Nowadays, a lot of energy is spent for air-conditioning in the cities for offices and private-housing. A good knowledge of the urban climatology around the buildings allows using the wind for natural air ventilation. A combined approach based on a Computational fluid dynamics code (CFD) that models the wind in urban environment and on air network tool was developed to assess quickly the natural ventilation of buildings in urban area. The aerodynamic tool is a CFD-Network software named "UrbaWind which performs calculation of the outdoor wind combined with a macroscopic network method for the evaluation of the mass flow rates through the openings. Results depend on the external wind conditions, taking into account the local wind climatology. Giving the influence of the shape and the disposition of the buildings on the wind behaviors, CFD software solves the equations of fluid mechanics with a specific model which allows taking into account the urban environment effects such as vortexes, venturi or wise effects etc…. Finally, the software is able to compute the air mass flow rate inside each internal volume depending on the opening's behaviors attached to the wall building. The mass flow rates across air inlets and openings, as well as the pressure fields on the building envelope are computed for every wind incidence, considering a wind reference velocity. The local climatology is introduced to deliver statistical data useful to assess the energy consumption and the thermal comfort (air exchange rate, indoor velocity). Through many architectural and urban design projects, Meteodyn will introduce during this session the characteristics of the software UrbaWind and the results delivered to the design teams.

1. Assessment of the natural air ventilation
of buildings in urban area with the CFD tool UrbaWind
Dr Stéphane SANQUER
Meteodyn
www.meteodyn.com info@meteodyn.com

2. About the thermal indoor comfort
 The indoor temperature depends on the air change rate and the
thermal characteristics of the envelope (conduction, radiation, storage)
 The thermal comfort depends on the indoor temperature, air speed on
occupants (>1 m/s, T=>-4°C), Air humidity, activity, clothing

3. About the thermal indoor comfort
 A building is well designed according to the thermal comfort criteria if :
• The Indoor operative temperature is close to the mean outdoor temperature
=> Standard ASHRAE 55-2010 Criteria
• The Indoor operative temperature and the air humidity are into the Givoni
comfort polygone curves

4. About the thermal indoor comfort
Example in warm tropical climate : T(outdoor)=30°C ; T(indoor)<32°C
=> DT=T(indoor)-T(outdoor)<2°C
Questions : How to reduce the
overheating of indoor air?
Who ‘s responsible?
Insulation and ventilation as well
0
1
2
3
4
5
6
7
8
9
10
0 10 20 30 40 50
T(indoor)-T(outdoor)(C)
ACH (1/h)
Roof without insulation
Roof insulation
Roof and NE wall insulation

5.  Flows go from the highest pressure areas to the lowest
pressure areas
 Velocity depends on the root square of the pressure gradient
Cross ventilation principles
From CSTB guidelines book (Natural ventilation in tropical warm climate)

6. The pressure coefficient is a parameter without dimension that
depends on the complex interactions between the wind and the
building
Upstream face : Cp from 0.5 to 0.8
Downstream face: Cp from -0.5 to -0.3
Side faces: Cp from -1 to -0.3
2
2
1 ref
ref
P
U
PP
C



Pressure = engine of the cross ventilation
UrbaWind

7. Air change rate and indoor velocity fundamentally depend on the external
wind pressure at the openings.
Basic formula for crosswind ventilation (one volume , 2 openings)
• Aeq =Aerodynamic area of the openings
• Uwind = wind speed
• Cp1 et Cp2: pressure coefficients
We need UWIND and Cp to calculate the mass flow rate.
Tables (Liddament, Eurocode) and parametric models can be used for
standard cases, that means for simple, detached and isolated buildings.
WINDeqWIND UCpCpAU
CACA
CpCp
Q 21
2
2
2
2
2
1
2
1
21
11




How to assess the air change rate ?

