Thermal comfort and indoor air quality in rooms with integrated personalized ventilation and under-floor air distribution systems.
Under-floor air distribution (UFAD) systems have been identified with potentially better performance in providing enhanced indoor air quality (IAQ), thermal comfort, and energy efficiency in comparison to conventional mixing-ventilation (MV) systems with ceiling supply diffusers (Bauman 2003; Cermak and Melikov 2006). It is apparent that the supply of conditioned air through the UFAD outlets in the close proximity of occupants' seats would have a strong effect on their thermal sensation. In spaces served by UFAD systems, the "cold feet" complaint due to the low supply air temperature and relatively high velocity of the air movement has often been reported by occupants as an uncomfortable thermal sensation (Bauman et al. 1991, 1995; Matsunawa et al. 1995; Sekhar and Ching 2001; Kobayashi and Chen 2003; Chao and Wan 2004; Leite and Tribess 2006; Lau and Chen 2007). Supplying the conditioned air in a warmer range is usually recommended as a means of alleviating the uncomfortable sensation (Bauman 2003; Loudermilk 1999). However, by raising the supply air temperature of the UFAD system, the temperature at the breathing zone may become unacceptably warm due to the vertical temperature gradient (Webster et al. 2002). In environments with thermal stratification, the "warm head" complaint and preference for a cooler environment was found to be mainly due to the vertical temperature stratification and insufficient air movement around head level (Zhang et al. 2005; Cheong et al. 2007).
Personalized ventilation (PV) systems have the ability to deliver clean, cool, and dry air to the breathing zone of each workplace, thus making it possible to customize the environment to individual preference (Melikov 2004; Sekhar et al. 2003a, 2003b, 2005; Kaczmarczyk et al. 2004). PV in conjunction with total volume ventilation system has been studied by several researchers (Sekhar et al. 2005; Cermak 2004; Cermak et al. 2006; Cermak and Melikov 2007; Halvoriova and Melikov 2008). Studies of human response to a PV system in conjunction with MV found that the acceptability of inhaled air provided by PV increases at a higher background room air temperature in comparison with MV alone (Melikov 2004; Kaczmarczyk et al. 2004).
Integrating a PV system with the UFAD system is seen as a strategy to potentially cool the facial region while keeping the feet region warm. The inhaled IAQ was also found to be substantially improved when PV was used in space served by a UFAD system (Cermak 2004; Cermak and Melikov 2007). However, human response to PV in conjunction with UFAD has not been studied.
This study is part of a comprehensive research project that aimed to investigate the overall performance of a desk-mounted PV system integrated with UFAD system. The fundamental hypothesis of the overall research project was that the PV system, by providing cool air at the face region, would reduce the uncomfortable sensation of a warm head and would improve inhaled air quality, while the UFAD system, operating at relatively warm supply air temperature, would address the cold feet issue. This article explores associations between the physical environment and human response in room served by the PV-UFAD system. The details of the human response measurements are presented elsewhere (Li et al. 2010).
A field environmental chamber (FEC), having dimensions of 11.0 m x 7.8 m x 2.6 m (36 ft x 25.6 ft x 8.5 ft) and shown in Figure 1, is used for conducting the experiments. The air distribution system used in this study consisted of desktop PV and UFAD system. In addition, MV with ceiling supply diffusers was used as a reference. The PV system, which is served by its own dedicated primary air-handling unit (AHU), provided clean, conditioned PV air to occupant breathing zones, while the UFAD system, which has its own secondary AHU, provided conditioned recirculated air to the space through floor outlets. In order to decrease the difference between different diffusers, the system used in this study employs three separate under floor plenums to deliver the conditioned air. This strategy also serves to minimize the temperature variations in the under floor plenum. The supply air temperature is an average of the three plenums. The variations of temperature through the three plenums are within [+ or -]0.4[degrees]C(0.7[degrees]F) with a 22[degrees]C (72[degrees]F) supply air temperature and [+ or -]0.7[degrees]C (1.4[degrees]F) with an 18[degrees]C (64[degrees]F) supply air temperature.
For the cases using a ceiling-supply MV (CSMV) and UFAD system alone, the primary airflow rate was kept at 140 L/s (298 CFM). The PV air terminal devices are of round shape with a diameter of 0.1 m (3.9 in.) having a perforated front plate and an equalizer (Figure1d). The velocity measured near face is 0.3 m/s (1.0 ft/s) and 0.7 m/s (2.3 ft) with 5 L/s and 10 L/s (10.6 CFM and 21.2 CFM) PV airflow rates, respectively. The UFAD diffusers are circular perforated outlets with a diameter of 0.2 m (7.8 in.) (Figure 1c). There were 16 workstations in the FEC. Each of the workstation was equipped with a PC and a desk-mounted PV air terminal device, as shown in Figures 1a and 1b. The PV airflow rate was kept constant at each workstation. The 16 workstations were divided into 8 groups. Each group had one workstation with a 5 L/s (10.6 CFM) PV airflow rate and the other with a 10 L/s (21.2 CFM) (for example, Figure1a workstations 4B-C and 4D-E). Twenty one 50 W (171 Btu/h) (0.6 m x 0.6m[24in x 24 in]) fluorescent lighting fixtures were used in the FEC to mock up a typical open-plan office.
[FIGURE 1 OMITTED]
Experimental design and experimental procedure
The experimental conditions for the overall research project, including the physical and human response measurements, involved different combinations of UFAD supply air temperature (22[degrees]C and 18[degrees]C [72[degrees]F and 64[degrees]F]) and PV supply air temperature (22[degrees]C and 26[degrees]C [71.6[degrees]F and 79[degrees]]F) as well as three experiments at reference conditions without PV; i.e., UFAD with supply air temperature at 22[degrees]C and 18[degrees]C (71.6[degrees]F and 64[degrees]F) and MV with ceiling supply air diffuser. In Table 1, the eight experimental conditions are listed (air temperature kept in the room, air temperature supplied from the UFAD system, PV air temperature, PV supply flow rate, and total ventilation flow supplied to the room). During all experiments, the air temperature at four points of the room (A, B, C, and D in Figure 1a) at a 1.3 m (4.3 ft) room height was used as a target temperature and was controlled at 26[degrees]C (79[degrees]F). While increasing the level of room set-point temperature is interesting and worth exploring, it is to be noted that in practice, occupants will move around and cause mixing of the room air. At an increased activity level, the higher set-point room temperature might cause uncomfortable thermal sensation. It is due to this reason that 26[degrees]C (79[degrees]F) was selected as the room set-point temperature.
