Improved performance of displacement ventilation by a pipe-embedded window

Improved performance of displacement ventilation by a pipe-embedded window

Building and Environment 147 (2019) 1–10 Contents lists available at ScienceDirect Building and Environment journal homepage: www.elsevier.com/locat...

4MB Sizes 0 Downloads 53 Views

Building and Environment 147 (2019) 1–10

Contents lists available at ScienceDirect

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

Improved performance of displacement ventilation by a pipe-embedded window

T

Gonghang Zhenga, Chong Shena,b,∗, Arsen Melikovb, Xianting Lia a b

Department of Building Science, Beijing Key Laboratory of Indoor Air Quality Evaluation and Control, Tsinghua University, Beijing, PR China International Centre for Indoor Environment and Energy, Department of Civil Engineering, Technical University of Denmark, Lyngby, Denmark

A R T I C LE I N FO

A B S T R A C T

Keywords: Thermal environment Air quality Physical measurements Window Displacement ventilation

The air distribution in displacement ventilation (DV) mainly depends on the heat sources in the room. The solar radiation and cold window are strong heat source or heat sink in summer and winter. A pipe-embedded window (PEW) has been developed to address the heat gain/loss through the window. In this study, the performance of the system based on DV and radiant ceiling was compared with that based on DV and PEW. A room with two workstations and two thermal manikins was adopted in the experiment. The impact of human bioeffluents and passive contaminant sources were studied. The results show that the warm window and floor in summer and cold window in winter damaged the normal air distribution of DV. The vertical temperature gradient was weakened and the ventilation effectiveness was close to that of mixing ventilation. The normalized contamination concentration was almost 1 in both workstations in different conditions. On the contrary, the PEW was able to keep the nature of DV and eliminate the negative effect from window. The bioeffluents and heat was efficiently removed by the DV flow. The exhaust air temperature in PEW system was higher in summer and lower in winter compared with radiant ceiling system.

1. Introduction One of the main goals of heating, ventilation and air-conditioning (HVAC) systems is to maintain a healthy Indoor Air Quality (IAQ). The buoyancy-driven ventilation is considered as a promising approach to remove indoor contaminants and excess heat [1]. And the displacement ventilation (DV) is one of the most popular systems that obey this principle [2]. DV has been developed for several decades, and it is becoming increasingly popular in North America and Europe, especially in Scandinavian countries [3]. Compared with the conventional all-air system, the main advantage of DV is the ability of improving IAQ in the occupied zone where the warm air is rising through the stratified air rather than being mixed. The warm air is usually associated with the contaminant sources such as occupant's exhalation and odors [4]. Thus the polluted air can be removed through the extract air directly instead of transferring in the room. However, as concerns the passive pollution sources, they are independent of the heat sources and their transportation under the DV airflow is mainly affected by their locations. Sometimes, the contaminants at lower level can be raised by the occupant plume to the breathing zone and increase the personal exposure risk [5]. The airflow pattern and contamination distribution under DV



are much more complicated than the simple mixing and dilution modes. It is crucial to apply DV in a proper way. There are already extensive investigations on the impact factors of DV performance. The main factors include the location and intensity of heat sources, the room partitioning, the parameters of supply air, the human activities and the thermal plume around windows and walls [6]. Overall, the turbulence of occupied zone under DV is relatively low, and the ventilation effectiveness of DV may be easily destroyed by strong heat sources. Schmeling [7] investigated the effect of sensible heat release on ventilation efficiency and thermal comfort in DV. The results showed that the heat removal efficiency (HRE) decreased with increasing mean cabin temperature during human subject tests. Maier [8] researched the thermal comfort of different displacement ventilation systems and found that DV could provide comfortable climate situation in the aircraft passenger cabin with low air velocities. Mundt [9] conducted tracer gas measurements in a room with DV and found that the pollutants distribution is very sensitive to disturbances. Park and Holland [10] performed simulations and the results showed that the convective heat sources had significant influence on the flow and temperature fields in DV systems. If the DV system cannot accord with the heat sources or contaminant sources in the room, its performance is even worse than that of mixing ventilation [11].

Corresponding author. Department of Building Science, School of Architecture, Tsinghua University, Beijing, PR China. E-mail address: [email protected] (C. Shen).

https://doi.org/10.1016/j.buildenv.2018.10.006 Received 25 June 2018; Received in revised form 1 October 2018; Accepted 3 October 2018 Available online 04 October 2018 0360-1323/ © 2018 Elsevier Ltd. All rights reserved.

