Building and Environment 45 (2010) 1582–1593
Contents lists available at ScienceDirect
Building and Environment
journal homepage: www.elsevier.com/locate/buildenv
A novel approach to enhance outdoor air quality: Pedestrian ventilation system
Parham A. Mirzaei, Fariborz Haghighat*
Department of Building, Civil and Environmental Engineering, Concordia University, Montreal, Quebec H3G 1M8, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 August 2009
Received in revised form
27 October 2009
Accepted 4 January 2010
Higher population density has altered the cities’ old landscape with dense areas consisting of high-rise
buildings. As a result, detrimental phenomena appeared inside modern cities that threatened the
inhabitant’s health and comfort. Among these phenomena, the Urban Heat Island (UHI) is known as the
most harmful side effect of the urbanization which affects the Outdoor Air Quality (OAQ).
In addition to the reduction of wind velocity within the urban canopies, the accumulated pollution
decreases the OAQ and renders the pedestrian areas to hazardous level. According to earlier researches,
the UHI generally shows more intensity in higher aspect ratio (the height of building to street breadth)
canopies which mostly exist in high-density areas. These buildings’ canopies have typically higher
pollution concentration than low-rise residential building canopies due to lower air exchange rate and
heavy traffic load. The situation becomes worst under the stable atmospheric stratification condition
when the canopy ground is colder than the ambient air.
Many passive strategies have been proposed to enhance the OAQ. However, the variety of the UHI
makes the passive mitigation strategies ineffective in some cases. In this paper a novel approach, the
pedestrian ventilation system (PVS), is proposed to ventilate building canopy under various atmospheric
stability conditions: stable, neutral, and unstable. The capability of this system to enhance the pedestrian
level health and comfort parameters (i.e. velocity, temperature and air exchange rate) has been studied
using Computational Fluid Dynamics (CFD) simulation. The results of the simulations confirm that the
PVS can significantly improve the flow regime of the buildings’ canopy.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Urban heat island
Mitigation techniques
Comfort
CFD
Outdoor air quality
Ventilation
1. Introduction
Pedestrian’s health and thermal comfort become pivotal factors
in design and planning of metropolitan areas. Temperature,
humidity, solar radiation, wind velocity, and pollution concentration are known as outdoor comfort indices that are exacerbated by
inappropriate design of the urban landscape. In addition to traffic,
Urban Heat Island (UHI) is contributing to change the building
canopies, major elements of an urban area, to hazardous places.
Many studies have been conducted to understand the effect of
urban feature and meteorological conditions on Outdoor Air
Quality (OAQ) inside and over the building canopy; outdoor
thermal comfort [5,19,22], pollution dispersion [2,4,26], effects of
aspect ratio (the height of building to street breadth) and urban
density on flow pattern [12,24,31], and thermal stratification
[6,32,34]. Furthermore, research has been conducted to enhance
the outdoor environment, especially at pedestrian level. The
planting of trees and the greening of spaces [1,17], the material
* Corresponding author.
E-mail address: [email protected] (F. Haghighat).
0360-1323/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.buildenv.2010.01.001
alteration and shading [7,28], and urban design [21,25] are some
attempts at improving the OAQ.
Although it is possible to significantly improve the UHI and the
pedestrian thermal comfort with widespread implementation of
proposed strategies, it is not feasible to completely diminish the
UHI and to significantly enhance the outdoor comfort within all the
canopies spatially distributed in different points of a city. The
countermeasures may affect the UHI and the pedestrian thermal
comfort in many parts of a city, however, the range of comfort
indices may not be acceptable in some building canopies. As an
example, tree planting is normally a practical method to produce
fresh air inside canopies by the natural transpiration effect of trees
[17]. However, the trees themselves can become obstacles to the
vertical or horizontal air movement which benefits the air pollution
removal, thermal comfort, and the air conditioning of the building
[20]. Moreover, they may shade the buildings in winter time and
causes more energy consumption for heating purposes. All these
aspects can be summarized as shortcomings of the passive strategies to control the OAQ. Nevertheless, it is obvious that the quality
of outdoor environment has an important role on the quality of
indoor environment [10,27]. Therefore, there is an urgent need to
develop active strategy to improve the quality of outdoor air.
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
1583
on controlling pedestrian health and comfort parameters. For this
purpose a three-dimensional building canopy with the aspect ratio
of two is considered for this investigation. Moreover, the performance of the system is investigated under different atmospheric
stratification conditions ranging from stable to unstable.
2. Pedestrian ventilation system
2.1. System configuration
Fig. 1. New design approach: Pedestrian Ventilation System (PVS).
In this article a novel mitigation technique called the Pedestrian
Ventilation System (PVS) is proposed to actively control the pedestrian health and thermal comfort inside the canopies. Preliminary
investigation is conducted to demonstrate the feasibility of the PVS
Three flow patterns are characterized based on geometry of
a building canopy [24]: isolated roughness flow (IRF), wake interference flow (WIF), and skimming flow (SF). Many studies have
been carried out to find corresponding threshold for aspect ratio of
these flows [16,33]. It is found that the threshold from WIF to SF is
around 0.7 (H/W). Moreover, after a certain aspect ratio (around
1.5) the main vortex in SF will be deformed to two vortices [16].