The operating conditions were controlled by a building automation system (BAS). To keep the room air temperature at 26[degrees]C (79[degrees]F) at a 1.3 m (4.3 ft) height, the supply flow rates were different according to different UFAD supply air temperatures (480 L/s, 1021 CFM with 22[degrees]C [71.6[degrees]F] and 360 L/s 766 CFM with 18[degrees]C [64[degrees]F]). In each experiment, the floor diffuser discharge rate was kept constant at 20 L/s. The vertical throw of the diffuser is 0.5 m (1.6 ft) (maximum height at which the velocities still reach 0.25 m/s [0.82 ft/s]). The number of diffusers was different in the various cases investigated : 24 in the cases with UFAD supply air temperature at 22[degrees]C (71.6[degrees]F) and 18 in the cases with UFAD supply air temperature at 18[degrees]C(64[degrees]F). The supply flow rate of CSMV was 750 L/s (1596 CFM).
The thermal manikin measurements formed an integral part of the physical measurements. During the experiment, a thermal manikin was placed in front of a workstation in a sedentary posture to simulate a human-being with typical summer clothing (0.7 clo). The manikin was located at workstation J4 (Figure 1a) and operated with both thermal and breathing modes. The operating parameters of the manikin are listed in Table 2. The body of the manikin is divided into 26 segments. The surface temperature and heat flux for each of the body segments was recorded every 1 min. The "manikin-based equivalent temperature" (ISO 2004 [Standard 14502-2 2004]) was used as the index to determine the effects of the thermal environment on the body cooling.
Before the experiment, calibration of the manikin was performed in the FEC. During the calibration, the indoor condition was kept as close to homogeneous as possible. The manikin was exposed in the chamber to a given air temperature, dressed, and kept at sedentary posture as it was during subsequent actual experiments. The heat loss from the body segments was recorded. Under the homogeneous condition, the indoor air temperature was equal to [t.sub.eq].
Then, the constant C values were calculated based on Equation 1 and were used to calculate [t.sub.eq] for subsequent experiments with UFAD alone and PV combined with UFAD.
[t.sub.eq] = 36.4 - C x Q[t.sub.,] (1)
[t.sub.eq] is the manikin-based equivalent temperature, [degrees]C;
36.4 is the deep body temperature, [degrees]C;
[Q.sub.t] is the sensible heat loss, W/[m.sup.2];
C is the constant, dependent on clothing, body posture, chamber characteristics, and thermal resistance off set of the skin surface temperature control system, K.[m.sup.2]/W.
Air temperature, mean velocity, and draft rating (DR) were measured at a point (G10 in Figure 1a) in the occupied zone as well as at one work station (J4 in Figure 1a) simultaneously with the subjective measurement for 2.5 h. The temperature of the inhaled air was also measured near the manikin's mouth (0.03-0.05 m [11.8-19.7 in.]). Tracer gas measurements for each of the eight PV-UFAD experimental conditions were conducted at the conclusion of the human response measurements. The parameters that were measured, the corresponding locations, and the measuring instruments used are listed in Tables 3 and 4.
The air temperature, mean velocity, and DR in the occupied zone were measured at heights of 0.1 m (0.3 ft), 0.3 m (1.0 ft), 0.6 m (2.0 ft), 1.1 m (3.6 ft), 1.3 m (4.3 ft), and 1.7 m (5.6 ft) using the omni-directional transducer. At the work station, these parameters were measured near the manikin at heights of 0.1 m (0.3 ft), 0.2 m (0.7 ft), 0.6 m (2.0 ft), 1.0 m (3.3 ft), 1.1 m (3.6 ft), 1.2 m (3.9 ft), 1.3 m (4.3 ft), and 1.4 m (4.6 ft) using the omni-directional transducer. The location of the sensor was kept 0.15 m to 0.20 m (5.9 in. to 7.9 in.) away from the manikin's body, which is aimed to reduce the effect of the convection flow around the human body. The profiles of the mean value of those parameters along vertical height from floor to ceiling are analyzed. Moreover, in order to identify the mixing pattern close to the work station at floor level, the air temperature at the floor surface, which is recorded by the sensors of the BAS, is incorporated when analyzing the temperature profile close to the work station. The inhaled air temperature was measured in the beginning, middle, and end of the 2.5 h duration of the experiments, and the average value was used in the analysis. Tracer gas ([SF.sub.6]) measurements were performed to investigate the performance of the PV-UFAD system in terms of the ability to provide occupants with conditioned outdoor air. [SF.sub.6] was dosed at location G10 (Figure 1a) at 1.3 m (4.3 ft) height until the concentration measured at 1.3 m height of G10, D5, D12, return grille level of K4, and manikin's mouth (J4) increased to 100 ppm. The manikin was set with both thermal and breathing mode (Table 2). The 100% outdoor air supplied by the PV system was kept free of [SF.sub.6], which was continuously sampled at the above-mentioned locations inside the environmental chamber by an infrared photo-acoustic spectrometer multi-gas sampler and analyzer. The results of tracer gas concentration measurement were used to analyze the performance of the system with regard to inhaled air quality. Two indices, as defined below, were calculated: personal exposure effectiveness (PEE) and personal exposure index (PEI).
The PEE index expresses the percentage of personalized air in inhaled air. It is derived from the following equation (Melikov et al. 2002):
P E E = [C.sub.r,sf6] - [C.sub.I,SF6]/[C.sub.R,SF6] - [C.sub.PV,SF6] (2)
[C.sub.R,SF6] is the [SF.sub.6] concentration of the tracer gas in the exhaust/return air (ppm),
[C.sub.PV,SF6] is the [SF.sub.6] concentration of the tracer gas in personalized air (ppm), and
[C.sub.I,SF6] is the [SF.sub.6] concentration of the tracer gas in the inhaled air (ppm).
The concentrations are average values taken over concentration measurement curves when steady-state conditions were reached. This index is equal to one if 100% of the personalized air is inhaled and equal to zero if no personalized air is inhaled.
