1、306 2009 ASHRAEThis paper is based on findings resulting from ASHRAE Research Project RP-1373.ABSTRACTTraditional displacement ventilation (TDV) and under-floor air distribution (UFAD) systems have been used widely because they create better indoor air quality. Many previous studies have compared th
2、e TDV or UFAD systems with mixing ventilation systems. This study first compared experimentally the TDV with UFAD systems that use four different diffusers (perforated TDV diffusers, swirl diffusers, linear diffusers, and perforated-floor-panel diffusers) in an environmental cham-ber that can simula
3、te different indoor spaces of the same size. The two systems had higher ventilation performance than the mixing one under cooling mode as well as under heating mode. Also, the systems with low-height-throw diffusers (all except the linear diffusers) were better. This investigation used a vali-dated
4、CFD program to further study the ventilation perfor-mance of the TDV and UFAD systems for an office, a classroom, and a workshop of different sizes. The CFD results further confirmed the findings from the experiment, but with more detailed information and at a lower cost. The air distri-bution effec
5、tiveness with the TDV system and with the low-height-throw UFAD system was in proportion to the ceiling height of the indoor spaces.INTRODUCTIONTraditional Displacement Ventilation (TDV) and Under-Floor Air Distribution (UFAD) systems have been popular in buildings since the 1970s. Many studies have
6、 revealed that TDV and UFAD systems provide better indoor air quality than the traditional mixing ventilation systems (Chen and Glicks-man 2003, Bauman and Daly 2003). Some studies (Hu et al. 1999, Im et al. 2005) reported that the two systems can also reduce energy demand because of their high ener
7、gy efficiency. Typically, a TDV system uses a perforated-sidewall or perforated-corner diffuser that discharges supply air horizon-tally as shown in Figure 1(a). A UFAD system supplies fresh air from a raised floor panel through swirl diffusers as shown in Figure 1(b), linear diffusers in Figure 1(c
8、), and perforated-floor-panel diffusers in Figure 1(d). Among these diffusers, the swirl diffusers and perforated-floor-panel diffusers gener-ate low-height throws, which means that the air velocity from the diffusers decays quickly and becomes less than 0.3 m/s (60 fpm) at 1.35 m (4.5 ft) above the
9、 floor. The linear diffusers generate a high momentum so that the supply air can reach a location at least 1.35 m (4.5 ft) above the floor with an air velocity larger than 0.3 m/s (60 fpm). They also generate high-height throws. The perforated TDV diffuser mounted on the wall or in the corner of the
10、 room discharges fresh air in a hori-zontal direction. This diffuser generates low-height throw due to its flow direction and diffuser structure.On the one hand, the supply air temperature from the TDV or UFAD systems is typically lower than the room air temper-ature for cooling so that the cool and
11、 clean air can stay in the occupied zone. On the other hand, the heat sources in the room generate thermal plumes that bring warm and contaminated air to the upper region to be extracted at the ceiling level. There-fore, the air quality in the occupied zone is normally much better than in the mixing
12、 ventilation. The thermal stratification implies a higher extracted air temperature than in the mixing ventilation. Thus, the TDV and UFAD systems can have higher energy efficiency.Comparison of Airflow and Contaminant Distributions in Rooms with Traditional Displacement Ventilation and Under-Floor
13、Air Distribution SystemsKisup Lee Tengfei Zhang, PhD Zheng Jiang, PhD Qingyan Chen, PhDStudent Member ASHRAE Associate Member ASHRAE Fellow ASHRAEKisup Lee is a student and Qingyan Chen is a professor in the Department of Mechanical Engineering, Purdue University, Indiana. Tengfei Zhang is an assist
14、ant professor in School of Civil and Hydraulic Engineering, Dalian University of Technology, China. Zheng Jiang is a partner of Building Energy and Environment Engineering LLP, Lafayette, Indiana.LO-09-028 (RP-1373) 2009, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc
15、. (www.ashrae.org). Published in ASHRAE Transactions 2009, vol. 115, part 2. For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.ASHRAE Transactions 307Lin et al. (2005) and Kobayashi
16、and Chen (2003) found that diffuser type and ventilation systems could have a signif-icant impact on the ventilation and energy performance of the UFAD systems. Yuan et al. (1998) conducted a similar parametric study on the TDV systems and reached the same conclusions. Thus, unless one can properly
17、select diffusers and ventilation systems, the air distribution created by the TDV or UFAD systems may not always guarantee good performance. For example, Chao and Wan (2004) used linear diffusers with a UFAD system and found that this diffuser may cause discomfort due to high air velocity in the occ