Building and Environment 147 (2019) 1–10

G. Zheng et al.

Nomenclature Cx Cs Ce DR DSF DV DVCC

DVCW DVHC DVHW HVAC IAQ PEW WS ε

concentration at the measuring point concentration at the supply diffuser average concentration at the exhaust diffusers draft rate double skin façade displacement ventilation DV combined with chilled ceiling

DV combined with cooling window DV combined with hot ceiling DV combined with hot window heating, ventilation and air-conditioning Indoor Air Quality pipe-embedded window workstation normalized contamination concentration

systems [12]. To address the negative effects of windows, innovative solutions have been studied. Passive techniques such as the double skin façade were proposed to improve the shading in summer and insulation in winter [13]. In addition, the integration of passive techniques and active techniques has been investigated to improve the thermal

A window is no doubt an important and intense heat source in a room. In summer, a window and the floor near the window are warmed up by the solar radiation. In winter, the surface temperature of window is much lower than the room temperature. These may influence or even destroy the normal air pattern and the ventilation effectiveness of DV

Fig. 1. Plan and perspective view of the simulated room. 2

Building and Environment 147 (2019) 1–10

G. Zheng et al.

controlled to simulate a person in a neutral thermal condition [19].

performance of glass façade [14]. A pipe-embedded window (PEW) is a typical one of these systems. Pipes are embedded in the cavity of a double window with water flowing inside [15]. The PEW is able to provide cooling in summer [16] and heating in winter [17]. When PEW is applied in typical regions with strong solar radiation and various natural cooling sources, the built-in pipes can reduce near 80% of the solar heat gain through windows [18]. More important, the temperature of window can be regulated to a certain degree and the negative impact of window on the indoor environment may be eliminated. Hitherto investigations payed insufficient attention on the role of window in a DV system. The potential of newly developed windows (e.g. PEW) to protect the normal DV flow pattern and maintain a better IAQ should be evaluated. In this study, the experimental platform for simulating the effects of embedded window on indoor thermal environment and air quality was built. The performance of the system based on DV and radiant ceiling was compared with that based on DV and PEW. Both summer conditions and winter conditions and both human bioeffluents and passive contaminant sources were considered. The distribution of air temperature, velocity and contamination concentration were researched respectively. The objective of the investigation was to assess the impact of window on DV airflow and the ability of PEW to generate a healthy indoor environment with DV.

2.2. Experimental conditions The operation parameters are listed in Table 1. The internal heat sources included two manikins, two laptops, and one lamp, with a total of 275 W. The relative humidity was 35 ( ± 2) %. The room temperature was 23( ± 0.15) °C in winter conditions and 26( ± 0.15) °C in summer conditions. The reference temperature and humidity were measured in the room center at the height of 1.1 m. Two air-conditioning strategies were compared in the experiments: DV combined with radiant ceiling and DV combined with pipe-embedded window. Both winter and summer conditions were considered, summer: DV combined with chilled ceiling (DVCC) and DV combined with cooling window (DVCW), winter: DV combined with hot ceiling (DVHC) and DV combined with hot window (DVHW). The supply air temperature has been reported as an important factor for the performance of DV [20,21], thus two levels of temperature difference between supply air and room air were investigated: 3 °C and 6 °C. The lowest DV flowrate was set according to the Category II in the EN standard [22]. The temperature of the radiant ceiling or the pipe-embedded window was adjusted to address all the heat gains in the room together with DV. Three kinds of pollution conditions were simulated under each thermal condition. First condition, nitrous oxide (N2O) was dosed from the armpits of the manikin seated in WS1 while carbon dioxide (CO2) was dosed from the groin of the same manikin to simulate human bioeffluents. Second condition, N2O and CO2 were dosed from the armpits and groin of the manikin seated in WS2. Third condition, N2O and CO2 were released from the points F and G in Fig. 1 to simulate passive contaminant sources. The height of F and G was 1.1 m. The tracer gas experiments based on the similar principle have been reported in Refs. [23,24]. Those three kinds of conditions were performed one by one.