The PEI, also called pollutant removal efficiency, is the effectiveness of an air-distribution system in removing internally generated pollutants from the ventilated space. It can be expressed either as an average or overall relative effectiveness for the whole occupied zone or as a local relative effectiveness. The local ventilation effectiveness for the removal of pollutants [[epsilon].sub.V], also called the PEI, is expressed as (Awbi 2003):
P E I = [C.sub.R] - [C.sub.[varies]]/[C.sub.I]-[C.sub.[varies]], (3)
[C.sub.R] is the contaminant concentration ([SF.sub.6]) in the exhaust/return air (ppm),
[C.sub.I] is the contaminant concentration ([SF.sub.6]) in the inhaled air of a person (ppm), and
[C.sub.[varies]] is the contaminant concentration ([SF.sub.6]) in the outdoor supply air (ppm).
Human response measurements
The physical measurements to characterize the IAQ and thermal comfort provided by the PV-UFAD system were accompanied with human response measurements involving university students as subjects. These experiments were conducted by a strict adherence to experimental design and protocol that complied with the requirements of the institutional review board (IRB) of the university. The details of the experimental protocol for the human response studies and their results are presented elsewhere (Li et al. 2010). A computerized questionnaire survey was used to obtain the responses from the various groups of subjects in several series of experimental conditions. The experimental conditions for the human response measurement are the same as the physical measurements. Thirty tropically acclimatized subjects, 15 males and 15 females, were divided into 4 groups (Table 5) and were asked to wear typical summer clothing (about 0.59 clo).
During the experiments, the subjects followed a predetermined protocol and were engaged in sedentary office work for 2.5 h. They were asked to sit at the workstation with a 5 L/s (10.6 CFM) PV airflow rate in the first 15 min and then given the option of changing their seats to the adjacent workstation with 10 L/s (21.2 CFM) PV airflow rate. They could make a further change after the second 15-min interval, following which the subjects were asked to stay in that current workstation for the next 2 h. Mean values of the data obtained from the responses of the subjects for the last 15 min of each experiment were used for analysis, since it is believed that subjects would have achieved steady-state conditions after they were exposed to the particular environment for long enough time (2.5 h) (Sekhar et al. 2005). The subjects responded to several questions concerning perceptions for the whole body as well as various body parts, details of which can be found elsewhere (Li et al. 2010). Some of the key responses sought include the following:
* thermal sensation at face and feet, based on the ASHRAE seven-point thermal sensation scale (cold = -3, cool = -2, slightly cool = -1, neutral = 0, slightly warm = 1, warm = 2, hot =+3) (ASHRAE 2004 [Standard 55-2004]);
* perception of air movement at face/feet (scale categories and weights were adopted on a sketch of human body: +3 = much too breezy, +2 = too breezy, +1 = slightly breezy, 0 = just right, -1 = slightly still, -2 = too still, and -3 = much too still);
* acceptability of air movement at face/feet and acceptability of perceived air quality (PAQ) on a visual-analog scale (end-points coded as 0 [very unacceptable] and 100 [very acceptable] with a clear distinction at 50); and
* perceived inhaled air temperature (PAT) ("cold" to "hot") and perceived inhaled air freshness (PAF) ("stuffy" to "fresh") on a linear scale.
The subjective responses and physical parameters were analyzed using linear correlations and involved [t.sub.eq] and thermal sensation at face; DR and thermal sensation at feet; measured inhaled air temperature and PAF/PAT and also measured inhaled air temperature and the acceptability of inhaled air quality; facial velocity (mean air velocity at 0.15 m [5.9 in.] from manikin's face and 1.3 m [4.3 ft.] height, Figure 3) and PAF/PAT and also facial velocity and the acceptability of inhaled air quality.
Results and discussion
Room air temperature/velocity/DR distribution
The warmer UFAD supply air temperature resulted in warmer room air temperature, especially at the lower part of the space. There are two height scenarios for the measurements. One is at the center of the room (Figure 2), and the other is at the workstation (Figure 3). As shown in Table 3, the measurements at the center of the room are from 0.1 to 1.7 m (0.3 to 5.6 ft) (Figure 2), and those at the workstation are from 0.1 to 1.4 m (0.3 to 4.6 ft) (Figure 3).
[FIGURE 2 OMITTED]
The air temperature measured at 0.1 m (0.3 ft) at the center of the room was 23.0[degrees]C (73.4[degrees]F) in case UV18 and 24.6[degrees] C (76.3[degrees]F)in case UV22 (Figure 2). Due to the heat load distribution in the lower space, it is observed from Figure 2 that the gradients of the room air temperature were steeper at the lower space and rather gentle in the upper space. Moreover, with lower level of supply air volume (cases with UFAD supply air temperature 18[degrees]C [64[degrees]F]), the temperature profiles were steeper than those with 22[degrees]C (71.6[degrees]F) supply air temperature, especially close to the floor level. This result is consistent with previous studies on temperature stratification in rooms with UFAD (Webster et al. 2002; Akimoto et al. 1995). The temperature distribution at the center of the room (Figure 2) revealed that in the region away from the workstation (i.e., in the center of the chamber), the thermal environments were only affected by the operating condition of the ambient total volume ventilation system (i.e., UFAD or CSMV). The effect of the PV system on the temperature distributions at the center of the room was negligible.
However, in the region close to the workstation, the patterns of temperature distribution were different than those at the center of the room. The furniture and human body obstruct the UFAD air discharged from the floor outlet and thus promote mixing, which resulted in a slightly cooler temperature at ankle level in the region close to the workstation than at the center of the room (Figures 2 and 3, 0.1 m [0.3 ft] height). In Figure 3, the temperature shown for height equal to 0 is the floor surface temperature.
At the work-station (Figure 3), the temperature distribution pattern above the table surface was different to that under the table surface. The temperature increased continuously under the table surface, but the rate of increase above the table surface was not as steep. It can be found from Figure 3 that in the space under the table, the temperature gradient between ankle level (0.1 m [0.3 ft]) and the "under surface" of the table (0.6 m [2.0 ft]) is about 3[degrees]C to 4[degrees]C (5.4[degrees]F to 7.2[degrees]F) in all experimental cases. The thermal flow generated by the heated manikin increases the temperature below the desk, resulting in air temperature as high as 28[degrees]C (82.4[degrees]F) at 1 m (3.3ft) height. This effect disappears above 1 m (3.3 ft), which results in negative vertical temperature gradient, i.e., a decrease in the temperature. This is attributed to the thermal plume of the manikin being affected by the table surface and higher temperatures being observed immediately under and above the table surface. Moreover, in cases with PV-UFAD, at the space above the table surface, the conditions of the PV air show a dominant effect on the thermal environment. There is no difference in the temperature profiles obtained with and without use of PV up to the height of 0.6 m (2.0 ft) above the floor. The use of the PV changed the vertical temperature distribution above 0.6 m (2.0 ft). The vertical temperature distribution is affected by the parameters of the PV airflow. The temperature decreases, and the vertical temperature profiles are more inclined to the cooler side when the PV airflow rate increases and its temperature decreases.