Again, the vortices number will be increased by raising aspect ratio
of the building canopy. As most of available studies in literature,
this research focuses on the skimming flow where consistent
vortex/vortices that retain pollution and result weak air quality
develop at the building canopy.
In building canopies mean wind speed is not only an important
parameter in air exchange, turbulence also plays a significant role
on canopy ventilation, especially in skimming flow. Kim and Baik
[11] demonstrated the importance of turbulence on the removal of
pollutants. On the other hand, it is proven that buoyancy can
increase or dominate turbulence inside some canopies by changing
Fig. 2. Different strategies of the PVS.
1584
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
Pedestrian
Ventilation
System
Prevailing Wind
A
A
B
B
H
L
Pedestrian
Ventilation
Zone
Upward/
Downward
Flow
Dampers
W
W
Pedestrian Walking
Area
Lateral Flow
Fig. 3. (Left) Homogeneous array of buildings - (Right) Pedestrian ventilation system and pedestrian walking area.
the street and the buildings’ wall temperature [35]. This usually
happens under different atmospheric stability conditions.
Therefore, to control the air movement and pollutant dispersion
inside building canopies, it is necessary to modify the air movement created by turbulence and buoyancy by imposing a controlled
air movement. This air movement is different from unpredictable
and stochastic vortex which is created by the top-canopy prevailing
wind. It is postulated that the required air movement is obtainable
with an active control system in the form of a pedestrian ventilation
system.
As shown in Fig. 1, the PVS induces air movement in a region
near the ground, the Pedestrian Ventilation Zone (PVZ), using
ventilation ducts. The PVZ volume is extended around the building
up to 3 m in height (sidewalks or region in which most pedestrian
activities occur). The mechanism for ventilation is based on guiding
air through a designed vertical duct system from roof of the
building to the surrounding street level.
In stratified situation, the street temperature is lower than the
prevailing wind temperature, and thus the pollutant is mostly
accumulated in the PVZ. In this case, the PVS can replace the
pedestrian level air with fresher air from the top-canopy level. On
the other hand, this system is also useful to accelerate the movement of cooler air from the top-canopy to the PVZ where the
weather is under unstable condition (when the prevailing wind
temperature is colder than the canopy temperature). Therefore, the
pedestrian air velocity, temperature and pollution concentration
can be placed under control by changing the airflow rate within the
building canopy: both natural and force convection can be used to
provide the required pressure gradient for the system.
Heating the duct can be used to provide the required air
movement (stack flow). The required energy for heating can be
provided by heat exhausted from condenser of the air conditioning
systems, and/or solar energy which is mostly available during
severe heat island episode. Alternately, force convection can be
achieved using supply or exhaust fan.
When the ambient air relative humidity is not within the
thermal comfort range, the pedestrian ventilation system can
Table 1
Proposed case studies based on thermal wind tunnel experiment by Uehara et al.
[32].
Case I
Case II
Stability
condition
Bulk-Richardson
number
Ta ¼ Wind
temperature (K)
Tf ¼ Ground
temperature (K)
Stable
Unstable
0.89
0.18
351
293
294
352
humidify the PVZ with some water sprays (Fig. 1). Solar radiation
can also be prevented by placing flexible pergolas (Fig. 1). In this
research, however, only applicability of the PVS in air removal and
changing air velocity and temperature within the building canopy
has been studied using electrical fan.
2.2. Combined pedestrian ventilation system
It is feasible to have various way of integrating the PVS inside
a canopy by installing two systems on adjacent buildings (Fig. 2).
These systems strengthen or weaken vortex/vortices of the
building canopy. Strategy (A) uses two exhaust fans to intensify
a downward flow. In strategy (B), an upward flow toward the topcanopy can be achieved using two supply fans. Strategies (C) and
(D) are capable of establishing a washing flow through one sidewalk to another using a supply and an exhaust fan. It is noteworthy
that closest vortex to the ground is either clockwise or counterclockwise depending on aspect ratio and number of vortices. Thus,
always one of the strategies (C) or (D) is strengthening the flow and
one is weakening that. Obviously, the required pedestrian comfort
situation is an important factor in order to choose the effective
strategy. This flexibility is investigated in the following sections
under both stable and unstable conditions.
2.3. Air exchange concept
The proposed PVS in this paper is a strategy to enhance
pedestrian health and thermal comfort, during heat island and
stratification period, using any available energy sources such as
electricity which is used in this research. The air exchange rate
(ACH) concept [3] is used to quantify air movement from the PVZ.
The air exchange rate is defined as the total air that is entering
Table 2
Boundary conditions and solution schemes.
Inflow boundary
Outflow boundary
Ground boundary
Upper and side surface
of domain
Building surface boundary
Turbulent scheme
Momentum discretization
Computational domain
Logarithmic flow from experiment [30]
Zero gradient assumption
Logarithmic law with roughness
length (0.024 m)
Free slip wall condition
Logarithmic law for smooth wall
Standard k 3
Second order Upwind
180 m(x) 280 m(y) 120 m(z)
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
Inflow
40
47
54
37
41
48
55
20 26 29 32
35 38
42
49
56 62
27 30 33
36 39
4
11
18
1
5
12
19
2
6
13
3
25 28
31 34
7
14
21
43
50
57
8
15
22
44
51
58
9
16
23
45
52
59
10
17
24
46
53 60
61
63
Fig. 4. Half-domain measured points by Tominaga et al. [30] located 2 mm above the
ground (Top view).
