Assotech breeze high rise apartments at sector 88 gurgaon 8447730201
NATURAL VENTILATION IN HIGH RISE APARTMENTS IN HOT … · 2018-03-05 · ii Natural Ventilation in...
Transcript of NATURAL VENTILATION IN HIGH RISE APARTMENTS IN HOT … · 2018-03-05 · ii Natural Ventilation in...
NATURAL VENTILATION IN HIGH-RISE
APARTMENTS IN HOT-HUMID CLIMATES
Sara Omrani
Bachelor of Architecture
Submitted in fulfilment of the requirements for the degree of
Doctor of Philosophy
School of Design
Creative Industries Faculty
Queensland University of Technology
2018
Natural Ventilation in High-rise Apartments in Hot-Humid Climates i
Keywords
Balcony
Case study approach
CFD
Cross ventilation
Full-scale experiment
High-rise buildings
Hot-humid climate
Meteorological data
Natural ventilation
Prediction methods
Residential buildings
Single-sided ventilation
Thermal comfort
Wind-driven ventilation
ii Natural Ventilation in High-rise Apartments in Hot-Humid Climates
Abstract
Today, cooling energy demand in buildings represents more than 10% of the
total world energy usage. Phenomena such as global warming, population growth, and
an increase in income are the key reasons for this increasing energy demand. Global
warming increases the cooling energy demand, population growth introduces more
people with such demand, and an increase in income makes air-conditioners more
affordable, hence, a greater proportion of the population can afford to use them. As
these factors are experiencing an upward trend, so does the cooling energy demand.
Since the negative effects of high energy consumption have become clearer, there is
an urgent need for energy conservation.
Economic development and population growth have also resulted in urban
consolidation and rapid emergence of high-rise buildings. The majority of these high-
rise buildings are designed with the sole reliance on air-conditioners that make them
highly energy-intensive, even though there is a great potential for energy conservation
by application of passive strategies for space conditioning in such buildings. In hot-
humid climates, the cooling energy consumption can effectively be reduced by the use
of natural ventilation.
This research aims to improve natural ventilation design in high-rise buildings
in hot-humid climates. The study focuses on two areas that largely affect the design of
natural ventilation in buildings: natural ventilation prediction and evaluation, and the
effect of design related parameters on ventilation performance. Given that natural
ventilation performance evaluation plays a crucial role in design improvement, the first
part of this study is dedicated to facilitating the natural ventilation evaluation process
for architects. This is mainly achieved through the review, identification, and analysis
of the available evaluation methods with regards to high-rise building projects.
Accordingly, a process model for integration of these methods into the design process
is proposed. The proposed model is then explored by employing full-scale
experiments, Computational Fluid Dynamics (CFD), and empirical models for natural
ventilation investigations. Using full-scale experimental measurements and
meteorological data, the correlation between wind speed and air velocity at building
Natural Ventilation in High-rise Apartments in Hot-Humid Climates iii
openings and indoor spaces was explored. Revealing a linear relation, natural
ventilation performance of a design can be estimated using meteorological data.
Effective natural ventilation also relies on a successful design of the elements
that influence natural ventilation performance. The second part of this thesis, therefore,
is focused on the effect of different design related parameters on natural ventilation
performance. Firstly, the effect of ventilation mode (single-sided and cross ventilation)
on ventilation performance was analysed using the full-scale experimental data. Full
scale experiments and CFD analyses on the high-rise apartment proved that cross
ventilation was 70% to 400% more effective than single-sided ventilation in terms of
providing thermal comfort and indoor air velocity respectively. Then, a CFD model
was validated using the same dataset and was used for simulation of the effect of
various balcony properties on natural ventilation performance under different wind
directions. Among wind direction, balcony type, and balcony depth, natural ventilation
performance was found to be most sensitive to the change of wind direction, which
highlights the importance of building orientation. It was also found that the addition
of a balcony mostly improves ventilation performance of single-sided ventilation
while it worsens that of the cross ventilation. Finally, the findings of these studies
along with the information extracted from the literature were gathered, prioritised, and
presented as a natural ventilation design flowchart. This chart can guide building
designers in natural ventilation design of buildings.
Given the limited available information about natural ventilation design of high-
rise buildings in the literature, this study can assist architects in designing naturally
ventilated high-rise buildings.
iv Natural Ventilation in High-rise Apartments in Hot-Humid Climates
Table of Contents
Keywords .................................................................................................................................. i
Abstract .................................................................................................................................... ii
Table of Contents .................................................................................................................... iv
List of Figures ........................................................................................................................ vii
List of Tables ........................................................................................................................... xi
List of Abbreviations .............................................................................................................. xii
Statement of Original Authorship ......................................................................................... xiii
Acknowledgements ............................................................................................................... xiv
List of publications ................................................................................................................. xv
Chapter 1: Introduction ...................................................................................... 1
1.1 Background and research problem ................................................................................. 1
1.2 Research aim, objectives, and questions ........................................................................ 3
1.3 Organisation of the thesis (thesis outline) ...................................................................... 5
1.4 Research progress linking the research papers ............................................................... 6
Chapter 2: Literature Review ........................................................................... 13
2.1 Natural ventilation in buildings .................................................................................... 13 2.1.1 Natural ventilation mechanism .......................................................................... 13 2.1.2 Advantages and disadvantages .......................................................................... 15 2.1.3 Codes and standards........................................................................................... 16
2.2 Design related parameters ............................................................................................ 18 2.2.1 Natural ventilation modes .................................................................................. 18 2.2.2 Building height .................................................................................................. 20 2.2.3 Windows and openings ...................................................................................... 22 2.2.4 Balconies and wing walls .................................................................................. 25 2.2.5 Plan layout and internal obstacles ...................................................................... 26
2.3 Natural ventilation and thermal comfort ...................................................................... 27 2.3.1 Fanger’s PMV/PPD model ................................................................................ 28 2.3.2 Adaptive model .................................................................................................. 29 2.3.3 Extended PMV model ........................................................................................ 31 2.3.4 SET* index ........................................................................................................ 32
2.4 Literature review summary .......................................................................................... 33
Chapter 3: Methodology .................................................................................... 35
3.1 Research design ............................................................................................................ 35 3.1.1 Design performance prediction and evaluation ................................................. 35 3.1.2 Effect of design related parameters on natural ventilation ................................ 37
3.2 Employed methods and data sources ........................................................................... 38 3.2.1 Case study approach .......................................................................................... 40 3.2.2 Brisbane climatic conditions .............................................................................. 44
Natural Ventilation in High-rise Apartments in Hot-Humid Climates v
3.2.3 Reference weather stations .................................................................................46 3.2.4 Full-scale experiment .........................................................................................47
3.3 Summary .......................................................................................................................54
Chapter 4: Natural ventilation in multi-storey buildings: design process and
review of evaluation tools ........................................................................................ 56
4.1 Introduction ..................................................................................................................58
4.2 Natural ventilation design of multi-storey buildings challenges ..................................61
4.3 Evaluation methods for natural ventilation ...................................................................63 4.3.1 Analytical and empirical methods ......................................................................63 4.3.2 Computational simulation ..................................................................................64 4.3.3 Experimental methods ........................................................................................71
4.4 Discussion .....................................................................................................................73 4.4.1 Method Evaluation .............................................................................................73 4.4.2 A design process model for integration of natural ventilation analysis into
overall building design .......................................................................................78
4.5 Conclusion ....................................................................................................................82
4.6 Appendix ......................................................................................................................85
Chapter 5: Predicting environmental conditions at building site for natural