8. How to assess the air change rate ?
In urban configurations, wind velocity
and pressure on buildings may not be
easily evaluated.
Tables and analytic models can not
be used.
Experimental approach (Wind tunnel)
Numerical approach (CFD)
Mass flow rate could be evaluated with
a network model. The inputs are :
-External pressure field
-Characteristics of openings (A,C)
-Indoor volume dimensions

9. Find a tool for modeling the flow over complex terrains, in urban area,
into buildings...Lots of applications
The effects created by the buildings make the modeling
of urban flows more difficult.
Some typical effects :
 Vortex at the base of the towers
 High wind speed near the edges of the
upwind face
 Wake effects behind a building
 Speed up in pedestrian ways under a building and between
buildings
Meteodyn developed UrbaWind, an automatic CFD software for
computing the wind between buildings…as well as possible
Challenge in Wind Enginering
UrbaWind

10. UrbaWind solves the averaged equations of mass and momentum
conservations (Navier-Stokes equations) for steady flow and the
incompressible fluids.
The CFD calculation computes the outside flow and the pressure
field for every wind direction
Velocity field Pressure field
The CFD Tool: Directional computations
UrbaWind

11. The Network calculation computes ACH based on CFD pressure field
The indoor pressure Pi is unknown and the flow rates through the openings are
solved by a Newton-Raphson iterative process.
where F’(Pi) is the first derivative of F(Pi) with respect to Pi , and w is an under-
relaxation coefficient.
In the case of a multi-volume configuration, the k openings’ aerodynamic area Ak is
replaced by an equivalent aerodynamic surface taken into account the door
aerodynamic surface Adoor
The CFD Tool: Air Change Rate
   n
i
n
i
n
i
n
i PFPFPP '1
/

UrbaWind

12. The Network part computes ACH based on CFD pressure field
UrbaWind provides wind roses, distribution and time series of the air
Change Rate
The CFD Tool: Air Change Rate
-4000
-2000
0
2000
4000
6000
8000
10000
0 50 100 150 200
Massflowrate(m3/h)
Time (h)
Mass flow rate across each opening
WIND
VENTILATION

13. Example n°1
Round robin test
CFD vs experiments (scaledown model)

14. Round robin test case
Experimental measurements in Wind tunnel
French Working Group for Natural Ventilation Standardization
 Detached simple house
Sub-urban wind
Pressure and mass flow rate
comparisons

15. Pressure correlation CFD/Experience
y = 1.0235x + 0.0035
R² = 0.9438
-0.8
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
-0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
DonnéesUW
Données expérimentales
CorrélationCpsur les centresdes quatre faces de lamaisonPressure coefficient correlation Exp vs UrbaWind
Cp experiments
CpUrbaWind

16. Mass flow rate correlation CFD/Experience
Cross configuration – 2 Windows
<ACH> (UW)=9 vol/h
<ACH> (Exp)=10 vol/h
UrbaWind

17. Mass flow rate correlation CFD/Experience
Full cross configuration – 4 Windows
<ACH> (UW)=18 vol/h
<ACH> (Exp)=21 vol/h
UrbaWind

18. Example n°2
Natural ventilation of a urban block
La Réunion Island

19. Renovation of urban districts
Natural ventilation potential
First step :Numerical simulations of the wind flow into the urban area
UrbaWind
Time of computations :
30’/wind direction
500000 cells

20. Second step: ACH computations
(time step : every hour of a year)
Renovation of urban districts
Natural ventilation potential
UrbaWind

21. Third step: ACH statistics
Parameters:
• Position of the openings
• Wall porosity
Area of the windows
• Internal porosity
Area of the doors
Area of the internal openings
CASES <ACH> ACH (P<0.05)
ACH MINI
North: 3 windows
South: 1 window
Doors: opened
65 Vol/h 9 Vol/h
North: 3 windows
South: 1 window
Doors: enlarged
90 Vol/h 13 Vol/h
North: 3 windows
South: 2 windows
(1.4 m² + 1 m²)
Doors: opened
105
Vol/h
15 Vol/h
North: 3 windows
South: 2 windows
(1.4 m² + 2 m²)
Doors: opened
140
Vol/h
19 Vol/h
North: 3 windows
South: 2 windows
(1.4 m² + 1 m²)
Doors: closed
(louver 0.5 m²
above the door)
80 Vol/h 10 Vol/h
North: 3 windows
South: 2 windows
(1.4 m² + 2 m²)
Doors: closed
(louver 0.5 m²
above the door)
115
Vol/h
15 Vol/h
Renovation of urban districts
Natural ventilation potential