(ACHþ) or leaving (ACH) from the lateral and top faces of the PVZ
(Fig. 3 - Right). The magnitudes of ACHþ and ACH from PVS
dampers and top and lateral surfaces of the PVZ are equal due to the
mass conservation law:
ACH ¼ ðACHÞ þ ðACHþÞ ¼ ACH þ ACH0
(1)
ACHþ ¼ ðACHLateral þ Þ þ ACHTop þ þ ðACHPVS þÞ
(2)
ACH ¼ ðACHLateral Þ þ ACHTop þ ðACHPVS Þ
(3)
Without using PVS, ACHPVS is apparently zero in equations (2) and
(3). Also, as shown in equation (1), ACH is made up of two parts:
mean and fluctuation velocity. Vertical component (w) of these
velocities has significant magnitude in the top-surface air
exchange, while horizontal component (v) which is perpendicular
to the flow is important in two lateral surfaces. Various twodimensional studies have been carried out to establish criteria for
ACH which are only based on equation (1). Nonetheless, a well
defined criterion is not available for three-dimensional ACH within
the building canopies, and as presented later, a two-dimensional
concept is adapted to investigate air movement in this study. This
air movement, ACH, can be later used to understand the pollution
exchange rate (PCH) of the canopy [18]. In this study, it is assumed
that supply and exhaust fans control airflow rate of the PVS.
3. Methodology
3.1. Case study
A simple case study is chosen to show the potential of the PVS in
urban areas. An array of buildings with simple geometry has been
selected with a PVZ volume of 1200 cubic meter
(20 m(x) 20 m(y) 3 m(z)) for each building canopy (y is in the
direction of prevailing wind). As demonstrated in Fig. 3 - Left, the
1585
urban landscape has been assumed as homogenous cuboid buildings with aspect ratio of two (H ¼ 40 m and L ¼ W ¼ 20 m). The PVS
is also applied to a canopy between buildings (A) and (B) as
depicted in Fig. 3.
To include weather stability condition, simulations have been
conducted separately for both stable and unstable conditions. Since
there is a wide variation in wind speed and temperature values
below
the
building
height,
Bulk-Richardson
number
ðRb ¼ gHðTH Tf Þ=fðTÞðUH Þ2 gÞ has been introduced in the
building canopy literature as an appropriate dimensionless number
to represent stability weather conditions [32]. In this equation g (m/
s2) is the acceleration due to gravity, TH (K) is temperature at the top
of the building canopy, Tf (K) is temperature at the ground level, T
(K) is the mean temperature, and UH (m/s) represents the mean
wind speed at top of building canopy.
As shown in Table 1, two Bulk-Richardson numbers, 0.89 and
0.18 have been chosen which respectively represent stable and
unstable weather conditions. These numbers are in the range of
thermal wind tunnel test (0.79 and 0.21) which was carried out by
Uehara et al. [32]. The main concern in case (I), stable condition, is
to remove the accumulated pollution from the PVZ that mostly
occurs during nocturnal non-cloudy calm weather. In contrast, case
(II) is related to unstable situations where the priority is to take
advantage of the colder prevailing wind flowing over the canopy.
Illustrated in Fig. 2, four strategies have been applied in each case
study in this paper.
In this study, the cross-section of the ducts is assumed to be
rectangular for both cases installed on both buildings (A) and (B).
Three dampers are also considered inside each sidewalk with an
area of 1 m square (Fig. 3). Moreover, the duct surface is assumed to
be well insulated to prevent any heat transfer. To provide the
required airflow for ventilating the PVZ around once per minute
(10–20 m3/s), adequate supply and/or exhaust fans with pressure
differences of 100 pa are also assumed. Thus, the induced air
velocity by PVS dampers remains between 1 and 2.3 m/s which is
inside the light breeze norm [14]. Furthermore to simplify the
calculation, radiation modeling and humidity calculation are
neglected in this study.
Many parameters contribute to the PVS performance, including
building canopy aspect ratio and orientation, prevailing wind
velocity and its direction, cloud cover, and the PVS design. Thus,
various studies are necessary to understand the influence of the
above mentioned parameters on the pedestrian comfort.
3.2. Solution scheme
The PVS has been simulated using Computational Fluid
Dynamics (CFD) approach: FLUENT software was used in this study
Fig. 5. Comparison between measurement and CFD with different mesh size.
1586
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
Fig. 6. Comparison between measurement and CFD with different domain fetch.
[8]. Around 320,000 structured meshes have been generated using
a commercial package, GAMBIT [9]. Finer resolution has been
applied to monitor the performance of the PVS within the target
building canopy. Also, half of the domain has been computed due to
symmetry of the cases. Steady scheme has been used with standard
k 3 model for turbulent closure. Tominaga et al. [30] used
different turbulence models and concluded that the standard k 3
model provides almost the same result as the other models except
for the circulating flow region behind the building. Presented case
studies in Table 1 equipped with the PVS, shown in Fig. 3 - Right,
originated with the initial condition as tested by Uehara et al. [32].
The boundary conditions and solution schemes are also given in
Table 2.