ventilation design: Correlation of meteorological data to air speed at building
openings 95
5.1 Introduction ..................................................................................................................97
5.2 Background ...................................................................................................................98
5.3 Methodology .................................................................................................................99 5.3.1 Case study...........................................................................................................99 5.3.2 Weather stations ...............................................................................................101
5.4 Results and discussion ................................................................................................102 5.4.1 Weather stations ...............................................................................................102 5.4.2 Wind speed at building’s openings ...................................................................104
5.5 Conclusion ..................................................................................................................106
5.6 Future work .................................................................................................................107
5.7 Epilogue ......................................................................................................................108
Chapter 6: Effect of natural ventilation mode on thermal comfort and
ventilation performance: Full-scale measurement .............................................. 109
6.1 Introduction ................................................................................................................112
6.2 Methodology ...............................................................................................................115 6.2.1 Climate Conditions ...........................................................................................116 6.2.2 Case study building ..........................................................................................117 6.2.3 Experimental setup and instrumentation ..........................................................119 6.2.4 Meteorological data ..........................................................................................121 6.2.5 Thermal comfort models ..................................................................................122
6.3 Results and discussion ................................................................................................124 6.3.1 Measurements summary ...................................................................................124 6.3.2 Thermal comfort ...............................................................................................128 6.3.3 Reference wind speed and resulting airflow.....................................................130 6.3.4 Air flow distribution .........................................................................................131
vi Natural Ventilation in High-rise Apartments in Hot-Humid Climates
6.3.5 Reference wind direction and internal air flow direction ................................ 133 6.3.6 Wind direction and internal air flow ................................................................ 135
6.4 Conclusion ................................................................................................................. 140
6.5 Limitations and future work ....................................................................................... 140
Chapter 7: Thermal comfort evaluation of natural ventilation mode: case
study of a high-rise residential building ............................................................... 143
7.1 Introduction ................................................................................................................ 145 7.1.1 Climate condition of Brisbane ......................................................................... 146
7.2 Methodology .............................................................................................................. 147 7.2.1 Full-scale measurements .................................................................................. 147 7.2.2 Evaluation criteria ............................................................................................ 149
7.3 Results and discussion ............................................................................................... 150 7.3.1 Cross ventilation .............................................................................................. 150 7.3.2 Single-sided ventilation ................................................................................... 151 7.3.3 Discussion ........................................................................................................ 152
7.4 Conclusion ................................................................................................................. 154
Chapter 8: On the effect of provision of balconies on natural ventilation and
thermal comfort in high-rise residential buildings.............................................. 155
8.1 Introduction ................................................................................................................ 157
8.2 Method of analysis ..................................................................................................... 159 8.2.1 Field measurement ........................................................................................... 159 8.2.2 Numerical method............................................................................................ 163 8.2.3 Tests configurations (case studies) .................................................................. 167 8.2.4 Thermal comfort model ................................................................................... 168
8.3 Results and discussion ............................................................................................... 169 8.3.1 Results summary .............................................................................................. 169 8.3.2 Sensitivity analyses .......................................................................................... 175 8.3.3 Thermal comfort analyses ................................................................................ 178
8.4 Conclusion ................................................................................................................. 179 8.4.1 Limitations and future work ............................................................................ 181
Chapter 9: Discussion ...................................................................................... 182
9.1 Methods of analysis ................................................................................................... 182
9.2 Design related parameteres ........................................................................................ 184
Chapter 10: Conclusion ..................................................................................... 189
10.1 Summary of key findings ........................................................................................... 190
10.2 Limitations and future work ....................................................................................... 193
Bibliography ........................................................................................................... 197
Natural Ventilation in High-rise Apartments in Hot-Humid Climates vii
List of Figures
Figure 1.1. Graphical display of thesis objectives and chapters’ hierarchy. ............. 11
Figure 2.1. Positive and negative pressure zones as a result of wind force. .............. 14
Figure 2.2. Buoyancy-driven ventilation: displacement ventilation (left) and
mixing ventilation (right). ............................................................................ 14
Figure 2.3. Combined wind and buoyancy forces when complementing each
other (left) and opposing one another (right) ............................................... 15
Figure 2.4. Single-sided ventilation. .......................................................................... 19
Figure 2.5. Cross ventilation. ..................................................................................... 19
Figure 2.6. Stack ventilation in a room with openings (left), and stack
ventilation with ventilation chimney (right). ............................................... 20
Figure 2.7. Schematic atmospheric boundary layer profile. ...................................... 21
Figure 2.8. Ventilation strategies in tall buildings: A) whole floor covered
(isolated), B) connected floors with central void, and C) segmentation
(based on a figure by Etheridge (Etheridge, 2011)). ................................... 22
Figure 2.9. Window types examined by Gao and Lee (2011b). ................................ 24
Figure 2.10. Window types examined by Grabe et al. (2014): a) double vertical
slide window, b) turn window, c) bottom-hung window, d) awning
window, e) horizontal pivot window, and f) vertical pivot window ............ 25
Figure 2.11. PPD as a function of PMV (Hazim B Awbi, 2003). ............................. 29
Figure 2.12. Acceptable operative temperature range for naturally ventilated
buildings (ASHRAE, 2013). ........................................................................ 30
Figure 2.13. Acceptable range of operative temperature as a function of air
speed (ASHRAE, 2013). .............................................................................. 32
Figure 3.1. Relation of design elements and evaluation methods in natural
ventilation design. ........................................................................................ 35
Figure 3.2. Relation of the design process model diagram with chapters of this
thesis. ........................................................................................................... 37
Figure 3.3. Diagram of the thesis methods. ............................................................... 39
Figure 3.4. Illustration of methods employed and their relation to the research
outcome. ....................................................................................................... 40
Figure 3.5. The case study building (right) and living area of the case study
unit (left). ..................................................................................................... 43
Figure 3.6. Case study’s site plan (left), and schematic east-west section (top-
right), and north-south section (bottom-right). ............................................ 44
Figure 3.7. Case study's plan (right) and its location within the building (left). ....... 44
Figure 3.8. Average temperature, wind speed, and relative humidity in
Brisbane (2010-2015). ................................................................................. 45
viii Natural Ventilation in High-rise Apartments in Hot-Humid Climates
Figure 3.9. Prevailing wind direction in Brisbane at 9 am (from (Australian
Government Bureau of Meteorology, 2016)). ............................................. 46
Figure 3.10. The weather station locations in relation to the case study building ..... 47
Figure 3.11. Sensors' placements for different test configurations ............................ 54
Figure 4.1. Schematic atmospheric boundary layer profile. ...................................... 61
Figure 4.2. Ventilation strategies in tall buildings: A) whole floor covered
(isolated), B) connected floors with central void, and C) segmentation
(based on a figure by Etheridge (Etheridge, 2011)). .................................... 62
Figure 4.3. Diagram of the coupled strategy (Carrilho da Graça et al., 2002). ........ 67
Figure 4.4. Coupling process between Building simulation and CFD (L. Wang