The air velocities measured at the workstation at 1.3 m (4.3 ft) (breathing zone) and 0.1 m (0.3 ft) (ankle level) are presented in Figure 4. The air velocities at the breathing zone height are apparently affected by the PV airflow rate rather than the PV and UFAD supply air temperatures. Higher PV airflow rate results in higher velocity at the facial part. When PV airflow rate was 10 L/s (21.2 CFM), the air velocities measured at 1.3 m (4.3 ft) height and 15 cm (5.9 in.) in front of the manikin were in the range of 0.6 m/s to 0.7 m/s (2.0 ft/s to 2.3 ft/s) and did not change with the changing of PV or UFAD supply air temperature (Figure 4; 18-22-10,22-22-10,18-2610, and 22-26-10). With this lower PV flow rate, the air velocities measured close to the manikin ([approximately equal to]0.3 m/s [1.0 ft/s]) were about half of that with 10 L/s (21.2 CFM) PV air. In the space close to the floor (0.1 m [0.3 ft]), similar to the distribution of the air temperature, the UFAD operating conditions have the dominant influence on the velocity at ankle level, and the effect of PV is negligible. The velocities at the lower sensor heights (0.1 m [0.3 ft] at locations G10 and J4) are in the range of 0.05 m/s to 0.15 m/s (0.16 ft/s to 0.49 ft/s). The air velocity around the diffuser at 0.1 m (3.9 in.) height was decreased to 0.04 m/s (0.13 ft/s) within a radius of 0.2 m (7.9 in.). Thus, the effect of the under-floor air diffuser on the sensor's measurement could be neglected.
[FIGURE 3 OMITTED]
The DR at the workstation at 0.1 m (0.3 ft) height (ankle level) and its relationship with occupants' thermal sensation at feet is presented in Figure 5. DR at the feet (0.1 m [0.3 ft]) is quite low since the velocity is very low. The effect of the air temperature was stronger than the effect of the air velocity. The warmer UFAD air supplied at temperature of 22[degrees]C (71.6[degrees]F) always brings a lower DR value at ankle level than air supplied at a temperature of 18[degrees]C (64[degrees]F). Although the velocities were slightly higher with the warmer UFAD supply air temperature, the DRs at the ankle level were always lower with the warmer UFAD supply air temperature (only about 5%). The association between the predicted DR and subjects' local thermal sensation and air movement perception, acceptability, and preference at the feet was analyzed. The relationship between DR and subjects' thermal sensation at feet was relatively stronger (Figure 5, R2 = 0.58). A negative linear relationship between DR and thermal sensation at the feet indicates that the warmer UFAD supply air temperature would result in lower DR and improved thermal sensation at feet.
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
It is to be noted that the experimental design adopted in this study represents a steady-state condition involving a small confined space under the desk with the manikin's legs underneath. In reality, most people will move around and not have their legs always under the desk. It is, therefore, quite likely that the heating effect was exaggerated under the desk, and this is corroborated from the measurement results. It might be interesting to study the difference between the steady state and the transient state when occupants move frequently.
Manikin-based equivalent temperature and subjective response
The manikin based equivalent temperatures are shown in Figure 6. In the lower region, equivalent temperatures differentiated according to different UFAD supply air temperatures, while at the head region, such as skull, face, and back of the neck, they were more influenced by PV air.
The equivalent temperature difference between UV22 and UV18 is presented in Figure 7. In case UV22, [t.sub.eq] at the foot segment is about 1.2[degrees]C (2.2[degrees]F) warmer than [t.sub.eq] at the feet in case UV18. At the facial part, [DELTA][t.sub.eq] decreased to only 0.3[degrees]C (0.5[degrees]F). When [DELTA][t.sub.eq] of the whole body is considered, the value is just between that at feet and face (0.7[degrees]C [1.3[degrees]F]). The equivalent temperatures in Figure 7 can be explained with the temperature and velocity distribution around the manikin. Since both air velocity and air temperature have an impact on convective heat loss, the increase of velocity and temperature at the same time may lead to either increase of the convection heat loss or decrease of convection heat loss when the temperature increase is high enough. Although the air movement close to the floor has a certain affect on increasing the convective heat loss from the manikin's feet level, the case with warmer UFAD supply air temperature can reduce the heat loss of the lower body segments. Moreover, the subjective responses on the thermal sensation at the feet (Table 6) were consistent with the results of the manikin-based equivalent temperature, which was warmer in case UV22 than in case UV18 at the feet. Both the manikin and the human response measurements indicated that the warmer UFAD supply air temperature (UV22) can improve the thermal comfort at the lower body parts.
The results in Figure 6 identify that the maximum change of [t.sub.eq] was determined at the facial region. Thus, the focus was placed on the equivalent temperature in the face segment to explore the effect of different PV parameters on the thermal environment in the breathing zone. To identify the effect of PV air, the difference between the manikin-based equivalent temperature in the cases with PV-UFAD and UFAD alone were used:
[DELTA][t.sub.eq] = [t.sub.eq] - [t.sup.*.sub.eq] (4)
where [t.sub.eq] is the manikin-based equivalent temperature obtained with PV-UFAD in [degrees]C, and [t.sup.*.sub.eq] is the manikin-based equivalent temperature in reference conditions with UFAD alone in [degrees]C.
[FIGURE 6 OMITTED]
[DELTA][t.sub.eq] at the facial region as obtained at the temperatures and PV flow rates studied is shown in Figure 8. The combination of cooler PV air (22[degrees]C [71.6[degrees]F]) and higher PV airflow rate (10 L/s [21.2 CFM]), i.e., cases 22-22-10 and 18-22-10 (Figure 8), always led to the maximum change of [t.sub.eq] ([DELTA][t.sub.eq] = -6[degrees]C [-10.8[degrees]F]) at the face when compared with the reference cases. The cooling effects of the combination of warmer PV air (26[degrees]C[79[degrees] F]) and higher PV airflow rate (10 L/s [21.2 CFM]), i.e., cases 22-2610 and 18-26-10 (Figure 8), were ranked as secondary ([DELTA][t.sub.eq] = -3[degrees]C to -4[degrees]C [-5.4[degrees]F to -7.2[degrees]F]). Significant changes of [t.sub.eq] were also found when the cooler PV air (22[degrees]C [71.6[degrees]F]) and lower PV airflow rate (5 L/s [10.6 CFM]), i.e., cases 22-22-5 and 18-22-5 (Figure 9), were used. As expected, the minimum changes of [t.sub.eq] were found in cases with warmer PV air (26[degrees]C [79[degrees]F]) and lower PV air volume flow rate (5 L/s [10.6 CFM]).
