, Developing an adaptive model of thermal comfort and preference.”, ASHRAE Transactions 104 (1a) (1998) 145–167. [11] W.S. Bouhamra, A.S. Elkilani, M.Y. Abdul-Raheem, Predicted and measured air exchange rates, ASHRAE Journal 40 (August) (1998) 42–45. [12] CAREL Industries, Evaporative Cooling, Carel Inductries, DR Resolving, Brugine, Italy, 2012, March 2012. [13] N. Luther, T. Anderson, T. Brain, Investigating glazing system simulated results with real measurements investigating glazing system simulated results with real measurements, in: Proceedings of the 12th Conference of International Building Performance Simulation Association, Sydney, 14–16 November, 2011. [14] S.J. Emmerich, A.K. Persily, State-of-the-Art Review of CO2 Demand Controlled Ventilation Technology and Application, National Institute of Standards and Technology, Technology Administration, US Department of Commerce, 2001. [15] R.K. Calay, B.A. Borresen, A.E. Holdo, Selective ventilation in large enclosures, Energy and Buildings 32 (2000) 281–289. [16] J.R. Jones, M.B. Luther, A summary of analytical methods and case study monitoring of atria, in: ASHRAE Winter Conference, ASHRAE Transactions, 1993. [17] M. Santamouris, C.A. Balaras, E. Dascalaki, A. Argiriou, A. Gaglia, Energy conservation and retrofitting potential in Hellenic hotels, Energy and Buildings 24 (1996) 65–75. [18] T. Lu, X. Lü, M. Viljanen, A novel and dynamic demand-controlled ventilation strategy for CO2 control and energy saving in buildings, energy and Buildings 43 (2011) 2499–2508.