2. Method 2.1. Experimental facility Measurements were performed in a climate chamber (4.20 m × 4.12 m × 2.87 m, W × L × H) under steady state. As shown in Fig. 1, the chamber was arranged as a two workstations (WS) office. DV was employed in the chamber and a semi-circular DV diffuser was mounted in the corner. The two workstations were kept away from the DV inlet to avoid draught. Two exhaust diffusers were deployed in the ceiling. Five panels with water pipes inside were attached to the left wall of the chamber to mimic window. The panels were made of thermally conductive material coated with a high emissivity gray paint. The area of window was 6.51 m2 and the distance between window and floor was 0.69 m. The temperature of window was regulated by the circulating water connected to a heat pump. In DVCC and DVHC, the water temperature in water pipes was hot in summer and cold in winter. The performance of panels was traditional window. The chilled or hot ceiling was operated. In DVCW and DVHW, the water temperature in water pipes was relatively cool in summer and relatively hot in winter. The performance of panels was pipe-embedded window. The chilled or hot ceiling was closed. Almost left half of the floor (4.0 m × 2.0 m) was covered by carbon flexible electric films to simulate the solar radiation projecting on floor. The thermal power of the foils was regulated by a transformer. Three fourths of the ceiling (13 m2) was mounted with radiant panels. Each workstation was occupied by a thermal manikin with a clothes insulation of 0.48 clo. The two sitting manikins had almost the same distance to the DV inlet. They were consisted of 23 and 17 body segments, respectively. Each segment was individually measured and

2.3. Measuring instrumentation and procedure The locations of physical measurement were illustrated in Fig. 1. The air temperature was measured in the supply air diffuser, exhaust air diffusers, and locations A, B, C, D, E (Fig. 1) at the heights of 0.1 m, 0.3 m, 0.6 m, 1.1 m, 1.4 m, 1.7 m, 2.1 m, and 2.4 m. The velocity and turbulent intensity were measured in A, B, C, D, E at the heights of 0.05 m, 0.1 m, 0.3 m, 0.6 m and 1.1 m. The concentration of tracer gas was measured in the supply air diffuser, exhaust air diffusers, noses of the manikins, and the room center (E) at the heights of 0.1 m, 0.6 m, 0.9 m, 1.2 m, 1.5 m, and 1.8 m. The temperatures of window, ceiling, wall, and heat sources were measured by substantial points. The air movement of DV airflow was visualized by smoke. The air temperature was measured using resistance temperature sensors (accuracy ± 0.15 °C). The velocity was measured using low velocity omni-directional wireless anemometers (accuracy 0.02 m/ s ± 0.1% of reading range 0.05–1 m/s). The air was sampled and analyzed by Multipoint Samplers INNOVA 1312. The air flowrate was measured by a micromanometer. The data were recorded by Aglient

Table 1 Boundary conditions in the experiment.

Room temperature (°C) Supply temperature (°C) Flowrate (L/s) Total heat load (W/m2) Ceiling temperature (°C) Window temperature (°C)

DVCC-3∗

DVCC-6

DVCW-3

DVCW-6

DVHC-3

DVHW-3

26 23 52 42.2 21 30

26 20 26 42.2 21 30

26 23 52 14.1 × 26.5

26 20 26 14.1 × 26.5

23 20 52 18.8 26 17

23 20 52 0.0 × 21.5

DVCC-3∗: displacement ventilation with chilled ceiling, the temperature difference between supply and room air was 3 °C. 3

Building and Environment 147 (2019) 1–10

G. Zheng et al.

DVCW-3, the vertical temperature profiles of different locations almost coincided. The horizontal temperature field was homogeneous. The DV flow had the same impact on different locations in a horizontal layer. And the stratified air flow pattern indicated the typical feature of DV.

data loggers. All the instruments have been calibrated before the measurements. This paper selected typical working condition and conducted the same experiment several times. The error of experimental data was within 5%. The repeatability of the experiments has been validated.

3.2. Air velocity distribution

2.4. Criteria for assessment

The vertical velocity variations at room center are measured. The velocity here was relatively slow, almost lower than 0.1 m/s in all the conditions. As typical examples, the velocity distributions in DVCC-3 and DVCW-3 are illustrated in Fig. 4. Overall, the velocity in DVCC was higher than that in DVCW. The locations C and D were close to the DV inlet. Thus, the velocity there was relatively large. In DVCC, the velocities of several points were higher than 0.2 m/s. The largest draft rate was 20.5%, which occurred at the height of 0.05 m in location D in DVCC-6. This value was slightly higher than the limit in Category II in EN ISO 7730 [26]. Though location A was far from DV inlet, its velocity was not slow due to the surrounding thermal plumes from the window and floor. In DVCW, the highest velocity was only 0.16 m/s and the largest draft rate was only 11.1%. In the locations higher than 0.2 m above floor, the velocities were close to 0. The turbulence and thermal convection in DVCW was low [26].