et al., 2007). .................................................................................................. 68
Figure 4.5. Natural ventilation design process model within the overall design
process .......................................................................................................... 82
Figure 5.1. Case study’s site plan. ........................................................................... 100
Figure 5.2. Case study’s plan (right) and photos of sensors (left). .......................... 101
Figure 5.3. The weather station locations in relation to the case study building. .... 102
Figure 5.4. Wind speed change (left) percentage of different wind directions
(right) at Brisbane, Brisbane Airport and Archerfield weather stations. ... 103
Figure 5.5. Regression lines between wind speeds recorded at Brisbane station
expressed according to Brisbane Airport and Archerfield stations wind
speed. .......................................................................................................... 103
Figure 5.6. Wind speed changes of Brisbane station, 2D and 3D for duration of
the data collection. ..................................................................................... 105
Figure 5.7.Variation of wind speed recorded at measurement points (2D and
3D) versus Brisbane station wind speed. ................................................... 105
Figure 6.1. Illustration of the employed methods and the relation to the
research outcome. ....................................................................................... 116
Figure 6.2. Average maximum daily temperature (A), wind speed (B), and
relative humidity (C) in Brisbane (2010-2015) (Australian
Government Bureau of Meteorology, 2016). ............................................. 117
Figure 6.3. Case study’s site plan (left), and schematic east-west section (top-
right), and north-south section (bottom-right). .......................................... 118
Figure 6.4. Plan layout of the case study ................................................................. 119
Figure 6.5. Openings’ configuration and measurement points for cross
ventilation (Test1-left), and single-sided ventilation (Test2- right). .......... 121
Figure 6.6. Local coordinate system (𝑵′) in relation to the true north.................... 124
Figure 6.7. Outdoor weather conditions: temperature (A), relative humidity
(B), and wind speed (C) ............................................................................. 126
Figure 6.8. Sensors P1 and CP-5 location ............................................................... 127
Natural Ventilation in High-rise Apartments in Hot-Humid Climates ix
Figure 6.9. Sun path on the measurement day relative to the case study
building and location.................................................................................. 127
Figure 6.10. Extended PMV and PPD (A), and SET* (B) values for single-
sided and cross ventilation ......................................................................... 129
Figure 6.11. Scatter plot of airspeed at P1 and reference wind speed (A), P1
and P2 (B), P2 and P3(C), and P3 and P4 (D) ........................................... 131
Figure 6.12. Mean wind speed ratio in single-sided and cross ventilation in
relation to the space length. ....................................................................... 133
Figure 6.13. Frequency of wind direction at reference weather station and
measurement points for Test-1 (left) and Test-2 (right). ........................... 134
Figure 6.14. Average wind speed ratio corresponding to the four main
directions along the case study for the cross ventilation test. .................... 137
Figure 6.15. Average wind speed ratio corresponding to the four main
directions along the case study for the single-sided ventilation test. ......... 138
Figure 6.16. Highest and lowest values of average wind ratio with regards to
the reference direction for Test-1 and Test-2 ............................................. 139
Figure 7.1: Brisbane’s mean monthly temperature and wind speed ........................ 147
Figure 7.2. Case study location within the building (left) and plan and
measurement point (right) .......................................................................... 148
Figure 7.3. Extended PMV and PPD results for the cross ventilation setting ......... 151
Figure 7.4. Extended PMV and PPD results for the single-sided ventilation
setting ......................................................................................................... 152
Figure 7.5. Extended PMV results for the single-sided ventilation setting ............. 153
Figure 8.1. Case study building (right) and case study surroundings (left). The
case study building and the case study unit are indicated with red
boundary. ................................................................................................... 160
Figure 8.2. Case study building plan layout (Omrani, Garcia-Hansen,
Drogemuller, & Capra, 2016b). .............................................................. 160
Figure 8.3. Case study plan and sampling location for cross ventilation (left)
and single-sided ventilation (right). ........................................................... 161
Figure 8.4. CFD domain size ................................................................................... 166
Figure 8.5. Comparison of measurement and simulation results for A) cross
ventilation, and B) single-side ventilation ................................................. 167
Figure 8.6. Balcony types, open balcony (left) and semi-enclosed balcony
(right) ......................................................................................................... 168
Figure 8.7. Results summary for both ventilation modes (A), cross ventilation
(B), and single-sided ventilation(C) ........................................................... 171
Figure 8.8. Indoor average velocity for single-sided ventilation subject to
various balcony type, depths and prevailing wind direction...................... 173
Figure 8.9. Indoor average velocity for cross ventilation subject to various
balcony type, depths and prevailing wind direction. ................................. 174
x Natural Ventilation in High-rise Apartments in Hot-Humid Climates
Figure 8.10. A) Velocity magnitude around the building for baseline cases of
single-sided ventilation (left) and cross ventilation (right), B) Velocity
magnitude plan at 1.2m (top) and section A-A (bottom) for cross
ventilation baseline case, and C) Velocity magnitude plan at 1.2m
(top) and section A-A (bottom) for single-sided ventilation baseline
case. ............................................................................................................ 176
Figure 8.11. Sensitivity percentage of average air speed to different variables
for single-sided ventilation ......................................................................... 177
Figure 8.12. Sensitivity analyses of average air speed to different variables for
cross ventilation configuration ................................................................... 178
Figure 8.13. Investigated parameters potential cooling effect ................................. 179
Figure 9.1. Natural ventilation design process model within the overall design
process ........................................................................................................ 183
Figure 9.2. Natural ventilation design flowchart. .................................................... 188
Natural Ventilation in High-rise Apartments in Hot-Humid Climates xi
List of Tables
Table 1.1: Research outcomes relative to research objectives and questions. ............ 6
Table 2.1. The minimum opening area requirements by Australian Standard
1668.4 (Australian Standard, 2012) ............................................................ 17
Table 2.2. Expectancy factor based on the location and weather (P Ole Fanger
& Toftum, 2002) ........................................................................................... 31
Table 3.1. Weather stations information .................................................................... 46
Table 3.2. Instruments' specifications ........................................................................ 49
Table 4.1. Summary of Methods’ Features ................................................................ 77
Table 4.2. Methods’ Advantages and Limitations ...................................................... 78
Table 4.3. Summary table........................................................................................... 85
Table 5.1. Weather stations information .................................................................. 101
Table 5.2. Linear regression equations of Brisbane station wind speed (VBr) on
wind speed for Brisbane Airport (VAi) and Archerfield (VAr) stations ....... 104
Table 6.1. Summary of the instrumentation ............................................................. 120
Table 6.2. Measurement summary ........................................................................... 125
Table 6.3. Average wind speed ratio corresponding to the four main directions
for Test-1. ................................................................................................... 136
Table 6.4. Average wind speed ratio corresponding to the four main directions
for Test-2. ................................................................................................... 137
Table 7.1: Weather condition and measured values summary for the cross
ventilation setting. ...................................................................................... 150
Table 7.2. Weather condition and measured values summary for the single-
sided ventilation setting ............................................................................. 151
Table 8.1. Sensors' specifications. ........................................................................... 162
Table 8.2. Configuration parameters. ...................................................................... 168
xii Natural Ventilation in High-rise Apartments in Hot-Humid Climates
List of Abbreviations
BES Building Energy Simulation
BT Balcony Type
CBD Central Business District
CFD Computational Fluid Dynamics
CV Cross Ventilation
IAQ Indoor Air Quality
NCC National Construction Code
OB Open Balcony
PMV Predicted Mean Vote
PPD Predicted Percentage of Dissatisfaction
RANS Reynold-Averaged Navier-Stokes
RNG Renormalisation Group
SB Semi-enclosed Balcony
SET Standard Effective Temperature
SSV Single-Sided Ventilation
WWR Window to Wall Ratio
Natural Ventilation in High-rise Apartments in Hot-Humid Climates xiii
Statement of Original Authorship
The work contained in this thesis has not been previously submitted to meet
requirements for an award at this or any other higher education institution. To the best
of my knowledge and belief, the thesis contains no material previously published or
written by another person except where due reference is made.
Signature: QUT Verified Signature
Date: ________22/02/2018________
xiv Natural Ventilation in High-rise Apartments in Hot-Humid Climates
Acknowledgements
First and foremost, I would like to thank my supervisory team whom made my
PhD journey much more enjoyable. My principal supervisor Dr Veronica Garcia-
Hansen, for her mentorship, valuable advice, and the invaluable teaching opportunities
that she provided me during my PhD. My associate supervisor Dr Bianca Capra, who
patiently guided me through the mechanical engineering and technical parts of the
research and this thesis. The experimental part of this study would not have been
possible without her great support in purchasing the equipment. My deepest gratitude
goes to my associate supervisor Professor Robin Drogemuller, whom was more like a
father to me rather than just a supervisor. My appreciation for everything that he has
been doing for me during the past few years is just beyond words.