3.3. Mesh and domain size test
The CFD model was firstly verified with wind tunnel experiment
for case (C) of AIJ [30]. This test has been conducted to an array of
building similar to Fig. 3 (L ¼ H ¼ W ¼ 0.2 m). An inflow velocity has
been chosen with a recorded log-profile. The wind velocity has
been measured at 63 points located 2 mm above the ground surface
(Fig. 4). Several cases have been simulated to find the appropriate
mesh size, fetch length and vertical height in order to optimize
mesh numbers, and decrease the computational cost of the final
study.
Different mesh sizes have been tested to find the proper
dimension to simulate the study domain, these included 0.2 H,
0.25 H, and 0.3 H. The result clearly demonstrates that a 0.25 H
mesh size is good enough to model the case study (Fig. 5). A similar
size is applied by [36]. They concluded that an appropriate mesh
size is around H/10. The reason that 0.2 H for mesh size is not better
than 0.25 H is related to the wall-function assumption for the walls
[13]. In the wall-function approach, semi-empirical formulas, the
viscous sub-layer and the buffer layer is not resolved. This means
that the wall-function is used to connect the viscosity-effected
region between the wall and the fully-turbulent region.
It is necessary for the horizontal computational domain to be
maintained on a certain length extending outside of the urban
block border. As illustrated in Fig. 6, three cases are compared with
fetch sizes of 2 H, 4 H, and 10 H. It is obvious from this figure that
the results do not change significantly when the fetch length is
increased from 4 H to 10 H.
Also, as demonstrated in Fig. 7, a height of 5 H provides almost
the same result as the case where the height is 6 H (wind tunnel
height). This conclusion is corroborated by Tominaga et al. [29]:
they suggest a vertical domain height of 3 H or more.
From results shown in Figs. 5–7, it can be concluded that
a domain with a fetch size of 4 H, a height of 5 H, and a mesh size of
0.25 H may optimize the computational cost of the simulation.
Therefore, the case (C) was again simulated with the obtained
mesh, height and fetch size (around 150,000 meshes in total). As
shown in Fig. 8, air velocities are in good agreement to wind tunnel
measurements.
3.4. Model verification
Choosing appropriate strategy of the PVS significantly depends
on the air stability regime. Therefore, it is necessary to include this
effect in the simulations. To verify the thermal stratification of the
building canopy, the simulation was verified with thermal wind
tunnel experiment by Uehara et al. [32]. The test was performed at
the National Institute for environmental Studies of Japan [23]. The
turbulence was modeled with an array of Styrofoam cubes. Also,
stratification was produced by changing the ground and air inflow
Fig. 7. Comparison between measurement and CFD with different domain height.
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
1587
Fig. 8. Comparison between measurement and CFD with suggested mesh size, domain height and fetch length.
temperature similar to Table 1. The buildings consisted of 0.1 m
cubes placed 0.1 m and 0.05 m apart along the length and width of
the tunnel, respectively. The ground and air temperature were set
in order to attain Bulk-Richardson number of 0.21 and 0.79 for
unstable and stable stratification, respectively. Fetch length and
mesh size of the domain were respectively chosen 4 H and 0.25 H
based on last section test. Domain height, however, was selected
10 H to capture the thermal stratification effect. The number of
meshes was around 350,000; k 3 model was used as turbulent
model. As shown in Fig. 9, the simulation results agreed well with
the measurements. Here U700 is attributed to air velocity at height
of 0.7 m from bottom of the target building canopy in the wind
tunnel experiment.
4. Results and discussion
The result for various strategies of the PVS integration is illustrated in Table 3. Presented airflow rate in this table is attributed to
air entering (positive number) or leaving (negative number) top
pedestrian zone and two lateral faces (Fig. 3-Right). All numbers are
normalized by PVZ volume per minute (Q ¼ PVZ/minute).
In stable weather conditions (Rb ¼ 0.89), a balance exists
between entering air from the top-surface (ACHTop ¼ 0.66) and
leaving air from lateral surfaces (ACHLateral ¼ 0.66) of the pedestrian zone. This air circulation is much stronger in unstable
condition where the entering and leaving airflow rates are 2.41 and
2.41, respectively. Apparently, the PVS is capable to produce a fair
Fig. 9. Comparison between measurement and CFD (Left) Velocity (Right) Temperature.
1588
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
Table 3
Air removal capability (normalized by Q) for the PVS strategies.
Stable condition (Bulk-Richardson ¼ 0.89)
Unstable condition (Bulk-Richardson ¼ 0.18)
Strategy
No PVS
(A)
(B)
(C)
(D)
No PVS
(A)
(B)
(C)
(D)
ACHLateral
ACHTop
ACHPVS
0
ACHTop
0.66
0.66
0.00
1.20
0.44
1.12
0.68
1.37
0.85
0.11
0.74
1.83
0.68
0.63
0.05
1.55
0.73
0.68
0.05
1.98
2.41
2.41
0.00
3.39
1.21
1.81
0.60
2.57
2.79
2.18
0.61
3.19
1.49
1.45
0.04
2.43
2.69
2.69
0.00
3.24
1.2
0.098
0 .0 7 7
1
0 .0
77
77
0.6
0 .0
z /H
0.8
0.4
-0.5
-0.25
0
0 .0 3 2
0.0 20
0
07
0.050
0 .0 0 9
0.