The concentration of contaminant was normalized according to [25]:

ε=

Cx − Cs Ce − Cs

(1)

where Cx is the concentration obtained at the measuring point, Cs is the concentration obtained at the supply diffuser, Ce is the average concentration obtained at the exhaust diffusers, and ε is the normalized concentration at the measuring point. 3. Results 3.1. Thermal environment and flow pattern 3.1.1. Air temperature distribution The vertical temperature stratifications in all the conditions are shown in Fig. 2. DVCC and DVCW were both performed in summer conditions with a target temperature of 26 °C, but their temperature profiles were different. The stratification in DVCW was more significant than that in DVCC. In DVCW, the boundary layer between the upper fully-developed flow and lower stratified flow was appeared at the height of 1.1 m. The temperature increased noticeably with height below this boundary layer, and kept increasing slightly with height above this boundary layer. In DVCC, the stratification height was only 0.6 m. Below this layer, the temperature gradient was smaller than that in DVCW. Above this layer, the temperature decreased with height owing to the low temperature ceiling. In both DVCC and DVCW conditions, the difference caused by the supply air temperature was very small. DVHC and DVHW were conducted in winter conditions with a target temperature of 23 °C. Their vertical temperature distributions were different, but the difference was smaller than that in summer conditions. In DVHW, the height of the boundary layer was around 1.1 m and the variation of the curve was analogous to those in DVCW. In DVHC, the boundary between the upper fully-developed flow zone and lower stratified flow zone was vague. Below 1.1 m, the temperature profile of DVHC was close to that of DVHW. Above 1.1 m, the temperature in DVHC grew continuously due to the warm ceiling. The variations of the curves in DVHC and DVCC were different because the ceiling operated at different temperatures. In all the conditions, the vertical air temperature differences between 0.1 m and 1.1 m above floor were lower than the limit given from Category II in EN ISO 7730 [26]. The exhaust air temperature in the experiment was shown in Table 2. In summer conditions, the exhaust air temperature in DVCC was lower than room temperature by 0.8 °C whereas the exhaust air temperature in DVCW was higher than room temperature by 0.8 °C. In winter conditions, the exhaust air temperature in DVHC was higher than room temperature by 1.7 °C while the exhaust air temperature in DVHW was higher than room temperature by 0.5 °C. The temperature fields in DVCC-3 and DVCW-3 are compared in Fig. 3. In DVCC-3, the vertical temperature profiles varied among different locations. The warm window and floor were located in the left part of the room. The heat source conducted heat transfer with local air and the heat convection was enhanced. The farther away from the heat source, the weaker the heat convection. Thus, the room temperature decreased with the distance to the left wall and window. In the location near the window (e.g. A), the vertical temperature variation was small. The typical temperature stratification in DV has been destroyed. In

3.3. Air flow visualization The development of DV air jet in the adjacent zone of supply terminal was shown in Fig. 5. If the supply air flowrate was low (26 L/ s), the introduced air dropped down to the floor and kept at a low level. In the initial period, the flow pattern was triangular. If the supply air flowrate was high (52 L/s), the initial airflow was strong and the layer of supply air jet was thick. In this zone, recirculation was weak. The cool supply air was drawn horizontally toward the heat sources. The comparison of thermal plumes around the manikin between DVCC and DVCW was shown in Fig. 6. In DVCW, a clear natural convection around the manikin was captured by the camera. After reaching the manikin, the supply air joined the rising air in the thermal plumes around the manikin. The plumes expanded gradually until they approached the mixed air in the upper area of the room. And the warmer air could be easily extracted by the exhausted diffuser instead of reentering to the occupied zone. In DVCC, the flow distribution was disordered. The thermal plumes developed around the manikin was very few. Instead, small-scale convective flows were observed almost in the entire occupied zone. The mix effect was strong.