I am honoured to dedicate this thesis to my beloved family. My lovely mother
Fariba, who has been giving me her never-ending love and support. My father Jafar,
who has trusted me with my decisions and has always had my back. My little sister
Ava, who had to play the daughter role for both of us while I was away studying for
my PhD. I would never be where I am now without their love, sacrifice, support, trust,
and patience.
I would also like to thank my best friend Freshteh Banakar, who demonstrated
the true meaning of friendship. Freshteh patiently listened to me complaining almost
every day and provided me the greatest mental support in the last few years.
Last, but by no means least, I would like to thank all the friends and colleagues
that were by my side during this journey.
Natural Ventilation in High-rise Apartments in Hot-Humid Climates xv
List of publications
Queensland University of Technology (QUT) allows the presentation of a thesis
for the degree of Doctor of Philosophy in the format of papers, where such papers have
been published, accepted or submitted during the period of candidature.
The outcomes of the current thesis is composed of five papers and each paper is
presented as a chapter. Two additional papers were also written during this PhD but
do not constitute chapters in this thesis.
Publications presented as part of this thesis
• Omrani, S., Garcia-Hansen, V., Capra, B., & Drogemuller, R. (2017).
Effect of natural ventilation mode on thermal comfort and ventilation
performance: Full-scale measurement. Energy and Buildings.
doi: https://doi.org/10.1016/j.enbuild.2017.09.061
• Omrani, S., Garcia-Hansen, V., Capra, B., & Drogemuller, R. (2017). On
the effect of provision of balconies on natural ventilation and thermal
comfort of residential buildings. Building and Environment.
doi: http://dx.doi.org/10.1016/j.buildenv.2017.07.016
• Omrani, S., Garcia-Hansen, V., Capra, B., & Drogemuller, R. (2017).
Natural ventilation in multi-storey buildings: Design process and review of
evaluation tools. Building and Environment.
doi: http://dx.doi.org/10.1016/j.buildenv.2017.02.012
• Omrani, S., Garcia-Hansen, V., Drogemuller, R., & Capra, B. (2016).
Predicting environmental conditions at building site for Natural ventilation
design: Correlation of meteorological data to air speed at building openings.
50th International Conference of the Architectural Science Association
2016, Adelaide, Australia.
https://eprints.qut.edu.au/103498/
xvi Natural Ventilation in High-rise Apartments in Hot-Humid Climates
• Omrani, S., Garcia-Hansen, V., Drogemuller, R., & Capra, B. (2016).
Thermal comfort evaluation of natural ventilation mode: case study of a
high-rise residential building. 50th International Conference of the
Architectural Science Association 2016, Adelaide, Australia.
https://eprints.qut.edu.au/103494/
Publications not included in the thesis
• Omrani, S., Capra, B., Garcia-Hansen, V., & Drogemuller, R. (2015).
Investigation of the effect of balconies on natural ventilation of dwellings in
high-rise residential buildings in subtropical climate. 49th International
Conference of the Architectural Science Association 2015, Melbourne,
Australia.
https://eprints.qut.edu.au/90026/
• Omrani, S., Drogemuller, R., Garcia-Hansen, V., & Capra, B. (2014).
Natural ventilation heuristics in high-rise residential buildings: evaluation
and prediction. 48th International Conference of the Architectural Science
Association (ANZAScA) 2014.
https://eprints.qut.edu.au/78438/
Chapter 1: Introduction 1
Chapter 1: Introduction
This chapter first provides background, significance, and problem for this
research. Then the research aim and objectives of the current thesis are presented.
Finally, the organisation of this document and research progress linking the research
papers are given.
1.1 BACKGROUND AND RESEARCH PROBLEM
The world’s energy consumption has increased by more than 50% over the last
few decades (1971-2013) (International Energy Agency, 2015; L. Wang & Hien, 2007)
due mainly to economic and population growth (Pachauri et al., 2014). This rapid
increase in non-renewable energy use is not only monetarily expensive, but also has
numerous negative environmental impacts (greenhouse gases (GHG) emissions,
climate change, global warming, etc). The utilisation of natural energy resources,
therefore, has gained more attention.
The building sector is one of the main energy consumers globally. In Australia,
being responsible for nearly 20% of the total energy usage, buildings are the third
largest energy consumers following the manufacturing and transport sectors (CIE
(Centre for International Economics), 2007). Indoor heating and cooling systems are
responsible for approximately half of the Australian buildings’ energy usage (Pears,
2007) while this number goes up to 65% in some other developed countries (Orme,
2001). Global warming, population growth, and an increase in income are parameters
that affect the widespread use of air-conditioners. Since these parameters are
experiencing an upward trend, the energy demand for space conditioning, particularly
space cooling, is expected to increase drastically. Buildings’ cooling energy demand
is predicted to increase up to 750% in 2050 (Mat Santamouris, 2016). Such high levels
of energy consumption, as well as the consequent negative environmental effects, have
made energy efficiency strategies a priority in building regulations in many countries
(Pérez-Lombard, Ortiz, & Pout, 2008; Roetzel, Tsangrassoulis, Dietrich, & Busching,
2010). Although buildings are energy intensive, there is a significant potential for
reduction of energy usage in buildings (W. Miller & Buys, 2012), particularly by
2 Chapter 1: Introduction
utilising passive cooling and heating in cooling-dominant, and heating-dominant
climates respectively.
Natural ventilation is one the most effective passive cooling solutions, especially
for cooling-dominant climates (Liping & Hien, 2007; Matheos Santamouris & Allard,
1998). Application of natural ventilation, however, is not limited to cooling-dominant
climates and is increasingly being adopted in different building types across various
climate zones (R. De Dear, 2010). Natural ventilation can benefit occupants of
buildings from different aspects including thermal comfort and a healthier indoor
environment. Furthermore, 30% to 40% less energy consumption is reported in
naturally ventilated buildings compared to mechanically ventilated buildings (Gratia
& De Herde, 2004b; Kolokotroni & Aronis, 1999; Oropeza-Perez & Østergaard, 2014;
Schulze & Eicker, 2013; Shameri, Alghoul, Sopian, Zain, & Elayeb, 2011). In spite of
the proved advantages of natural ventilation, it has largely been ignored in the design
of high-rise buildings, resulting in them being highly energy intensive (Cheung, Fuller,
& Luther, 2005; R. Kennedy, Buys, & Miller, 2015).
Economic development and population growth have resulted in urban
consolidation and an increase in the manifestation of high-rise buildings (Cheung et
al., 2005). The Australian Bureau of Statistics defines high-rise buildings as multi-
storey structures with four or more storeys ("Australian Bureau of Statistics,"). In
Australia, construction of high-rise buildings is experiencing significant growth. In a
20-year period (1981-2001), the number of residents of high-rise buildings has nearly
doubled ("Australian Bureau of Statistics," 2004). Today, the number of approvals for
high-rise construction is higher than ever with an approximate increase of 300% over
the last ten years (Kusher, 2016) from ("Australian Bureau of Statistics,"). Considering
this large volume of unit apartments, an adoption of passive strategies such as natural
ventilation offers great potential for energy conservation in such buildings, whereas,
sole reliance on air-conditioning may impose an excessive burden on both energy
suppliers and the environment.
Natural ventilation design of buildings, however, is not a straightforward task
due to turbulent flow and complex fluid-flow physics and interactions (Chen, 2004).
This becomes even more challenging when the subjects of natural ventilation design
are high-rise buildings in dense urban areas. In such a scenario, not only does building
height affect the magnitude of the airflow (Etheridge, 2011), but also the surrounding
Chapter 1: Introduction 3
environment influences the turbulent intensity and wind pattern, making it more
difficult to predict. This complexity, combined with the lack of sufficient information
related to natural ventilation of high-rise buildings, are some of the reasons for
inefficient natural ventilation of majority of high-rise buildings.