The trend line in Figure 9 shows that the relationship between these two parameters is relatively strong (with [R.sup.2] = 0.8). The relationship between thermal sensation and [t.sub.eq] at face can be represented by a positive line. The mean values of thermal sensation at face (Table 6) were also affected by the combined effect of supply air conditions of PV and UFAD. As expected, the coolest sensation was reported at the lowest UFAD supply air temperature, lowest PV supply air temperature, and highest PV airflow rate (18-22-10, Table 6), while the warmest sensation at face was reported at the warmest UFAD supply air temperature, warmest PV supply air temperature, and lowest PV air volume flow rate (22-26-5, Table 6). In cases 22-2610 and 18-26-5, the facial thermal sensations were comparable to each other. This is consistent with the results of manikin-based equivalent temperature. Thus, the subjects' thermal sensation could be predicted by the manikin-based equivalent temperature. The overall thermal sensations have been presented elsewhere (Li et al. 2010), in which it is shown that the overall thermal comfort is optimal at an overall thermal sensation that is slightly cooler.
Measured inhaled air quality and subjective responses
The PEE and PEI values obtained for the experimental conditions in the present study are shown in Figure 10. The findings of previous studies (Cermak 2004; Cermak and Melikov 2006) with UFAD alone are shown as well. The percentage of PV air or the percentage of outdoor air in the inhaled air does not change a lot when different PV airflow rates (10 L/s and 5 L/s [21.6 CFM and 10.6 CFM]) are used. When comparing the PEE in pairs of 10 L/s and 5 L/s (21.6 CFM and 10.6 CFM), for example, cases 22-22-10 and 22-22-5, the difference between these two PV airflow rates is only 0.03. Compared with PV airflow rates, the supply air temperature of PV air has a stronger effect on the PEE. It was observed that the cooler PV air was always able to deliver a higher percentage of outdoor air. The PEI shows a similar pattern as PEE. Better performance is achieved with PV-UFAD when compared with total MV, which has a PEI = 1. The PEI increases with the decrease of PV supply air temperature. The higher supply volume of PV air results in slightly higher PEI but is not as apparent as that caused by PV supply air temperature. This indicates that the air in the inhalation zone was rather mixed and was not 100% clean PV air. The mixing pattern of the air at the inhalation region might be affected by a multiple combination effect of PV and ambient conditions (i.e., PV supply air temperature and flow rate, ambient total volume ventilation supply air temperature and flow rate, etc.).
[FIGURE 7 OMITTED]
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
The UFAD supply conditions also have a marginal effect on the PEE and PEI but are not comparable to that of PV The cooler UFAD supply air temperature tends to result in lower PEE and PEI. This might be due to the thicker thermal plume generated by the thermal manikin that is then harder for penetration by the PV air.
The relationship between the PEE/PEI and the acceptability of PAQ is shown in Figures 11a and 11b. At an 18[degrees]C (64[degrees]F) UFAD supply air temperature, the PAQ values follow similar trends as that of PEE and PEI. The PAQ and measured inhaled air quality are higher when the cooler PV supply air temperatures are used. However, with a 22[degrees]C (71.6[degrees] F) UFAD supply air temperature, the patterns of the distribution of PAQ values are not consistent with that of the PEE and PEI. It is observed that the supply air temperature of PV air has a stronger effect on the PEE and that the PAQ is most likely to be affected by the PV supply airflow rate than PV supply air temperature. For example, the cases were ranked as 22-22-10, 22-22-5, 22-26-10, and 22-26-5 according to the value of PEE from high to low, while this rank was changed to 22-22-10, 2226-10, 22-22-5, and 22-26-5 regarding the value of PAQ. This indicates that the subjects' PAQ is correlated not only with the pollution level in the inhaled air but also with other factors. Fang et al. (1998) reported that PAQ improves when temperature and relative humidity of the inhaled air decrease. The analyses of the present results support the finding that elevated temperature of the inhaled air has negative impact on the PAQ. As shown in Figure 12 the correlation between the freshness of the air as reported by the subjects decreases with the increase of the temperature of the inhaled air. However the correlation is not strong. It is also seen from Figures 12b and 12c and Table 3 that the perceived inhaled temperature and acceptability of PAQ have a weak linear relationship with measured inhaled air temperature and are not significantly correlated (at 0.05 confidence level) with measured inhaled air temperature. The reason for the weak correlations can be that the used PV air terminal device (ATD) device promoted mixing of the supplied PV air with the warm and polluted room air. As a result the temperature and pollution of the inhaled air did not change in a wide enough range to be felt clearly by the subjects. In this research, the inhaled air temperatures were measured in the breathing zone close to the mouth. Research shows that inhalation through the mouth or through the nose does not show any significant difference in the characteristics of the inhaled air (Melikov and Kaczmarczyk 2007).
[FIGURE 11 OMITTED]
[FIGURE 12 OMITTED]
Melikov and Kaczmarczyk (2008) and Melikov et al. (2008) reported that elevated facial velocity diminishes the negative impact of increased air temperature, relative humidity, and pollution level on PAQ. Further analyses were performed to study the impact of the facial velocity of the personalized flow on PAQ. Table 7 shows the Pearson correlation between PAQ and facial velocity (mean air velocity measured at 0.15 m [5.9 in.] from face, 1.3-m [4.3ft] height, Figure 3). The correlations between facial velocity and PAF/PAT and acceptability of PAQ are significant (at 0.05 confidence level). The facial velocity has a positive linear relationship with PAF and acceptability of PAQ and has a negative linear relationship with PAT. This is reasonable because the facial velocity and measured inhaled air temperature has a negative relationship (R = -0.77, P = 0.009). Thus the results of the present study confirm the positive impact of elevated facial velocity on PAQ and reported in previous studies.