Fig. 2. Vertical temperature gradient at room center. 4

Building and Environment 147 (2019) 1–10

G. Zheng et al.

Table 2 Exhaust air temperature in the experiment.

Exhaust air temperature (°C)

DVCC-3

DVCC-6

DVCW-3

DVCW-6

DVHC-3

DVHW-3

25.24

25.22

26.85

26.82

24.74

23.52

Fig. 3. Detailed temperature distribution in DVCC-3 and DVCW-3.

The comparison of the flow pattern near the window between DVCC and DVCW was shown in Fig. 7. The stratified effect was found in DVCW. The air moved across different stratified air layers with a low momentum. A so-called cool air lake was formed in the stratified zone. In DVCC, the warm window was the strongest heat source. The cool and clean supply air was induced by the warm window surface from floor level to the ceiling. The stratified effect was weak and illegible.

Fig. 4. Overall velocity distribution in DVCC-3 and DVCW-3.

6, DVCC-3 had a double ventilation rate. A weak concentration gradient was found in DVCC-3. But the height of the boundary layer (stratification height) between the upper polluted zone and lower clean zone was lower than 0.6 m. The clean zone was useless for the occupants because it was much lower than the breathing zone. In DVCW, a clear concentration gradient was found at room center. The height of boundary layer at room center was between 0.9 m and 1.2 m. The difference between DVCW-3 and DVCW-6 mainly occurred above the height of 0.9 m. Surprisingly, the concentration of DVCW-3 was higher than that of DVCW-6 at some heights. In winter conditions, DVHW performed better than DVHC in terms of ventilation effectiveness.

3.4. Contamination concentration 3.4.1. Contamination stratification The contamination distributions at room center when WS1 dosed from groin and armpits are shown in Fig. 8 (a) and (b), respectively. In DVCC-6, the normalized contamination concentration was around 1 at the vertical line in the room center no matter the sources were groin or armpits. The characteristics of DV disappeared. Compared with DVCC5

Building and Environment 147 (2019) 1–10

G. Zheng et al.

Fig. 5. Pattern of DV jet under different flowrate.

Above the height of 0.9 m, the ε in DVHC-3 was close to 1 except for an extremely high concentration point observed at the height of 1.5 m. The contamination distributions at room center when WS2 dosed from groin and armpits are shown in Fig. 9 (a) and (b), respectively. In DVCC-3 and DVCC-6, the vertical contamination distributions are almost the same as WS1 dosed. The stratification was weak. In DVCW, a positive concentration gradient was observed. Because the exhaust diffusers were mounted near WS2, the contaminants were transported into the upper zone by the convection flows and directly removed from the room. The contaminants did not widely spread in the room. Compared DVHW-3 with DVHC-3, the air was cleaner in DVHW-3. The contamination distributions at room center when passive source dosed from F and G are shown in Fig. 10. Points F and G were located between WS1 and WS2. Point G was close to the DV inlet. The DV supply air passed G at first, and then went to WS1 and room center. Point F was opposite to the DV inlet. The DV supply air first went through WS2 and room center, then passed F and took the polluted air to the exhaust diffusers. Fig. 10 (a) indicates that DVCW and DVHW were more efficient to remove the contaminants from F than DVCC and DVHC. However, if the contaminants were from G, DVCC-3 performed the best. The normalized concentration in DVHW-3 was relatively high

and DVHW-3 had negative effect in this condition. 3.4.2. Breathing area air quality The contamination concentration of breathing area air was measured by a point between the mouth and nose. The values of ε when WS1 and WS2 dosed are shown in Fig. 11 and Fig. 12, respectively. The results in these two figures were analogous. The ε was almost 1 in DVCC-6, which was the highest among these conditions. On the contrary, the ε in DVCW-6 was only 0.3. When the supply air flowrate increased to the double (the supply air temperature increased simultaneously), the ε decreased by 0.1 in both DVCC and DVCW. The lowest ε was 0.2, which was found in DVCW-3. The positive effect of DV was kept in both DVHC and DVHW. But the ε in DVHW-3 was only the half of ε in DVHC-3. In most conditions, the ε when armpits dosed was a bit higher than the ε when groin dosed. The ε of breathing area air in WS1 and WS2 when passive sources dosed was shown in Fig. 13. When the contaminants were released from G (near DV inlet), the ε was high at WS1 in DVCW and DVHW conditions. The DV flow took the contaminants to WS1. In the meantime, the ε was 0.6 and 0.8 at WS2 in DVCW and DVHW, respectively. The DV flow still had positive effect. In DVCC and DVHC, the ε was around 1.