Various parameters affect natural ventilation performance, some of which are
beyond the control of designers (uncontrollable variables), whereas others can be
addressed through design (design related parameters). Although uncontrollable
variables (e.g. wind, weather conditions, and surrounding environment) cannot be
modified and controlled by designers, they need to be considered in design for natural
ventilation. Therefore, they affect the decisions on design related parameters. Design
related parameters such as building height, openings, internal layout, and balconies
play an important role in the determination of natural ventilation performance.
Accordingly, a detailed consideration of both uncontrolled variables and design related
parameters need to be undertaken for successful natural ventilation design.
There is a good body of knowledge around the integration of these parameters
into the natural ventilation design of buildings (discussed in detail in Chapter 2). The
majority of these studies, however, are based on low-rise buildings and simple
geometries. Most of the available building codes are also developed based on low-rise
buildings and no specific guidelines were found related to natural ventilation in high-
rise buildings. Accordingly, despite the great potential for energy conservation and
resulting benefits from the integration of natural ventilation into the ever-growing
number of high-rise buildings, there is a significant gap in the literature addressing this
matter.
In summary; the need for energy conservation, the energy saving potential of
natural ventilation within cooling-dominant climates, the ongoing emergence of high-
rise buildings, the challenges associated with effective integration of natural
ventilation into the building design, and the limited available studies and guidelines
with regards to natural ventilation design of high-rise buildings, highlight the necessity
for further investigation of natural ventilation design of high-rise buildings in hot-
humid climates.
1.2 RESEARCH AIM, OBJECTIVES, AND QUESTIONS
The aim of this research is to improve natural ventilation design of high-rise
residential buildings in hot-humid climates.
4 Chapter 1: Introduction
Design evaluation and knowledge on the effect of design related parameters on
natural ventilation performance are the keys for a successful ventilation design.
Evaluation of design in terms of natural ventilation performance is a significant stage
which results in improvement of the design before construction, where every change
would be very costly and sometimes impossible. Furthermore, knowledge on the effect
of various parameters on natural ventilation performance leads to more informed
design decisions, therefore, effective design improvements. Therefore, these two key
factors shape the design and structure of this thesis. The following objectives were
defined to achieve the aim of this study:
1. Facilitate the process of natural ventilation prediction and evaluation for
designers.
2. Investigate the effect of design related parameters on natural ventilation
performance and thermal comfort of high-rise dwellings in hot-humid
climates.
Each of these objectives is associated with a number of research questions as
presented below.
Objective 1- Facilitate the process of natural ventilation prediction and
evaluation for designers.
Corresponding questions:
• What are the available natural ventilation evaluation and prediction methods
for high-rise buildings?
• What are the strengths and weaknesses of these methods?
• How can these methods be integrated into the overall design process of high-
rise buildings?
• How can meteorological data -the main available source of data to
designers- be used in natural ventilation prediction of high-rise buildings?
Objective 2- Investigate the effect of design related parameters on natural
ventilation performance and thermal comfort of high-rise dwellings in hot-humid
climates.
Corresponding questions:
Chapter 1: Introduction 5
• How do different ventilation modes (cross ventilation and single-sided
ventilation) perform in a typical apartment in a high-rise building in a hot-
humid climate?
• What is the effect of ventilation mode on indoor thermal comfort?
• What is the effect of the façade design on ventilation performance of high-
rise buildings?
• How can different parameters be designed to deliver effective natural
ventilation and thermal comfort in high-rise buildings?
In this research among the design related parameters, only ventilation mode and
balconies were investigated. The reasons for that are: 1) among the design related
parameters, ventilation mode is the main determinant of ventilation rate, 2) balconies
are one of the most common and desired elements in hot-humid climates. In addition,
the survey of the literature (Chapter 2) shows a rich body of knowledge around some
other influential design related parameters such as openings. Additionally, to address
the last question of objective 2, results of this study were combined with the outcomes
of other studies that were explored in the literature review chapter, and a flowchart
model for natural ventilation design of high-rise buildings is proposed.
1.3 ORGANISATION OF THE THESIS (THESIS OUTLINE)
This thesis is presented in the format of published papers. The outcomes,
therefore, are peer reviewed journal and conference papers. The thesis consists of 10
chapters. The first chapter (Introduction) provides context to the research problem,
aims, and objectives. The second chapter is the literature review, in which the previous
related works are critically reviewed and the gaps in the knowledge are identified.
Chapter three, methodology, describes the design of the research, relation of the
published work to each other, and explains the methods and equipment used in the
current study. The forth chapter is a published journal paper that reviews the
commonly used methods in natural ventilation studies and it further proposes a model
that will be explored in the next chapters. Chapters five to eight are journal and
conference papers that were presented as outcomes of this thesis. The relation of these
papers with the research objectives and research questions are presented in the next
section. Following the publications chapter nine, Discussion, discusses the outcome of
this thesis and presents all the findings of different chapters as a whole. Discussion
6 Chapter 1: Introduction
chapter also provides a natural ventilation design flowchart that is developed based on
the outcome of different chapters along with the information found through survey of
the literature. Chapter 10, conclusion, provides key findings and conclusion of the
study, in addition to discussion of future work.
1.4 RESEARCH PROGRESS LINKING THE RESEARCH PAPERS
The outcome of the current thesis is presented in five publications. Table 1.1
presents an overview of the relation of these publications with the research objectives
and research questions. Discussion chapter and its relation to the research questions is
also included in this table, however, it is not published as a paper.
Table 1.1: Research outcomes relative to research objectives and questions.
Research objectives Questions Publications/Chapters
1. Facilitate the
process of natural
ventilation
prediction and
evaluation for
designers.
• What are the available
natural ventilation
evaluation and prediction
methods for high-rise
buildings?
• What are the strengths
and weaknesses of these
methods?
• How can these methods
be integrated into the
overall design process of
high-rise buildings?
Chapter 4:
Natural ventilation in
multi-storey buildings:
design process and review
of evaluation tools
(journal paper)
• How can meteorological
data -the main available
source of data to
designers - be used in
natural ventilation
prediction of high-rise
buildings?
Chapter 5:
Predicting environmental
conditions at the building
site for natural ventilation
design: Correlation of
meteorological data to air
speed at building openings
Chapter 1: Introduction 7
(conference paper)
Chapter 6:
Effect of natural
ventilation mode on
thermal comfort and
ventilation performance:
Full-scale measurement
(journal paper)
2. Investigate the
effect of design
related parameters
on natural
ventilation
performance and
thermal comfort
of high-rise
dwellings in hot-
humid climates.
• How do different
ventilation modes (cross
ventilation and single-
sided ventilation) perform
in a high-rise building?
Chapter 6:
Effect of natural
ventilation mode on
thermal comfort and
ventilation performance:
Full-scale measurement
(journal paper)
• What is the effect of
ventilation mode on
indoor thermal comfort?
Chapter 7:
Thermal comfort
evaluation of natural
ventilation mode: case
study of a high-rise
residential building
(conference paper)
• What is the effect of the
façade design on
ventilation performance
of high-rise buildings?
Chapter 8:
On the effect of provision
of balconies on natural
ventilation and thermal
comfort in high-rise
residential buildings
(journal paper)
• How can different
parameters be designed Chapter 9:
8 Chapter 1: Introduction
to deliver effective
natural ventilation and
thermal comfort in high-
rise buildings?
Discussion
Chapter 4 is a review paper about the available tools for natural ventilation
evaluation, in addition to a design process model proposed for integration of the
identified methods into the whole design process. This paper is published in the
“Building and Environment” journal.
This review paper not only identifies the available natural ventilation evaluation
methods, but also assesses their advantages and limitations with regards to high-rise
building projects. From this, the most appropriate arrangement for application of these
methods in a design process of high-rise buildings is presented. The outcome of this
study contributes to objective 1 as it provides directions for an effective integration of
natural ventilation prediction and evaluation tools into the design process. This paper
is important in structuring the thesis since it forms a basis for determination of the
methodology applied in this study.