0.2
0
0.25
x/H
0.5
Without PVS
A
B
1.2
1
0.100
0.0 68
1.2
0.080
1
0.0 90
0.071
0 .0
0.080
z /H
z /H
68
0.068
0.6
0.0
.6
58
0.4
0 .0
58
0.4
0.2
0 .0 0
9
0.029
-0.25
0.02 9
0
0.25
x/H
0.5
1.2
1
0.080
0.1 00
0.060
D
-0.25
0
x/H
0 .1
0.1 00
00
0 .0
0.8
67
.6
0.25
0.5
1.2
1
z/H
0 .0
80
-0.5
0 .0
72
72
0.
0.6
06
6
0 .0
67
z/H
0.8
0
0.
0.014
0.037
0 .0 5
C
0
0.037
0
-0.5
7
71
0 .0 1 7
0
0 .0 3
0 .0 5 8
0 .0
0.2
71
0.8
0 .0
0.8
0 .0
0
x/H
0.25
0.5
0
0 .0
-0.5
-0.25
0 .0 66
0.03 9
0. 02 1
72
2
8
-0.25
02
-0.5
5
0.035
0.
0
0.2
0 .0 3
0 .0
0.0 37
0.2
0.4
0 .0
0.060
0 .0 5 4
0.0 17
0.4
0.050
0
x/H
0.25
0.5
Fig. 10. Spatial contours of K/U2H for various PVS strategies inside the building canopy under unstable condition (Rb ¼ 0.18).
7
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
1589
1.2
05
32
0
1
0.032
0.8
z /H
0 .0
0.
0.050
0 .0 2
0.6
5
0.4
0.016
0.012
0.2
0.008
0.006
0
-0.5
-0.25
0
0.25
x/H
0.5
without
B 1.2
1.2
0.06 1
6
0 .0 6 1
1
0.
0
04
0 .0
0.8
6
37
02
0 .0
0.
0.4
12
0.0 17
0.
02
0.4
0.6
31
0.032
0.008
0.01 0
-0.25
0.00 8
0.006
0
0.25
x/H
0.
0
0.25
x/H
0.5
z/H
0 .0 22
0.03 8
0.0 56
0.018
0 .0 1 7
-0.5
0
-0.25
22
0
x/H
0.2
0.008
.
1
03
0.25
0.5
0
0.035
0.026
4
0 .0
0 .0 2
0.2
5
.6
0.4
0.031
0 .0 3
0.05 6
.8
38
0.01 7
0.4
1
5
.6
03 8
03
0.
0 .0
0
-0.25
1
03
0.03 5
-0.5
-0.25
0 .0
33
0.026
0
x/H
0.006
0.050
0.8
z /H
-0.5
0.
D 1.2
1.2
1
0
0.5
0 .0 3 7 0.026
17
C
-0.5
0.034
0 .0
0
0.2
0 .0
0.2
37
0 .0
z/H
0.6
0
z /H
0.8
85
0 .0 2
0 .0
0.060
1
0.018
A
0.25
0.5
Fig. 11. Spatial contours of K/U2H for various PVS strategies inside the building canopy under stable condition (Rb ¼ 0.89).
air exchange within the PVZ where supply and exhaust fans are
used. This means that these strategies can double the air exchange
of the entire PVZ volume (1200 m3). For example, strategy (A)
entrains more fresh air from top-PVZ (around 70%) to the PVZ
under stable condition using exhaust fan. The air exchange ratios
(ACH (No PVS)/ACH (Strategy A, B, C, or D)) are generally lower
under unstable conditions due to the stronger circulation of the
building canopy which partly opposes the induced flow by the PVS.
Moreover, fan air exchange rate (ACHPVS) has the same order of
magnitude with ACHLateral and ACHTop in stable case, however, the
ACHPVS is around four times smaller than ACHLateral and ACHTop in
unstable case. This can be compensated using more powerful fan.
Generally, it can be concluded that strategy (A) reduce horizontal
air exchange inside the PVZ and therefore, entrains more air from
top-PVZ. On the other hand, strategy (B) decreases the vertical air
exchange. Also, strategies (C) and (D) change the magnitude of the
initial balance between the vertical and horizontal air exchange
rate imposing a washing flow from one sidewalk.
As mentioned earlier, turbulence has significant effect on the air
exchange of PVZ. Nonetheless, only few works have been carried
1590
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
out to analyze this effect and most of them were two-dimensional
studies in which the influence of lateral surfaces is neglected. In this
study, the approach proposed by Li et al. [15] is used to calculate the
turbulence at top-PVZ surface. They assumed the transient air
exchange is divided evenly into entering and leaving parts. There0
can be determined as follows:
fore, ACHTop
1
ACHTop þ ¼ ACHTop ¼
2
0
0
Z
G
w00 w00 1=2 dG
(4)
roof
where w00
is the vertical velocity fluctuation. They also assumed the
isotropic condition for turbulence because of the high-Reynolds
number ðw00 w00 ¼ v00 v00 ¼ u00 u00 Þ:
k ¼ ðw00 w00 þ v00 v00 þ u00 u00 Þ=2 ¼ 3w00 w00 =2
0
1
ACHTop þ ¼ pffiffiffi
6
Z qffiffiffiffiffiffiffiffiffiffiffi
kjroof dG
(5)
(6)
G
Depicted in Table 3, however, in this study K is calculated on the
PVZ top-surface which may cause some discrepancy.