Fig. 6. Pattern of thermal plumes around manikin. 6

Building and Environment 147 (2019) 1–10

G. Zheng et al.

Fig. 7. Pattern of thermal plumes near the window.

When the contaminants were released from F, the ε was lower than 0.5 at WS1 in DVCW and DVHW conditions. The DV flow efficiently took away the contaminants. Meanwhile, the ε was high at WS2 in DVCW and DVHW conditions. In DVCC and DVHC, the ε was around 1. 4. Discussion Natural convection flows are the crucial roles in DV. The thermal plumes in DV significantly influence the performance of DV. The magnitude of the buoyancy-driven flow is decided by the dimensions and strength of the corresponding heat source. The biggest difference between the room with DVCC and the room with DVCW was the characteristic of heat sources. In DVCC, the window and the floor shined by solar radiation were warm. In addition, they took up a large area in the room. It implied that they were the main heat sources in the room. The height of these sources was not high. They directly affect the DV flow in the occupied zone. For example, the DV flow was heated up when it encountered the warm floor. Furthermore, after reaching the warm vertical window, the DV flow was induced directly to the ceiling (Fig. 7). As a result, the normal flow pattern and temperature distribution of DV was destroyed. The DV flow rising with the thermal plumes around manikins was insufficient. On the contrary, the mixing effect in the room was strong. The vertical temperature profile at room center indicated the vertical temperature gradient in the room was small and the stratification height was low. The chilled ceiling even weakened the vertical temperature gradient. However, the horizontal temperature field was highly non-uniform. The location near the window was warmer than the location opposite of window due to the solar heat gain. In DVCW, the incident solar radiation was balanced by the cooling of window. The negative effect from the strong heat sources caused by solar heat gain was eliminated. The air distribution feature of a normal DV system was well kept. The “50%-rule” was a general experience that the air temperature at floor was half-way between the supply air temperature and the exhaust air temperature [27]. The vertical temperature in DVCC was in accordance with this approximation. Horizontal air layers were observed in Fig. 7 (a). And the buoyancy flow around the manikin was clear. The horizontal temperature distribution was homogeneous. In winter, the window temperature under DVHC was low due to the relatively poor thermal insulation of window. As a result, the cold

Fig. 8. Vertical contamination distribution at room center when WS1 dosed.

downdraught counteracted the development of the normal DV flow. The stratification near the window was weakened. In addition, the downdraught transported some air from the upper area back down to

7

Building and Environment 147 (2019) 1–10

G. Zheng et al.

Fig. 9. Vertical contaminant distribution at room center when WS2 dosed.

the stratified area and increased the mixing of room air. But the heat sink of window was not so strong. Thus, the negative effect caused by window was limited in winter than in summer. Actually, the ceiling heating was considered as a suitable method for DV because it counteracted the heat loss through the ceiling and stabilized the thermal stratification. However, the heating ceiling could not address the cooling from window. In DVHW, the window surface was no longer a heat sink for the indoor space. Thus, the DV flow was not affected by

Fig. 10. Vertical contaminant distribution at room center under passive sources.

the downdraught from the window. In terms of contamination concentration, the human bioeffluents were associated with the thermal plumes of the occupants. The air stratification in the room remarkably affected the distribution of these 8

Building and Environment 147 (2019) 1–10

G. Zheng et al.

Fig. 11. ε at manikin's nose in WS2 when WS1 dosed.

Fig. 14. Schematic illustration of the contamination distribution under DV in summer condition.