Chapter 5 correlates wind speed and direction at weather stations to air velocity
at openings of a high-rise residential building. This peer-reviewed conference paper
was presented at the ASA2016 conference in Adelaide, Australia.
Using measured velocity at the case study building, as well as weather data from
the three closest weather stations to the case study, this paper investigates; 1) possible
relationships between data from different meteorological stations, and 2) the
correlation of air velocity at openings of the case study with the wind speed at these
stations. The analysis reveals that air speed at building openings can be predicted using
the available wind data. Accordingly, this paper serves the last question of the first
objective of the thesis. Using data collected only at the openings, however, left the
possible correlation between wind data and air velocity at internal spaces unresolved.
Additionally, only the cross ventilation configuration was investigated in this paper.
Therefore, there was a need for further investigation under single-sided ventilation
configurations. These points, therefore, were addressed in chapter 6.
Chapter 1: Introduction 9
Chapter 6 investigates the effect of natural ventilation mode (i.e. single-sided
and cross ventilation) on ventilation performance and indoor thermal conditions in
high-rise buildings comparatively and with regards to the reference weather data.
Ventilation mode, as one of the main determinants of ventilation performance, is the
main focus of the paper. This paper is published in the “Energy and Buildings” journal.
The paper presented as chapter 6 addresses a number of issues. Firstly, it
complements objective two by investigating the correlation of the reference wind
speed with indoor air velocity at different points under both single-sided and cross
ventilation configurations. Then it presents the mechanism of both ventilation modes
in terms of airflow distribution as a result of different wind speed and directions.
Thermal comfort conditions for single-sided and cross ventilation, also, have been
assessed in this paper. Chapter 7, however, explored this issue more comprehensively.
Since ventilation mode is one of the major design related parameters, this paper
partially fulfils the second objective of this study.
Chapter 7 investigates indoor thermal comfort of a high-rise residential unit for
single-sided and cross ventilation configuration. The results of this paper are partially
presented in the Chapter 6 paper and also contribute to objective two, the impact of
design related parameters on ventilation performance and thermal comfort. This peer-
reviewed conference paper was presented at the ASA2016 conference in Adelaide,
Australia.
Chapter 8 studies the effect of balconies on natural ventilation performance of
high-rise buildings. This paper is published in the “Building and Environment”
journal.
In this paper, various balcony features such as type and depth were investigated
for single-sided and cross ventilation modes under different wind directions. Balconies
are one of the main architectural features in hot-humid climates and are categorised as
design related parameters that affect natural ventilation performance. Outcomes of this
paper, therefore, contribute to the second objective of this thesis.
Chapter 9 is the discussion chapter. This chapter ties all outcomes of this thesis
together.
Firstly, the methodological model proposed in Chapter 4 is explored using a case
study approach. Secondly, the effect of the design related parameters examined in
10 Chapter 1: Introduction
Chapters 6, 7, and 8 along with information extracted form literature review chapter
(Chapter 2) are used in developing a design flowchart model for natural ventilation
design of high-rise buildings. This chapter was not published as a paper.
To summarise, the outcome of this thesis consists of five papers in which the
first one develops the ground for the methodology employed in this research, as well
as guidelines for a better integration of natural ventilation evaluation methods into the
general design process. The other four serve the study’s aim at different scales. Firstly,
at a broader picture, the correlation between wind data from weather stations with the
expected airspeed at building openings is explored, allowing designers to predict
ventilation performance of their designs using the main source of data available to
them. Secondly, at building scale, the main determinant of natural ventilation
(ventilation mode) is comprehensively studied providing a clearer vision for decisions
on determination of ventilation mode. Finally, at a more detailed scale, balconies and
their impact on natural ventilation are investigated to shed light on the effectiveness of
this commonly used feature in hot-humid climates in terms of natural ventilation.
Figure 1.1 represents the chapters’ hierarchy and corresponding objectives.
Chapter 1: Introduction 11
Figure 1.1. Graphical display of thesis objectives and chapters’ hierarchy.
Chapter 2: Literature Review 13
Chapter 2: Literature Review
The focus of this thesis is mainly on natural ventilation design and evaluation.
This chapter, therefore, surveys and discusses the existing literature related to these
areas. To provide background to the subject, Section 2.1 covers the natural ventilation
mechanism, the advantages and disadvantages associated with it, and the
corresponding building codes and standards. Section 2.2 reviews the parameters that
affect natural ventilation performance of buildings and can be addressed through
design comprising ventilation mode, building height, openings, balconies and wing
walls, and plan layout and internal obstructions. One of the main purposes of natural
ventilation is to provide thermally comfortable environments for the occupants.
Thermal comfort, therefore, was used as a criterion for natural ventilation performance
evaluation in this study. Accordingly, Section 2.3 explores the impact of natural
ventilation on indoor thermal conditions, as well as different thermal comfort models
and their suitability for the current study. Finally, a summary of the reviewed literature
is provided in section 2.4.
2.1 NATURAL VENTILATION IN BUILDINGS
2.1.1 Natural ventilation mechanism
Natural ventilation driving forces are dynamic pressure and static pressure
differences. Accordingly, higher pressure differences result in higher ventilation rate.
The dynamic pressure differential is a result of incident wind while static pressure
difference is due to the temperature gradient, which is also known as buoyancy or stack
effect. Natural ventilation can also be driven by a combination of both static and
dynamic pressure differences ("BS 5925: Code of practice for ventilation principles
and designing for natural ventilation," 1991).
Wind striking a building surface causes a pressure differential by creating
positive pressure on the windward side and negative pressure on the leeward side and
the side walls (Figure 2.1). Having openings at the external walls, therefore, directs
the external air to flow through the internal spaces from the zone with positive pressure
to the zone with negative pressure (P. F. Linden, 1999). Greater pressure difference
results in higher indoor airflow rate. Parameters such as building shape, wind speed,
14 Chapter 2: Literature Review
wind direction, and surrounding environments affect the pressure distribution on the
building façade (Hunt & Linden, 1999).
Figure 2.1. Positive and negative pressure zones as a result of wind force.
Temperature difference affects the air density and produces buoyancy forces that
drive the air from high-density regions (lower temperature) to low-density regions
(higher temperature). Buoyancy driven ventilation can be categorised into two main
groups: mixing ventilation and displacement ventilation (P. Linden, Lane-Serff, &
Smeed, 1990). Mixing ventilation is normally characterized with one opening acting
as both supply and exhaust where cool air enters the enclosure from the lower part of
the opening and the warm air escapes from the upper part. Displacement ventilation,
however, works with two openings located at different heights, in which cool air flows
in from the lower opening and warm air flows out from the upper opening normally
located near the ceiling (Cooper & Linden, 1996) (Figure 2.2).
Figure 2.2. Buoyancy-driven ventilation: displacement ventilation (left) and mixing ventilation
(right).
Natural ventilation can also be driven by a combination of wind force and stack
effect. These forces may counteract or complement each other based on the location
of openings and the incident wind direction (Hunt & Linden, 1999). Indoor and
outdoor temperature differences in a room with openings at different heights produce
buoyancy forces and the stack effect. As illustrated in Figure 2.3, depending on the
Chapter 2: Literature Review 15
incident wind direction, the pressure differential created by wind forces can either
fortify (Figure 2.3-left) or oppose (Figure 2.3-right) the buoyancy forces.
Figure 2.3. Combined wind and buoyancy forces when complementing each other (left) and opposing
one another (right)
Wind driven ventilation is much more effective than stack ventilation but the
benefit of the stack ventilation is that it can circulate the air through a space even when
there is no wind (Walker, 2008).
2.1.2 Advantages and disadvantages
The main advantages of natural ventilation are reducing energy consumption and
consequent pollutants, providing thermal comfort, improving indoor air quality, and
low initial and operating costs.