Obviously, turbulence fluctuation considerably changes in
different strategies. For example, this number has been altered more
than 50% in strategy (B) of stable case. Figs. 10 and 11 also demonstrate the normalized K by square of air inflow velocity (U2H) in midplane of the building canopy without and with PVS strategies. Again,
it is evident that the turbulence kinetic energy roughly varies within
the PVZ. The value inside the sidewalks is extremely weak in stable
case; nonetheless it is remarkably increased using the PVS strategies.
In addition to the air removal ability, providing appropriate
velocity is another task of the PVS. Fig. 12 illustrates the vertical
profile of the velocity normalized by the inflow velocity (UH) in
middle of the pedestrian sidewalks (1 m from walls), located at the
center plane of the building canopy. The left and right graphs
respectively depict the right-sidewalk and left sidewalk. The prevailing wind is from left to right as demonstrated in Fig. 3. In
addition to applying different PVS strategies, the skimming flow air
circulation inside the canopy initiates an asymmetry between these
two sidewalks.
Normally without the PVS, the vertical velocity profile tends to
be the same under both stable and unstable conditions, even
though the velocity magnitude is higher during unstable situations.
This higher magnitude is related to the stronger initial air circulation (vortex).
The maximum velocity in Fig. 12 is related to the place of
dampers where air is supplied or exhausted. Similar to strategies
(B) and (D), the vertical air velocity profile in the right-sidewalk is
the same for strategies (A) and (C). This is due to the existence of
a similar supply or exhaust fan in these cases. Correspondingly, in
the left-sidewalk, the vertical velocity profile is almost the same for
strategies (A)–(D) and (B)–(C). However, the velocity magnitude of
strategy (D) is higher than strategy (A) in this case due to the
direction of the horizontal washing flow produced by synchronizing of the supply and exhaust fans. This means that vortex flow
(a secondary weak clock-wise vortex opposing with main counter
clock-wise vortex of the building canopy) is intensified in the leftsidewalk for strategy (D). However, this effect is not seen for
strategy (C) on the right-sidewalk (no considerable difference
between strategy (A) and (C)). The reason is that vortex flow does
Fig. 12. Vertical Air velocity profile comparison for various PVS strategies through the pedestrian walking area (Left) Left-sidewalk of the canopy (Right) Right-sidewalk of the
canopy.
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
1591
Fig. 13. Horizontal air velocity profile comparison for the PVS strategies through the pedestrian walking area (Left) Left-sidewalk of the canopy (Right) Right-sidewalk of the canopy.
not considerably affect the exhaust flow in this case. It can be
generally concluded that the vertical velocity profile is slightly
affected by the stability condition inside pedestrian sidewalks. On
the contrary, velocity magnitude is highly dependent on the
stability situation.
Horizontal velocity profiles normalized by inflow velocity (UH)
through a line located 1.5 m above the ground (parallel to the
building canopy surface) in middle of right-sidewalk and leftsidewalk (1 m from walls) are demonstrated in Fig. 13. The graphs
only show half of the canopy (due to the symmetry condition)
where the zero in the (y) direction signifies the middle of the
canopy.
The peak values for air velocity in these graphs are assigned to
the outlet dampers of the PVS. In this research three outlet dampers
were used, and it is obvious that increase of the damper’s number
will produce smoother velocities through the sidewalks. It seems
that the effect of atmospheric stability is not significant on the
horizontal velocity profile. However, vortex circulation within the
building canopy has little influence on the velocity magnitude
through the left-sidewalk. This effect is much weaker in rightsidewalk.
Air supplying strategies on the right-sidewalk, (B) and (D),
change the velocity ratio from approximately 0.1 to a maximum of
0.5. Strategy (B) has the same effect on the left-sidewalk. However,
strategy (D) cannot produce a similar velocity because of the
mixing of the supplied air with existed circulation regime of the
building canopy. It can be seen that the mixing is reduced under
stable conditions where the vortex circulation regime is weak.
Generally, the exhaust strategy, (A), provides smooth velocity in
both right and left sidewalks, but normally with a lower velocity
magnitude.
The vertical air temperature profile in the middle of the building
canopy is shown in Fig. 14. Obviously, the air temperature does not
alter considerably above the pedestrian area. However, the air
temperature fluctuation mostly occurs in pedestrian level as
depicted in Fig. 15.
Fig. 15 shows intense air temperature fluctuations inside sidewalks. These results are again related to the middle of the right-side
and left-side pedestrian zones (1 m from walls) situated 1.5 m
above the ground level. Also, the intensive increase and decrease of
the air temperature again are related to place of the dampers.
When atmospheric conditions are unstable, it means that the
prevailing wind is cooler. Therefore, as demonstrated in Fig. 15,
strategies (B) and (D) are appropriate to reduce sidewalk’s air
temperature (greater number of (TTf)/(TaTf)). This temperature
decrease is more predominant in the left-sidewalk than rightsidewalk. Strategy (C) is evidently not capable of making a significant change in the air temperature of the zones. Although using an
exhaust fan in strategy (A) increases the air temperature of the leftsidewalk pedestrian zone, it reduces that of on the right-sidewalk.