18、upied zone. Hu et al. (2003) also found that a UFAD system may cause significant overcooling near the diffusers.In addition, most studies in the literature assessed the performance of TDV or UFAD systems by comparing each of them with the mixing ventilation. The TDV system is not the same as the UFA
19、D one (McDonell 2003) because of different types of flow driven forces. Unfortunately, little research has been conducted to compare TDV systems with UFAD ones. Furthermore, few comparisons have been made to assess the performance of the various diffusers used in TDV and UFAD systems. It is essentia
20、l to compare the two ventilation systems and to assess the performance of various diffusers for different types of indoor spaces, such as offices, classrooms, and work-shops. This paper addresses these issues in order to assist designers in selecting appropriate ventilation systems and diffusers for
21、 their applications.RESEARCH METHODSThere are two approaches that can be used to compare the performances of different ventilation systems and to assess various diffusers: experimental measurements, and computa-tional simulations by computational fluid dynamics (CFD). In principle, direct measuremen
22、ts give the most realistic infor-mation concerning ventilation performance such as the distri-bution of air velocity, air temperature, relative humidity, and contaminant concentrations. This is why many engineers are still using measurement to evaluate the performance of ventilation systems in field
23、s. However, it is very expensive and time consuming to measure detailed data in an indoor spacewith a TDV or UFAD system. Unfortunately, the time, the amount of measuring equipment, and the size of the test chamber were very limited in this investigation of the ventilation performances of TDV and UF
24、AD systems invarious indoor spaces.On the other hand, CFD simulations can be used as exten-sions of the experiment if these simulations are well calibrated by the experimental data. CFD simulations solve a set of conservation equations for flow, energy, and species concen-trations in an indoor space
25、 and can quickly obtain detailed Figure 1 Common TDV and UFAD systems: (a) TDV with a perforated- sidewall or perforated-corner diffuser, (b) UFAD with swirl diffusers, (c) UFAD with linear diffusers, and (d) UFAD with perforated-floor-panel diffusers.308 ASHRAE Transactionsinformation concerning ve
26、ntilation performance at very little cost. The CFD technique is a powerful tool, but it uses approx-imations to model the flow physics (Versteeg and Malalase-kera 1999).These approximations could bring uncertainties in the numerical results. The uncertainties could come from grid resolution, wall su
27、rface and diffuser boundary conditions, and turbulence models. Zhai and Chen (2004) investigated the impact of grid resolution on the CFD results for indoor airflow. They suggested a grid size of 0.005 m (0.2 in) for the natural convection and 0.1m (4 in) for the forced convection, respec-tively.The
28、 uncertainties due to wall surface boundary conditions come from how the heat transfer is specified: by surface temperature or by heat flux. This investigation used surface temperature, which was not always known. For example, the surface temperature of a lighting fixture was estimated from the powe
29、r input and the radiation between the lighting and the room, but this estimation could produce some errors.The uncertainties caused by the approximations used to simulate a diffuser in the CFD could be significant. This inves-tigation used the momentum method (Chen and Moser (1991) to simulate compl
30、ex diffusers. Srebric and Chen (2002) compared the momentum method with other methods and recommended it for indoor airflows.Since indoor airflows are turbulent, the capacity and speed of existing computers cannot calculate the turbulence details without approximation. The most popular approach is t
31、o simulate the flow details by a turbulence model. Zhang et al. (2007) demonstrated that a turbulence model may work better for one type of indoor airflow but not well for another. There-fore, it is essential to validate the CFD technique by appropri-ate experimental data to ensure that all the appr
32、oximations used are appropriate.This study used experiments to measure the parameters of ventilation performance of the TDV and UFAD systems and to validate the computational simulation for different indoor spaces such as the distributions of air velocity, air temperature, and contaminant concentrat
33、ions in several locations. Since the measured data resolution was not sufficiently fine, it was diffi-cult to obtain a complete picture of the ventilation perfor-mance in these indoor spaces. Then a CFD program was used to further simulate these cases in great detail. The CFD program was also used t