Natural ventilation is one of the major determinants of indoor thermal comfort
conditions, especially in cooling-dominant climates (Papakonstantinou, Kiranoudis, &
Markatos, 2000). An elevated air velocity can eliminate the excessive heat of the
human body and provide it with a thermally comfortable environment. Uncomfortable
indoor thermal conditions encourage the use of air-conditioners, and hence the
consequent energy consumption (ASHRAE Fundamentals Handbook, 2009).
About 30% of the energy use by the building sector is fed to space conditioning
(M. W. Liddament, 1996). Natural ventilation replaces the hot air inside a space with
cooler air from the outside through natural processes. Thus, natural ventilation can
result in a reduction of energy consumption and the resultant pollution emissions
(Matheos Santamouris & Allard, 1998).
Furthermore, natural ventilation has the potential of improving the indoor air
quality by replacing the aged air inside the space with fresh air from the outside
(Matheos Santamouris & Allard, 1998). Improvements in indoor air quality result in
16 Chapter 2: Literature Review
improvement in occupants’ health and performance (Fisk & Rosenfeld, 1997). In
contrast to naturally ventilated buildings, mechanically conditioned buildings have
often reported problems such as sick building syndrome (Mendell et al., 1996). In
addition, a study on the effect of using natural ventilation instead of mechanical
ventilation on airborne infection transmission in hospitals suggests natural ventilation
decreases the chance of airborne contagion by 6-28% (Escombe et al., 2007).
In terms of installation and maintenance costs, natural ventilation is also much
more cost-effective than mechanical ventilation, particularly for residential buildings
(Etheridge, 2011).
Despite the aforementioned advantages, there are some limitations associated
with the application of natural ventilation in buildings such as: limited control, noise,
and pollution from outside.
Unlike mechanical ventilation, natural ventilation is highly dependent on natural
forces such as wind speed and direction (Bailey, 2000). Thus, ventilation rate cannot
be easily adjusted by the occupants in naturally ventilated buildings. In extreme, hot
climates, therefore, overheating in some days will be inevitable (Etheridge, 2011). In
addition, the dependence of building ventilation performance on wind conditions
requires adequate consideration of building location and design in order to facilitate
natural ventilation which adds additional challenges to the building design (Walker,
2008).
Furthermore, open windows used for natural ventilation make the enclosure
prone to outside noise and pollution (Kwon & Park, 2013) especially in high-traffic
areas and regions close to pollution sources.
In spite of the limitations associated with the application of natural ventilation
in buildings, this passive cooling system still remains an attractive solution for space
cooling. This becomes even more feasible for hot-humid climates where cooling is
most needed. In addition, residential buildings with limited numbers of occupants
(compared to office and commercial buildings) and more flexibility in clothing choice
have great potential for successful application of natural ventilation.
2.1.3 Codes and standards
As previously mentioned, natural ventilation serves a number of purposes such
as improving IAQ and improving thermal comfort in warm environments. A number
Chapter 2: Literature Review 17
of national and international standards have established criteria for natural ventilation
design. These criteria are usually set to dictate the minimum standards for natural
ventilation and they mainly only focus on IAQ requirements. Criteria for satisfactory
natural ventilation design in different standards are expressed in a variety of forms
such as opening area, and ventilation rate. The required specifications related to
ventilation are presented below.
Australian Standard 1668.4 (2012) specifies the minimum opening area
according to the space size, the number of occupants, and the intended activity level.
Presented in Table 2.1, the minimum required opening area is defined as a percentage
of floor area. As can be seen this number varies between 5%-7.5% of the floor area.
The recommended numbers are to provide sufficient IAQ.
Table 2.1. The minimum opening area requirements by Australian Standard 1668.4 (Australian
Standard, 2012)
Use of space Average adjusted
metabolic rate
Watts/occupant
Net floor area per occupant, 𝑚2
<2 2 to5 5 to 15
Low activity Up to 160 7.5% 5% 5%
Medium activity 161-200 7.5% 5% 5%
In the Australian National Construction Code (NCC), the minimum required
opening area for each habitable room with and without ceiling fan is 7.5 and 10 percent
of the floor area respectively (Australian Building Codes, 2011). Whereas, this number
is specified to be 4 percent of ventilated space floor area in the International Building
Code (IBC, 2006).
While Australian standards are only concerned with the minimum size of the
openings, other standards such as British Standard (BS 5925) ("BS 5925: Code of
practice for ventilation principles and designing for natural ventilation," 1991) and
ASHRAE 62.1 (A. ASHRAE, 2010) define the ventilation requirements based on the
ventilation rate. The BS 5925 ventilation rate requirement is defined based on the CO2
and activity level, varying between 0.8 L/s to 14 L/s ("BS 5925: Code of practice for
ventilation principles and designing for natural ventilation," 1991). ASHRAE 62.1 (A.
ASHRAE, 2010) requires 2.5 liters of fresh air per second per person plus 0.3 liters of
fresh air per second per square meter of space.
The building codes and standards requirements are mainly to satisfy the
minimum requirements for sanitary and respiratory purposes and they are deemed
18 Chapter 2: Literature Review
rather general for an effective application of natural ventilation as a cooling system.
Determination of openings only as the percentage of floor area and neglecting other
influential parameters such as opening types is largely implicit (see section 2.2.3).
Since high-rise buildings are exposed to higher wind speed and turbulence, and the
obstructions are normally less for higher levels, these parameters need be considered
for natural ventilation design of such buildings. However, no guidelines were found
that consider these parameters for high-rise buildings. More importantly, different
requirements for the same subject were found in different resources. This leads to
confusion on the part of designers.
2.2 DESIGN RELATED PARAMETERS
Natural ventilation in buildings can be affected by a different range of
parameters, some of which are not controllable by the designers, such as outside
weather conditions and site density, while some can be addressed through appropriate
design. In the current study, the latter are termed “design related parameters” and will
be explained in this section.
2.2.1 Natural ventilation modes
Among the design related parameters that influence natural ventilation,
ventilation mode has the greatest effect (Fung & Lee, 2014). Natural ventilation mode
can be defined based on the aperture placements and the ventilation mechanism.
Natural ventilation, therefore, can be divided into three main categories:
• Single-Sided Ventilation,
• Cross-Flow Ventilation, and
• Stack ventilation
Single-sided ventilation occurs when air enters and leaves from one side of the
space through one or more openings located on the same side as illustrated in Figure
2.4 (H. B. Awbi, 1994; P. F. Linden, 1999). As the air in single-sided ventilation
supplies and exhausts from the same side of the enclosure, it may not circulate through
the whole space.
Chapter 2: Literature Review 19
Figure 2.4. Single-sided ventilation.
Single-sided ventilation can be driven by buoyancy forces, wind forces, or both.
Buoyancy forces are dominant when wind speed is low (e.g. up to 2 m/s) and there is
a temperature difference between inside and outside. At higher wind speeds, however,
wind forces take over and buoyancy forces become negligible (Allocca, Chen, &
Glicksman, 2003). In the case of wind-driven single-sided ventilation, total flow rate
is a result of the mean and pulsating components of wind in which pulsating flow is
dominant for small openings whereas mean flow is the major cause for airflow in large
openings (J. Zhou et al., 2017).
Cross-flow ventilation occurs in the case of two or more openings installed on
opposite walls (Figure 2.5) where air flows in from the opening at the windward side
(inlet) and escapes from the opening at the opposite side (outlet) (Ohba & Lun, 2010).
Cross ventilation is highly affected by wind velocity and the resultant pressure
distribution around openings. Cross-ventilation is proven to be much more effective
than single-sided ventilation (H. B. Awbi, 1994; Fung & Lee, 2014; Givoni, 1969; M.
W. Liddament, 1996; Visagavel & Srinivasan, 2009) due to the greater pressure
differential between the inlet and outlet.