Fig. 12. ε at manikin's nose in WS1 when WS2 dosed.

contaminants. The difference between WS1 and WS2 was small. The comparison of the mechanisms under DVCC and DVCW is illustrated in Fig. 14. In DVCC, the low stratification height resulted in the high contamination concentration in the occupied zone. In DVCW, the contaminants were efficiently removed by the DV flow. The breathing area contamination concentration was less than one fourth of that in DVCC. Both armpits and groin belonged to the bioeffluents sources, their distributions were similar. The height of armpits was higher than that of groin. It was beneficial for the DV flow to transport the contaminants from armpits to the upper zone. However, the armpits were more exposed to the surrounding air. The entrainment of the contaminants from armpits to the fresh air was stronger. Thus, the spread of the contaminants from armpits was a bit easier than that from groin. On the contrary, the groin was located in the core of a seated manikin. The natural convection flow easily transported the contaminants to the upper zone and the surrounding air can hardly be involved. Overall, the concentration of the contaminants from armpits was a bit higher than that from groin in the occupied zone. For passive sources, their locations had crucial influence on the contamination distribution in DVCW and DVHW. For example, the DV flow transported the contaminants from point F to WS1 and transported the contaminants from point G to WS2. But the impact of point F on WS2 and point G on WS1 was small. It was in accordance with the general nature of DV. With regard to DVCC and DVHC, their contamination distribution was close to the mixing ventilation. The normalized concentration was approximately 1. The flowrate of DV affected the concentration in both DVCC and DVCW. In DVCC, the ε was close to 1. Though large flowrate deceased the concentration, the relative rate was only about 10%. In a pure mixing ventilation, the flowrate will not affect the ε value at all. In DVCW, the flowrate of 52 L/s decreased nearly 40% of the ε value compared with the flowrate of 26 L/s. But the large flowrate may cause draught in the lower zone near the supply terminal. The effect of

Fig. 13. ε at manikin's nose when passive sources dosed.

9

Building and Environment 147 (2019) 1–10

G. Zheng et al.

flowrate on the temperature distribution was small (Fig. 2). In terms of energy grade, the exhaust air temperature in DVCC was 1.5–2 °C lower than that in DVCW due to the chilled ceiling in summer. The ventilation was inefficient to remove the heat. And the water provided for cooling window was 8–10 °C higher than that for chilled ceiling. In winter, the exhaust air temperature in DVHC was 1.5–2 °C higher than that in DVHW owing to the warmer ceiling. Thus, more heat was wasted. And the water produced for hot window was 5–8 °C lower than that for hot ceiling.