In Figure 13, the velocities are clustered into three regions: [approximately equal to]0.1 m/s (0.3 ft/s) (no PV air), [approximately equal to]0.3 m/s (1.0 ft/s) (PV airflow rate 5 L/s [10.6 CFM]), and [approximately equal to]0.7 m/s (2.3 ft/s) (PV airflow rate 10 L/s [21.2 CFM]). The velocity vector is normal to the face in Figure 13. In the case of air velocity on the x-axis, each of the 11 experimental conditions plotted represents the mean value of 4 repeat experiments conducted to complete the entire sample size, as shown in Table 5. Similarly, the acceptability of perceived inhaled air quality on the y-axis in Figure 13 represents the mean value of all 30 subjects for each of the 11 experimental conditions.
The acceptability of PAQ increases with the increase of velocity. The linear relationship between acceptability of PAQ and facial velocity is relatively stronger than that between acceptability of PAQ and measured inhaled air quality. Moreover, when comparing the correlation coefficient (R-value) and significance (P-value), the correlation between PAQ and facial velocity are always stronger than that between PAQ and measured inhaled air temperature. As already discussed, the mixing of the personalized air with the room air as promoted by the used PV ATD might be the reason.
[FIGURE 13 OMITTED]
The physical measurements (temperature, velocity, and DR distributions and manikin-based equivalent temperature) obtained in this study validate the hypotheses for improved thermal comfort in the case of PV in conjunction with UFAD when compared with the CSMV system and UFAD system alone.
The subjects' thermal sensation is shown to have a positive linear relationship with the manikin-based equivalent temperature in the facial region and a negative linear relationship with DR in the feet region. This implies that cooler PV air would be beneficial for the facial region as it would provide a cooler sensation, and the warmer UFAD supply air temperature would result in lower DR and improved thermal sensation at feet. A room set-point temperature of 26[degrees]C (79[degrees]F) was adopted in this study due to the consideration that higher temperatures may become uncomfortable and actually be less preferred by standing and walking occupants.
It is also shown that the measured inhaled air quality indices (PEI and PEE) are strongly influenced by PV supply air temperature. The PEI increases with the decrease of PV supply air temperature, i.e., the portion of clean outdoor personalized air in the inhaled air is larger when the supplied air is cool. Perceived air freshness, perceived air temperature, and acceptability of PAQ improve with the increase of the facial velocity. The impact of air velocity on these parameters was stronger than the impact of PV supply air temperature.
In this study of an integrated UFAD-PV system, a steady-state condition involving a small confined space under the desk with the manikin's legs always underneath has been explored. This may not be typical as people will move around in reality. Future research should, therefore, include transient-state conditions that will be more realistic in representing frequent movement of occupants in a typical building.
This research was performed with the financial assistance of the National University of Singapore in the form of research grant RP 296-000-101-112. The support of the International Centre for Indoor Environment and Energy at The Technical University of Denmark and The Daloon Foundation is also acknowledged.
Akimoto, T., T. Nobe, and Y. Takebayashi. 1995. Experimental study on the floor-supply displacement ventilation system. ASHRAE Transactions 101(2):912-25.
ASHRAE. 2004. ASHRAE Standard 55-2004, thermal environmental conditions for human occupancy. Atlanta, GA: American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.
Awbi, H.B. 2003. Ventilation ofBuildings. London: Spon Press.
Bauman, F. 2003. Under Floor Air Distribution (UFAD) Design Guide. Atlanta, GA: American Society of Heating, Refrigerating, and Air-Conditioning Engineers.
Bauman, F., E. Arens, S. Tanabe, H. Zhang, and A. Baharlo. 1995. Testing and optimizing the performance of a floor-based task conditioning system. Energy and Buildings 22:73-186.
Bauman, F., L.P. Johnston, H. Zhang, and E. Arens. 1991. Performance testing of a floor-based, occupant-controlled office ventilation system. ASHRAE Transactions 97(1):553-65.
Cermak, R. 2004. Design strategies for personalized ventilation. PhD Thesis (Chapter 5, pp. 63-92), International Centre for Indoor Environment and Energy, Department of Energy Engineering, Technical University of Denmark.
Cermak, R., and A.K. Melikov. 2006. Air quality and thermal comfort in an office with underfloor, mixing and displacement ventilation. International Journal of Ventilation 5(3):323-32.
Cermak, R., and A. Melikov. 2007. Protection of occupants from exhaled infectious agents and floor material emissions in rooms with personalized and under floor ventilation. HVAC&R Research 13(1):23-38.
Cermak, R., A.K. Melikov, L. Forejt, and O. Kovar. 2006. Performance of personalized ventilation in conjunction with mixing and displacement ventilation. International Journal of Heating, Ventilation and Refrigeration Research 12(2):295-311.
Chao, C.Y. and M.P. Wan. 2004. Airflow and air temperature distribution in the occupied region of an under floor ventilation system. Building and Environment 39:749-62.
Cheong, K.W.D., W.J. Yu, S.C. Sekhar, K.W. Tham, and R. Kosonen. 2007. Local thermal sensation and comfort study in a field environment chamber served by displacement ventilation system in the tropics. Building and Environment 42:525-33.
Fang, L., G. Clausen, and P.O. Fanger. 1998. Impact of temperature and humidity on the perception of indoor air quality. Indoor Air 8(2):80-90.
Halvonova, B., and A.K. Melikov. 2008. Displacement ventilation in conjunction with personalized ventilation. Proceedings of the 11th International Conference on Indoor Air Quality and Climate--Indoor Air 2008, Copenhagen, August 17-22, Paper ID: 411.
ISO. 2004. ISO 14505-2, ergonomics of the thermal environment--Evaluation of thermal environment in vehicles--Part 2: Determination of equivalent temperature. Geneva, Switzerland: International Organization for Standardization.
Kaczmarczyk, J., A. Melikov, and P.O. Fanger. 2004. Human response to personalized ventilation and mixing ventilation. Indoor Air 14(Suppl. 8):17-29.
Kobayashi, N., and Q.Y. Chen. 2003. Floor supply displacement ventilation in a small office. Indoor and Built Environment 12:281-91.
Lau, J., and Q.Y. Chen. 2007. Floor-supply displacement ventilation for workshops. Building and Environment 42:1718-30.
Leite, B.C.C., and A. Tribess. 2006. Analysis of thermal comfort in an office environment with under floor air supply in a tropical climate. HVAC&R Research 12:215-29.