22. Example n°3
Summer night cooling
for an industrial building
(wind and stack effects mixed)

23. 0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
10/7 12/7 14/7 16/7 18/7 20/7 22/7 24/7 26/7
Outdoor temperature
Indoor Temperature
Natural air ventilation (wind and stack effects)

24. Natural air ventilation (wind effects)
NE wind, NE view
NE wind, SO view NE wind, SW view
SW wind, NE view
Pressure coefficient
UrbaWind
UrbaWind
UrbaWind UrbaWind

25. Natural air ventilation (wind effects)
SUMMER FREE COOLING
Average ACH
9600 m3/h – ACH = 2.0
vol/h
7600 m3/h ACH = 1.6
vol/h
INDUSTRIAL
AREA
Input door 4830 m3/h 4300 m3/h
Output door 4290 m3/h 2910 m3/h
Average ACH
3860 m3/h ACH = 6.7
vol/h
2900 m3/h ACH = 5.1
vol/h
OFFICE Input door 2000 m3/h 1600 m3/h
Output door 1600 m3/h 1000 m3/h

26. Natural air ventilation (wind and stack effects)
Add thermal pressure gradient due to stack effects
0
5
10
15
20
25
30
2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000
Fréquence(%)
Mass flow rate (m3/h)
Wind Wind+stack
0
5
10
15
20
25
750 1250 1750 2250 2750 3250 3750 4250 4750 5250 5750 6250
Fréquence(%)
Mass flow rate(m3/h)
Wind Wind+Stack
Tall volume (ACH +15%) 2 storeys building (ACH +40%)

27. Example n°4
Natural ventilation of a secondary scholl
Kourou – French Guiana

28. Whole geometry of the secondary school
Optimisation of the cooling of the class rooms
• depends on the configuration (line, V or grid)
• depends on the windows positions and aerodynamic behaviors

29. Pressure potentiel depends on configuration configuration
DCP=0.8 DCP=0.9 DCP=1.1
(href=10 m)

30. Mean speed up factor Mean pressure coefficient
UrbaWind UrbaWind

31. Two louvers with sizes 1x2 m² and 2x2 m² (downstream)
ACH average = 71 Vol/h
18% of time with ACH< 30 Vol/h

32. Decreasing of the air change rate compared to the detached
configuration without the main bluidings
-18%
-22%
-6%
-9%
UrbaWind

33. • ACH Statistics for average models
Distribution, frequence, mean, RMS
Day and night data, monthly…
• Indoor air velocity statistics
• Time series for the dynamic thermic
models
Mass flow rate for each opening
-10000
-8000
-6000
-4000
-2000
0
2000
4000
6000
8000
10000
0 50 100 150 200
Débitvolumique(m3/h)
Temps (h)
Débit volumique : ouvrant côté vent dominant
Further data for thermic tools
Mass flow rate vs time
ACH distribution

34. Others examples

35. Architectural design competition
assistance
Validation of the project designed by the
architect according to the local
climatology
Wind pressure on buildings Choice of the openings Ventilation strategy
Assistance to define the ventilation strategy

36. Diagnostic of the preliminary
project design
Validation of the project
proposed by the Project
designer to the Project owner
Diagnostic of the building
ventilation scenario and mass flow
assessment
Visualization of the natural
ventilation potential

37. From the preliminary to the
detailed project design
Validation of the project
Interior architectures and openings
validation
Data extraction for thermal software
Visualization of the
natural ventilation potential and diagnostic

38. Design of “green” district
Planners’ project
Diagnostic of the ventilated
buildings => Proposal of a new
improved version
Natural ventilation potential of the buildings
Time of computations :
120’/wind direction
3 Millions cells

39. Keep in touch
stephane.sanquer@meteodyn.com
www.meteodyn.com
info@meteodyn.com