Conditioning Engineers, Inc., Atlanta, 2003. [4] M.J.A. Sabbagh, Field Evaluation of a Displacement Ventilation System for a Cold Climate School, UNIVERSITY OF CALGARY, ALBERTA, 2013. [5] H. Brohus, P.V. Nielsen, Personal exposure in displacement ventilated rooms, Indoor Air 6 (3) (1996) 157–167. [6] G. He, X. Yang, J. Srebric, Removal of contaminants released from room surfaces by displacement and mixing ventilation: modeling and validation, Indoor Air 15 (5) (2005) 367–380. [7] D. Schmeling, J. Bosbach, On the influence of sensible heat release on displacement ventilation in a train compartment, Build. Environ. 125 (2017) 248–260. [8] J. Maier, C. Marggraf-Micheel, T. Dehne, J. Bosbach, Thermal comfort of different displacement ventilation systems in an aircraft passenger cabin, Build. Environ. 111 (2017) 256–264. [9] E. Mundt, Contamination distribution in displacement ventilation—influence of disturbances, Build. Environ. 29 (3) (1994) 311–317. [10] H.-J. Park, D. Holland, The effect of location of a convective heat source on displacement ventilation: CFD study, Build. Environ. 36 (7) (2001) 883–889. [11] I. Olmedo, P.V. Nielsen, MRd Adana, R.L. Jensen, P. Grzelecki, Distribution of exhaled contaminants and personal exposure in a room using three different air distribution strategies, Indoor Air 22 (1) (2012) 64–76. [12] A.K. Melikov, Advanced air distribution: improving health and comfort while reducing energy use, Indoor Air 26 (1) (2016) 112–124. [13] M.A. Shameri, M.A. Alghoul, K. Sopian, M.F.M. Zain, O. Elayeb, Perspectives of double skin façade systems in buildings and energy saving, Renew. Sustain. Energy Rev. 15 (3) (2011) 1468–1475. [14] X. Li, C. Shen, C.W.F. Yu, Building energy efficiency: passive technology or active technology? Indoor Built Environ. 26 (6) (2017) 729–732. [15] C. Shen, X. Li, Solar heat gain reduction of double glazing window with cooling pipes embedded in Venetian blinds by utilizing natural cooling, Energy Build. 112 (2016) 173–183. [16] C. Shen, X. Li, Thermal performance of double skin façade with built-in pipes utilizing evaporative cooling water in cooling season, Sol. Energy 137 (2016) 55–65. [17] C. Shen, X. Li, S. Yan, Numerical Study on Energy Efficiency and Economy of a Pipeembedded Glass Envelope Directly Utilizing Ground-source Water for Heating in Diverse Climates, Energy Conversion and Management, 2017. [18] C. Shen, X. Li, Potential of utilizing different natural cooling sources to reduce the building cooling load and cooling energy consumption: a case study in urumqi, Energies 10 (3) (2017) 366. [19] S. Tanabe, E.A. Arens, F. Bauman, H. Zhang, T. Madsen, Evaluating Thermal Environments by Using a Thermal Manikin with Controlled Skin Surface Temperature, (1994). [20] F. Causone, F. Baldin, B.W. Olesen, S.P. Corgnati, Floor heating and cooling combined with displacement ventilation: possibilities and limitations, Energy Build. 42 (12) (2010) 2338–2352. [21] M. Dalewski, A.K. Melikov, M. Vesely, Performance of ductless personalized ventilation in conjunction with displacement ventilation: physical environment and human response, Build. Environ. 81 (2014) 354–364. [22] EN15251, Indoor Environmental Input Parameters for Design and Assessment of Energy Performance of Buildings Addressing Indoor Air Quality, Thermal Environment, Lighting and Acoustics, European Committee for Standardization, Brussels, 2007. [23] A. Lipczynska, J. Kaczmarczyk, A.K. Melikov, Thermal environment and air quality in office with personalized ventilation combined with chilled ceiling, Build. Environ. 92 (2015) 603–614. [24] Z.D. Bolashikov, M. Barova, A.K. Melikov, Wearable personal exhaust ventilation: improved indoor air quality and reduced exposure to air exhaled from a sick doctor, Science and Technology for the Built Environment 21 (8) (2015) 1117–1125. [25] C. Shen, N. Gao, T. Wang, CFD study on the transmission of indoor pollutants under personalized ventilation, Build. Environ. 63 (2013) 69–78. [26] EuropeanCommitteeforStandardization EN ISO 7730, Ergonomics of the Thermal Environment - Analytical Determination and Interpretation of Thermal Comfort Using Calculation of the PMV and PPD Indices and Local Thermal Comfort Criteria, (2005) Brussels. [27] REHVA, Displacement Ventilation in Non-industrial Premises, second ed., Federation of European Heating and Air-conditioning Associations, Brussels, 2004.

5. Conclusion The indoor thermal environment and contamination distribution under DV with ceiling cooling/heating and DV with window cooling/ heating were experimentally investigated. Human bioeffluents and passive contaminant sources were considered. The conclusions can be drawn as follows: (1) With a normal window, the strong solar radiation in summer and cold window surface in winter had negative effect on the temperature distribution in DV. The vertical temperature gradient was weakened and the horizontal temperature difference was increased. The pipe-embedded window (PEW) can avoid these drawbacks. (2) The normal DV air pattern in DVCC was destroyed by the solar radiation. The ability of DV flow to remove human bioeffluents was gone. The characteristic of DV in DVHC was weakened by the downdraught from window. The nature of DV was reserved in DVCW and DVHW. (3) Compared with ceiling cooling/heating, the PEW can apply high temperature cooling and low temperature heating and the exhaust air temperature was relatively high in summer and low in winter. (4) A high DV flowrate was more beneficial in PEW than in a normal window condition. The contaminants from armpits were easier to spread than those from groin. Acknowledgement This study was conducted at the International Center for Indoor Environment and Energy (ICIEE) at the Technical University of Denmark and supported by the ICIEE. It was funded by the National Natural Science Foundation of China (Grant No. 51638010 and 51578306). References [1] O. Seppanen, W.J. Fisk, J. Eto, D.T. Grimsrud, Comparison of Conventional Mixing and Displacement Air Conditioning and Ventilating Systems in US Commercial Buildings, Lawrence Berkeley Lab., CA (USA), 1989. [2] H.M. Mathisen, Displacement ventilation ‐ the influence of the characteristics of the supply air terminal device on the airflow pattern, Indoor Air 1 (1) (1991) 47–64. [3] Q. Chen, L. Glicksman, System Performance Evaluation and Design Guidelines for Displacement Ventilation, American Society of Heating, Refrigerating and Air-

10