34、o calculate various spaces of different sizes: an office, a classroom, and a workshop.Experimental ApproachThe experiment was conducted by using an environmen-tal chamber as shown in Figures 2(a) and (b). The chamber was a well-insulated room with a window. It had a few pieces of Figure 2 Environmen
35、tal chamber and diffusers used for the TDV and UFAD systems: (a) and ( b) environmental chamber, (c) perforated-corner diffuser, (d) swirl diffuser, (e) linear diffuser, and (f ) perforated-floor-panel diffuser.ASHRAE Transactions 309furniture and some heated boxes that were used to simulated equipm
36、ent and occupants in the room. Table 1 gives some information about the chamber size and enclosures. More detailed information about the furniture and heat sources will be presented in the next section. The chamber can simulate a TDV or a UFAD system with various kinds of diffusers. The TDV system e
37、mployed a perforated-sidewall or perforated-corner diffuser as shown in Figure 2(c) to supply fresh air.The UFAD system utilized a swirl, linear, or perforated-floor-panel diffuser as depicted in Figures 2 (d), (e), and (f), respectively. The specifications of the diffusers provided by the manufactu
38、rers are listed in Table 2. It should be noted that the information provided by a manufacturer may not always be useful for our research. For example, the variation of the throw and the surface distribution of air velocity from a diffuser depends on the supply air temperature and room temperature, w
39、hich are typically not provided by manufacturers. Thus, this investigation obtained the information by additional experi-mental measurements.The chamber has an HVAC system to maintain desirable thermal and flow conditions. Table 3 shows the capacities of the HVAC system, and Figure 3 describes the s
40、chematic of the system.This investigation measured the distributions of air veloc-ity, air temperature, and tracer-gas concentrations. Fifty-four omni-directional hot-sphere anemometers were used to measure air velocity and temperature in the chamber. Table 4 provides detailed information about the
41、sensor probes. The instantaneous air velocity can be used to determine turbulence intensity. A multi-gas monitor and analyzer system, based on the photo-acoustic infrared detection method, was used to measure tracer-gas concentrations in the chamber. The speci-fications of the tracer-gas analyzer ar
42、e listed in Table 5.The air velocity, air temperature, and tracer-gas concentrations were measured at six to eight different vertical positions in nine vertical poles. The experiment used a tracer-gas (SF6or CO2) to simulate a gaseous contaminant generated in the chamber.The measurements were conduc
43、ted under steady-state conditions with constant supply airflow rate, supply air temperature, enclosed surface temperatures, and heat sources in the chamber. The experiment had a good repeatability for the air velocity and air temperature. The only uncertainty came from the measurements of contaminan
44、t concentrations. Unlike the air velocity and air temperature, the multi-gas analyzer can measure concentrations just a few points at a time. Thus, the distributions of the measured tracer-gas concentrations were not obtained at the same time for all the measuring locations. However, the system was
45、run under steady state, so the tracer-gas concentrations measured over time should be valid.Numerical ApproachThe CFD program used in this study solved a set of partial differential equations that can be written in a general form:(1)Table 1. Dimensions and Enclosure Properties of the Environmental C
46、hamberDimensionChamber4.2 m (13.7 ft) wide 4.8 m (15.7 ft) long 2.73 m (8.95 ft) high (including 0.3 m (0.98 ft) high floor plenum height)Window 4.65 m (15.25 ft) wide 1.55 m (5.1 ft) highThermal ResistanceWalls, ceiling, door5.45 Km2/W (30Fft2h/Btu)Floor 5.45 Km2/W (30Fft2h/Btu) Window 0.25 Km2/W (
47、1.42Fft2h/Btu) (double glazing)Table 2. Specifications of Various Diffusers Provided by ManufacturersDiffusers SpecificationsPerforated- corner diffuser- Adequate flow rate: 250 430 m3/h (148 256 cfm)- Total pressure drop: 1.5 mm. w. (0.06 in. w.) at 270 m3/h (160 cfm)Swirl diffuser - Throw heights:
48、 0.73 m (2.41 ft) at 0.3 m/s (50 fpm) with 100 m3/h (60 cfm)- Horizontal discharging range: 0.3 m (1.0 ft) at 0.3 m/s (50 fpm) with 100 m3/h (60 cfm)- Pressure drop: 0.74 mm. w. (0.029 in. w.)Linear diffuser - Adequate flow rate: 85 550 m3/h (50 325 cfm)- Throw height: 2.43 m (8.0 ft) at 0.3 m/s (50
49、 fpm) with 128 m3/h (75 cfm)- Pressure drop: 0.127 mm. w. (0.005 in. w.)Perforated-floor-panel diffuser- Dimension: 600 600 mm (2 2 ft)Table 3. Specifications of the HVAC SystemEnvironmental ChamberSupply Fan 849.5 m3/h (500 cfm)Return Fan 849.5 m3/h (500 cfm)Preheater 8 kW (27,297 Btu/h)Reheater 8 kW (27,297 Btu/h)Humidifier 11.34 kg/h (25 lb/h)Chiller 17.59 kW (6 tons)()t- u()+ 2 S+=310 ASHRAE Transactionswhere, is density, is a dummy scalar variable, t is time, is the velocity vector, is the effective diffusion coefficient, and