Li, R., S.C. Sekhar, and A.K. Melikov. 2010. Thermal comfort and IAQ assessment of under-floor air distribution system integrated with personalized ventilation in hot and humid climate. Building and Environment Journal 45(9):1906-13.
Loudermilk, K.J. 1999. Under floor air distribution solutions for open office applications. ASHRAE Transactions 105(1):605-13.
Matsunawa, K., H. Iizuka, and S. Tanabe. 1995. Development and application of an under floor air-conditioning system with improved outlets for a "smart" building in Tokyo. ASHRAE Transactions 101(2):887-901.
Melikov, A.K. 2004. Personalized ventilation. Indoor Air 14:157-67.
Melikov, A.K., R. Cermak, and M. Majer. 2002. Personalized ventilation: Evaluation of different air terminal devices. Energy and Buildings 34:829-36.
Melikov, A., and J. Kaczmarczyk. 2007. Indoor air quality assessment by a breathing thermal manikin. Indoor Air 17(1):50-9.
Melikov, A., and J. Kaczmarczyk. 2008. Impact of air movement on perceived air quality at different level of air pollution and temperature. Proceedings of the 11th International Conference on Indoor Air Quality and Climate--Indoor Air 2008, Copenhagen, August 17-22, Paper ID: 1033.
Melikov, A., J. Kaczmarczyk, and D. Silva. 2008. Impact of air movement on perceived air quality at different level of relative humidity. Proceedings of the 11th International Conference on Indoor Air Quality and Climate--Indoor Air 2008, Copenhagen, August 17-22, Paper ID: 1037.
Sekhar, S.C., and C.S. Ching. 2001. Indoor air quality and thermal comfort studies of an under-floor air-conditioning system in the tropics. Energy and Buildings 34:431-44.
Sekhar, S.C., N. Gong, C.R.U. Maheswaran, K.W.D. Cheong, K.W. Tham, A. Melikov, and P.O. Fanger. 2003a. Energy efficiency potential of personalized ventilation system in the tropics. Proceedings of Healthy Buildings 2003, Singapore, 2:686-91.
Sekhar, S.C., N. Gong, C.R.U. Maheswaran, K.W.D. Cheong, K.W. Tham, A. Melikov, and P.O. Fanger. 2003b. Preliminary findings of a pilot study of personalized ventilation in a hot and humid climate. Proceedings of Healthy Buildings 2003, Singapore, 2:825-30.
Sekhar, S.C., N. Gong, K.W. Tham, K.W.D. Cheong, A.K. Melikov, D.P. Wyon, and P.O. Fanger. 2005. Findings of personalized ventilation studies in a hot and humid climate. International Journal of Heating, Ventilating, Air-Conditioning and Refrigerating Research 11(4):603-20.
Webster, T., F. Bauman, and J. Reese. 2002. Under floor air distribution: Thermal stratification. ASHRAE Journal 44(5):28-33.
Zhang, H., C. Huizenga, E. Arens, and T. Yu. 2005. Modeling thermal comfort in stratified environments. Proceedings of Indoor Air 2005, September 4-9, 2005, Beijing, pp. 133-7.
Ruixin Li, (1,2) S. C. Sekhar, (1), * and A.K.Melikov (2)
(1) Department of Building, National University of Singapore, 4 Architecture Drive, Singapore 658882
(2) Department of Civil Engineering, Technical University of Denmark, Denmark
* Corresponding author e-mail: email@example.com
Received April 4, 2010; accepted November 15, 2010
Ruixin Li is PhD Student. S. C. Sekhar, PhD, Fellow ASHRAE, is Associate Professor. A. K. Melikov, PhD, Fellow ASHRAE, is Associate Professor.
Table 1. Experimental conditions. Temperature, [degrees]C([degrees]F) Exp. condition System type (session no.) [T.sub.room] [T.sub.UFAD supply] Mixing C (a) 26 (79) 16(61) UFAD UV22 26 (79) 22 (71.6) UFAD+PV 22-26-5 26 (79) 22 (71.6) UFAD+PV 22-26-10 26 (79) 22 (71.6) UFAD+PV 22-22-5 26 (79) 22 (71.6) UFAD+PV 22-22-10 26 (79) 22 (71.6) UFAD UV18 26 (79) 18 (64) UFAD+PV 18-26-5 26 (79) 18 (64) UFAD+PV 18-26-10 26 (79) 18 (64) UFAD+PV 18-22-5 26 (79) 18 (64) UFAD+PV 18-22-10 26 (79) 18 (64) Temperature, [degrees]C ([degrees]F) Total volume PV flow rate, ventilation system System type [T.sub.PV supply] L/s (CFM) flow rate, L/s (CFM) Mixing -- -- 750 (1596) UFAD -- -- 480 (1021) UFAD+PV 26 (79) 5 (10.6) 480 (1021) UFAD+PV 26 (79) 10(21.2) 480 (1021) UFAD+PV 22 (71.6) 5 (10.6) 480 (1021) UFAD+PV 22 (71.6) 10(21.2) 480 (1021) UFAD -- -- 360 (766) UFAD+PV 26 (79) 5 (10.6) 360 (766) UFAD+PV 26 (79) 10(21.2) 360 (766) UFAD+PV 22 (71.6) 5 (10.6) 360 (766) UFAD+PV 22 (71.6) 10(21.2) 360 (766) (a) C refers to the conventional CSMV Table 2. Manikin operating conditions during experiment. Parameters/condition Value/description Posture Seated Thermal operation Comfort Clothing 0.7 clo (undergarments, T-shirt, pants, and slippers) Respiration Inhalation Through mouth 2.5 s/breathing Exhalation Through nose 2.5 s/breathing, 34[degrees]C (93[degrees]F) exhaled air temperature Break 1.0 s/breathing Frequency 10 times/min Pulmonary 6 L/min (0.21 CFM) ventilation volume Table 3. Details of thermal comfort and IAQ parameters measured. Measured parameter Name of instrument Floor supply air temperature (Ts) HOBO data logger