Using a warmer air temperature increases the pedestrian zone
temperature in stable cases. Therefore, strategy (A) is the best
technique to avoid the warmer prevailing wind temperature
(smaller number of (TTf)/(TaTf)).
Generally, it can be concluded that strategy (A) provides a fair air
removal from the canopy. Although the velocity comfort parameter
1592
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
Fig. 14. Air temperature profile comparison of the PVS strategies in middle of the building canopy (Left) Unstable case (Right) Stable case.
is poor in this strategy, it is more practical than strategy (B) in
increasing air exchange. The reason is that some part of the
supplied air returns to the canopy by using strategy (B).
On the other hand, strategies (C) and (D) are appropriate options
when there is a high source of pollution (e.g. vehicles and pedestrians). With these options a proper air movement in addition to
cooler temperature within the pedestrian walking area can be
obtained, especially when the focus is on one side of the street.
Although air exchange rate is weak in these techniques, a strong
washing flow can be produced from one sidewalk to another.
Under the above mentioned assumption of this research, it can
be concluded that strategy (B), air supply mechanisms, produce
proper conditions for air velocity, temperature and air removal
indices. Both mean and fluctuation velocities are considerable in
this case. The air temperature decreases under unstable atmospheric conditions and remains almost constant under stable
condition. Also, the air velocity advances significantly in both
pedestrian sidewalks from a light-air situation to a light breeze
norm [14].
Although two cases, under stable and unstable weather situation, have been investigated, the presented results are not unique
answers of this system. As mentioned previously, the benefit of this
system is its adaptability under various flow regimes. For example,
the vortex circulation regime changes majorly in higher building
Fig. 15. Air temperature profile comparison of various PVS strategies on pedestrian walking area (Left) Left-sidewalk of the canopy (Right) Right-sidewalk of the canopy.
P.A. Mirzaei, F. Haghighat / Building and Environment 45 (2010) 1582–1593
aspect ratios, and this may cause different air movement than
discussed in this study. Future research will focus on a parametric
study of the pedestrian ventilation system to propose in situ control
strategies for the pedestrian ventilation zone.
5. Conclusion
A novel building canopy ventilation system, pedestrian ventilation system, has been introduced to improve pedestrian level air
quality, especially within high-rise building areas. An active control
technique is used in this system to enhance human comfort
parameters including, the air exchange, temperature and velocity
inside the pedestrian sidewalks. As its energy source, this system is
capable of using electricity and solar energy, and even waste energy
from the building’s condenser units. Effective system performance
parameters are categorized as street design parameters (i.e. ducting
design, energy source used, aspect ratio, etc.) and meteorological
conditions (i.e. weather stratification regime, wind velocity, etc.).
To obtain basic knowledge about the PVS and understand the
effectiveness of the above mentioned parameters, firstly the model
has been verified with two wind tunnel experiments. To show the
applicability of the proposed strategies two case studies have been
carried out using CFD simulation. The results fairly show the ability
of this system to provide air movement inside the building canopy
under stable and unstable weather condition considering the air
exchange rate criteria. An increase in this parameter can improve
pedestrian comfort, particularly during severe heat island episode.
Moreover, the simulation demonstrated that this system can bring
top-canopy air to the PVZ when the atmosphere is unstable.
To propose the PVS as a practical ventilation system, more
experimental and simulation based on influential parameters are
needed. Also, to achieve more realistic results, coupling of heat
storage, humidity and radiation models with CFD simulation is
strongly recommended.
Acknowledgements
The authors would like to express their gratitude to the Natural
Science and Engineering Research Council Canada (NSERC), and
Concordia University for their financial support.
References
[1] Alexandria E, Jones P. Temperature decreases in an urban canyon due to green
walls and green roofs in diverse climates. Building and Environment
2008;43:480–93.
[2] Bady M, Kato S, Huang H. Towards the application of indoor ventilation efficiency indices to evaluate the air quality of urban areas. Building and Environment 2008;43:1991–2004.
[3] Bentham T, Britter R. Spatially averaged flow within obstacle arrays. Atmospheric Environment 2003;37:2037–43.
[4] Cermak JE. Thermal effects on flow and dispersion over urban areas: capabilities for prediction by physical modeling. Atmospheric Environment
1996;30:393–401.
[5] Chen H, Ooka R, Harayama K, Kato S, Li X. Study on outdoor thermal environment of apartment block in Shenzhen, China with coupled simulation of
convection, radiation and conduction. Energy and Buildings 2004;36:1247–58.
[6] Cheng WC, Liu C-H, Leung DYC. On the correlation of air and pollutant
exchange for street canyons in combined wind-buoyancy-driven flow.
Atmospheric Environment 2009;43:3682–90.
[7] Doulos L, Santamouris M, Livada I. Passive cooling of outdoor urban spaces.
The role of materials. Solar Energy 2004;77:231–49.
[8] Fluent, http://fluent.com/; 2008.
[9] GAMBIT, http://fluent.com/software/gambit/; 2008.
[10] Hoppe P. Different aspects of assessing indoor and outdoor thermal comfort.
Energy and Buildings 2002;34:661–5.
1593
[11] Kim J-J, Baik J-J. Effects of inflow turbulence intensity on flow and pollutant
dispersion in an urban street canyon. Journal of Wind Engineering and
Industrial Aerodynamics 2003;91:309–29.
[12] Kubota T, Miura M, Tominaga Y, Mochida A. Wind tunnel tests on the relationship between building density and pedestrian-level wind velocity:
development of guidelines for realizing acceptable wind environment in
residential neighborhoods. Building and Environment 2008;43:1699–708.
[13] Launder BE, Spalding DB. Numerical computation of turbulent flows.
Computer Methods in Applied Mechanics and Engineering 1974;3:269–89.
[14] Lawson TV, Penwarden AD. The effects of wind on people in the vicinity of
buildings. 4th International Conference on wind effects on buildings and
structures. Cambridge University Press; 1975. pp. 605–622.
[15] Li X-X, Liu C-H, Leung DYC. Development of a k-[epsilon] model for the
determination of air exchange rates for street canyons. Atmospheric Environment 2005;39:7285–96.
[16] Li X-X, Liu C-H, Leung DYC, Lam KM. Recent progress in CFD modelling of wind
field and pollutant transport in street canyons. Atmospheric Environment
2006;40:5640–58.
[17] Lin B, Xiaofeng Zhu L, Qin Y. Numerical simulation studies of the different
vegetation patterns’ effects on outdoor pedestrian thermal comfort. Journal of
Wind Engineering and Industrial Aerodynamics 2008;96:1707–18.
[18] Liu C-H, Leung DYC, Barth MC. On the prediction of air and pollutant exchange
rates in street canyons of different aspect ratios using large-eddy simulation.
Atmospheric Environment 2005;39:1567–74.
[19] Mochida A, Lun IYF. Prediction of wind environment and thermal comfort at
pedestrian level in urban area. Journal of Wind Engineering and Industrial
Aerodynamics 2008;96:1498–527.
[20] Mochida A, Tabata Y, Iwata T, Yoshino H. Examining tree canopy models for
CFD prediction of wind environment at pedestrian level. Journal of Wind
Engineering and Industrial Aerodynamics 2008;96:1667–77.
[21] Murakami S. Environmental design of outdoor climate based on CFD. Fluid
Dynamics Research 2006;38:108–26.
[22] Nikolopoulou M, Lykoudis S. Thermal comfort in outdoor urban spaces:
analysis across different European countries. Building and Environment
2006;41:1455–70.
[23] Ogawa Y, Diosey PG, Uehara K, Ueda H. A wind tunnel for studying the effects
of thermal stratification in the atmosphere. Atmospheric Environment
1981;15:807–21.
[24] Oke TR. Street design and urban canopy layer climate. Energy and Buildings
1988;11:103–13.
[25] Priyadarsini R, Hien WN, Wai David CK. Microclimatic modeling of the urban
thermal environment of Singapore to mitigate urban heat island. Solar Energy
2008;82:727–45.
[26] Sarrat C, Lemonsu A, Masson V, Guedalia D. Impact of urban heat island on
regional atmospheric pollution. Atmospheric Environment 2006;40:1743–58.
[27] Stathopoulos T, Wu H, Zacharias J. Outdoor human comfort in an urban
climate. Building and Environment 2004;39:297–305.
[28] Synnefa A, Santamouris M, Apostolakis K. On the development, optical
properties and thermal performance of cool colored coatings for the urban
environment. Solar Energy 2007;81:488–97.
[29] Tominaga Y, Yoshie R, Mochida A, Kataoka H, Harimoto K, Nozu T. Cross
comparisons of CFD prediction for wind environment at pedestrian level
around buildings. part 2: comparison of results for flowfield around building
complex in actual urban area, The Sixth Asia-Pacific Conference on Wind
Engineering, Korea; 2005.
[30] Tominaga Y, Mochida A, Yoshie R, Kataoka H, Nozu T, Yoshikawa M, et al. AIJ
guidelines for practical applications of CFD to pedestrian wind environment
around buildings. Journal of Wind Engineering and Industrial Aerodynamics
2008;96:1749–61.
[31] Uehara K, Wakamatsu S, Ooka R. Studies on critical reynolds number indices
for wind-tunnel experiments on flow within urban areas. Boundary-Layer
Meteorology 2003;107:353–70.
[32] Uehara K, Murakami S, Oikawa S, Wakamatsu S. Wind tunnel experiments on
how thermal stratification affects flow in and above urban street canyons.
Atmospheric Environment 2000;34:1553–62.
[33] Vardoulakis S, Fisher BEA, Pericleous K, Gonzalez-Flesca N. Modelling air
quality in street canyons: a review. Atmospheric Environment 2003;37:
155–82.
[34] Xie X, Liu C-H, Leung DYC. Impact of building facades and ground heating on
wind flow and pollutant transport in street canyons. Atmospheric Environment 2007;41:9030–49.
[35] Xie X, Huang Z, Wang J, Xie Z. The impact of solar radiation and street layout
on pollutant dispersion in street canyon. Building and Environment
2005;40:201–12.
[36] Yoshie R, Mochida A, Tominaga Y, Kataoka H, Yoshikawa M. Cross comparisons
of CFD prediction for wind environmentat pedestrian level around buildings.
comparison of results for flow-field around a high-rise building located in
surrounding city blocks, The sixth Asia-Pacific conference on wind engineering; 2005.