Floor supply air velocity Omni-directional transducer PV supply air velocity Airflow anemometer Return air temperature (Te) HOBO data logger Room air temperature, velocity, Thermal anemometer with and DR omnidirectional transducer Local temperature, velocity, Thermal anemometer with turbulence intensity, and DR omni-directional transducer Inhaled air temperature Digital thermometer--Fluke 54 II Concentration SF6 Photo-acoustic spectrometer multi-gas analyzer (INNOVA) Measured parameter Location (Figure 1a) Floor supply air temperature (Ts) Floor diffuser Floor supply air velocity Floor diffuser PV supply air velocity Each PV outlet Return air temperature (Te) Return grille K4 Room air temperature, velocity, G10 at 0.1, 0.3, 0.6, 1.1, 1.3, and DR and 1.7 m (0.3, 1.0, 2.0,3.6, 4.3, and 5.6 ft) (Figure 2) Local temperature, velocity, Near the manikin J4 (0.1, 0.2, turbulence intensity, and DR 0.6, 1.0, 1.1, 1.2, 1.3, and 1.4 m [0.3, 0.7, 2.0, 3.3, 3.6, 3.9, 4.3, and 4.6 ft] height) (Figure 3) Inhaled air temperature Manikin's mouth Concentration SF6 1.3 m (4.3 ft) height of G10, D5, D12, return grille level of K4, and manikin's mouth (J4) Table 4. Accuracy of instruments. Time intervals Name of equipment of data collection (sec) Omni-directional 60 transducer HOBO data logger 60 Airflow anemometer -- INNOVA Digital thermometer--Fluke 60 54 II Name of equipment Accuracy Omni-directional V ~ [+ or -] 0.01 m/s (0.4 in./s), transducer T ~ [+ or -] 0.5 [degrees]C (0.9[degrees]F) HOBO data logger V ~ 0.03 [+ or -] 5% m/s (1.2 in./s), T ~ [+ or -] 0.4 [degrees]C (0.7[degrees]F) Airflow anemometer 0.1 [+ or -] 3% m/s (0.3 [+ or -] 3% ft/s) INNOVA [+ or -] 2% Digital thermometer--Fluke 0.05 [+ or -] 0.03[degrees]C 54 II (0.05% [+ or -] 0.05[degrees]F) Table 5. Subjects' groups. Number of female Number of male Group subjects subjects 1 4 3 2 3 4 3 4 4 4 4 4 Table 6. Statistical results of subjective survey. 22-22-5 22-22-10 22-26-5 22-26-10 18-22-5 Sample size 19 11 18 12 20 Thermal sensation at feet Mean -0.4 -0.5 -0.2 -0.1 -0.7 Std. error 0.1 0.3 0.2 0.3 0.2 Thermal sensation at face Mean -0.6 -1.0 -0.4 -0.7 -0.8 Std. error 0.2 0.2 0.1 0.1 0.2 Perception of air movement at face Mean 0.6 0.5 0.6 0.5 0.5 Std. error 0.2 0.2 0.1 0.2 0.1 Acceptability of air movement at face Mean 83 83 80 83 83 Std. error 3.7 4.9 4.4 4.9 3.9 Preference for air movement at face Mean -0.2 0.0 -0.1 0.1 -0.1 Std. error 0.1 0.1 0.1 0.1 0.1 Perception of air movement at feet Mean -0.1 -0.3 -0.2 -0.4 0.3 Std. error 0.2 0.2 0.2 0.3 0.3 Acceptability of air movement at feet Mean 80 82 74 70 78 Std. error 4.3 5.7 4.7 8.0 5.0 Preference for air movement at feet Mean 0.1 0.2 0.1 0.3 -0.1 Std. error 0.1 0.1 0.1 0.1 0.1 Acceptability of PAQ Mean 79 88 76 88 84 Std. error 3.5 4.3 3.0 3.5 3.4 PAT Mean 37 34 40 34 36 Std. error 3.6 3.2 2.7 2.8 3.0 PAF Mean 74 88 74 86 83 Std. error 4.9 4.9 5.6 4.6 4.2 18-22-10 18-26-5 18-26-10 C UV22 UV18 Sample size 10 17 13 30 30 30 Thermal sensation at feet Mean -0.6 -0.4 -0.5 -0.4 -0.2 -0.7 Std. error 0.2 0.2 0.2 0.1 0.2 0.2 Thermal sensation at face Mean -1.2 -0.7 -0.9 -0.3 -0.1 -0.3 Std. error 0.3 0.2 0.2 0.1 0.1 0.1 Perception of air movement at face Mean 0.4 0.2 0.8 0.1 0.1 -0.1 Std. error 0.2 0.2 0.2 0.1 0.1 0.1 Acceptability of air movement at face Mean 85 78 83 72 71 76 Std. error 4.8 4.9 4.4 3.6 5.1 4.0 Preference for air movement at face Mean 0.0 -0.1 -0.2 0.3 0.4 0.4 Std. error 0.0 0.1 0.1 0.1 0.1 0.1 Perception of air movement at feet Mean -0.2 -0.3 -0.1 0.2 -0.1 0.3 Std. error 0.1 0.2 0.1 0.1 0.1 0.2 Acceptability of air movement at feet Mean 65 68 78 74 84 76 Std. error 8.4 6.1 4.2 4.3 2.5 4.1 Preference for air movement at feet Mean 0.3 0.1 0.1 0.1 0.4 0.2 Std. error 0.2 0.1 0.1 0.1 0.1 0.1 Acceptability of PAQ Mean 84 80 81 81 77 80 Std. error 5.8 4.5 5.2 3.0 3.9 3.3 PAT Mean 33 38 33 39 41 38 Std. error 2.9 4.0 3.0 2.5 2.5 3.1 PAF Mean 83 73 77 73 71 69 Std. error 6.4 6.1 7.0 4.8 5.4 5.3 Table 7. Pearson correlation between measured inhaled air temperature ([T.sub.inhaled]) and human responses of inhaled air and facial velocity (mean air velocity at 0.15 m [5.9 in.] from face, 1.3 m [4.3 ft] height) and human responses of inhaled air. Acceptability of perceived inhaled air PAF PAT quality [T.sub.inhaled] R -0.649 0.434 -0.552 P 0.031 (a) 0.182 0.078 Facial R 0.827 -0.872 0.766 velocity P 0.003 (a) 0.001 (a) 0.01 (a) (a) Significant with 0.05 confidence level.
|Printer friendly Cite/link Email Feedback|
|Author:||Li, Ruixin; Sekhar, S.C.; Melikov, A.K.|
|Publication:||HVAC & R Research|
|Date:||Sep 1, 2011|
|Previous Article:||The influence of wind on outdoor thermal comfort in the city of Beirut: a theoretical and field study.|
|Next Article:||Probability analysis of wind-driven natural ventilation potential in areaway-attached basements.|