Figure 2.5. Cross ventilation.
Stack ventilation takes place when there is a height difference between the inlet
and outlet (Figure 2.6-left), where hot air rises and escapes through the higher opening
20 Chapter 2: Literature Review
and will be replaced by the cool air entering from the lower opening (Szokolay, 2004).
The effectiveness of stack ventilation is proportional to the height difference between
the inlet and outlet. Stack ventilation, therefore, functions more effectively in high
floor to ceiling heights spaces and/or through the application of ventilation chimneys
(Figure 2.6-right).
Figure 2.6. Stack ventilation in a room with openings (left), and stack ventilation with ventilation
chimney (right).
As elaborated in Section 2.1.1, higher pressure differences result in higher
ventilation rates. Pressure difference produced by wind is far greater than pressure
differential resulting from buoyancy and temperature difference (Evola & Popov,
2006). Accordingly, wind-driven ventilation is much more effective than stack
ventilation. In addition, the floor area utilized for the chimney can be a disadvantage
of stack ventilation, especially in high-density residential apartments. The current
study, therefore, only focuses on wind-driven ventilation (i.e. single-sided and cross
ventilation).
2.2.2 Building height
The main driving forces of natural ventilation (wind and buoyancy) are the same
for low-rise and high-rise buildings. However, the main challenge associated with
natural ventilation design in high-rise buildings comes from the greater pressure
differences created by both wind and buoyancy as a result of the higher heights
(Etheridge, 2011). Wind speed and wind pressure both increase with building height
(Günel & Ilgin, 2014), resulting in a building experiencing a wider pressure range
across the facade. Figure 2.7 illustrates a schematic of the atmospheric boundary layer
Chapter 2: Literature Review 21
showing the correlation between wind speed and height. As is evident from this figure,
the wind pressure loading on a building varies significantly with height, with upper
levels experiencing higher wind pressure loads than lower levels. Accordingly, in
upper levels of high-rise buildings, the higher wind pressure introduces additional
challenges for a natural ventilation design in terms of the size and design of the
openings (Wood & Salib, 2013).
Figure 2.7. Schematic atmospheric boundary layer profile.
Buoyancy forces result when there is a temperature and height difference
between inlets and outlets (Wood & Salib, 2013). In the case of high floor to ceiling
spaces and chimney like structures, the space height is the main determinant of
buoyancy driven pressure differences. Etheridge (2011) divides natural ventilation
strategies of tall buildings into three categories (Figure 2.8). In type A (Figure 2.8- A),
where the whole floor area is covered and openings of each floor are not connected to
vertical voids, the pressure differential generated by buoyancy forces are not
problematic and would act similar to the buoyancy forces of low-rise buildings. In this
condition, wind is usually the main driving force of natural ventilation. In type B
(Figure 2.8- B), high-rise buildings with central voids and large internal openings, this
pressure differential becomes challenging. In such cases, the building would act as a
single-cell and the overall height of the buildings would determine the pressure
difference made by buoyancy forces. Accordingly, the units at the lower parts
experience a great pressure drop that may result in an unacceptable force requirement
for opening the windows. Segmentation (Figure 2.8- C) is proposed by Liu et al. (2012)
to overcome this excessive pressure differential resulting from buoyancy forces in
22 Chapter 2: Literature Review
buildings with central voids. In this method, each segment is separated from the other
segments and as such, is analogous with a low-rise building.
Figure 2.8. Ventilation strategies in tall buildings: A) whole floor covered (isolated), B) connected
floors with central void, and C) segmentation (based on a figure by Etheridge (Etheridge, 2011)).
2.2.3 Windows and openings
Among the design related parameters, the effect of the openings on natural
ventilation is perhaps one of the most studied areas. Openings and their impact on
natural ventilation have been investigated according to their configurations and types.
Configuration of openings can refer to their form, size, and location on the façade
(Lukkunaprasit, Ruangrassamee, & Thanasisathit, 2009). A study on the influence of
the openings’ configuration on ventilation rate shows that placing two openings
opposite or perpendicular to each other would enhance the ventilation performance
(CF Gao & Lee, 2011a). Another study (Hassan, Guirguis, Shaalan, & El-Shazly,
2007) concerned with the openings’ configuration in single-sided ventilation reported
that placing two openings far apart improves the ventilation performance compared to
the case with two adjacent openings which supports the previous recommendations by
Santamouris and Allard (1998). Yin et al. (W. Yin, Zhang, Yang, & Wang, 2010),
Tantasavasdi et al. (Tantasavasdi, Srebric, & Chen, 2001), and more recently
Derakhshan and Shaker (Derakhshan & Shaker, 2017) investigated the opening
configuration for cross ventilation. Yin et al. (2010) pointed out that relative openings
heights affect ventilation performance considerably. Their results indicated that the
same level inlet and outlet results in a better ventilation in most cases. Tantasavasdi et
Chapter 2: Literature Review 23
al. (2001) found that a larger inlet accompanied with a smaller outlet would improve
the ventilation rate, although, this finding is in contrast with Santamouris and Allard’s
recommendation in their design handbook (Matheos Santamouris & Allard, 1998)
where an equal or larger outlet is suggested. Derakhshan and Shaker (2017) concluded
that rectangular windows with smaller width-to-height ratio would improve natural
ventilation.
Opening types and their impact on indoor airflow and natural ventilation have
been investigated in a number of studies (CF Gao & Lee, 2011b; Heiselberg, Svidt, &
Nielsen, 2001; O'Sullivan & Kolokotroni, 2017; von Grabe, Svoboda, & Bäumler,
2014; H. Wang, Karava, & Chen, 2015).
Heiselberg et al. (2001), and Gao and Lee (2011b) evaluated the effect of
different opening types for single-sided and cross ventilation modes. Although both of
these studies are concerned with the effect of window type on indoor airflow,
Heiselberg et al. (2001) look at the problem more from the draught risk perspective,
while Gao and Lee’s (2011b) study focuses on airflow as a passive cooling component.
Heiselberg et al. (2001) conducted laboratory experiments on side-hung and bottom-
hung windows. They concluded that in winter, the bottom-hung window is the most
appropriate type for both single-sided and cross ventilation configurations. In summer,
however, bottom-hung windows would not supply enough air to a single-sided room.
Gao and Lee (2011b) investigated four types of windows naming side-hung, top-hung,
full end-slider, and half end-slider (Figure 2.9) using CFD. It was found that full end-
slider and side-hung windows performed better for cross ventilation, while side-hung
windows were the most appropriate type for single-sided ventilation.
24 Chapter 2: Literature Review
Figure 2.9. Window types examined by Gao and Lee (2011b).
While Heiselberg et al. (2001) and Gao and Lee (2011b) focused on wind-driven
ventilation, Grabe et al. (2014) examined the ventilation performance of six different
window types (Figure 2.10) for buoyancy driven ventilation. From their experimental
results, horizontal pivot windows presented the best ventilation performance whereas
tilt windows were proven to be the worst.
Lately, Wang et al. (2015) developed semi-empirical models for ventilation
prediction of side-hung, top-hung, and bottom-hung for single-sided ventilation mode.
They evaluated the performance of the aforementioned window types under various
wind directions as a part of their study. Their result demonstrated that side-hung
windows performed better for windward conditions while bottom-hung windows
showed a better overall performance.
Most recently, the effect of slot louvres on air change rate of single-sided
ventilation was investigated (O'Sullivan & Kolokotroni, 2017). An average increase
of 6.5% in air change rate was reported using louvres compared to the plain opening.
Chapter 2: Literature Review 25
Figure 2.10. Window types examined by Grabe et al. (2014): a) double vertical slide window, b) turn
window, c) bottom-hung window, d) awning window, e) horizontal pivot window, and f) vertical
pivot window
2.2.4 Balconies and wing walls
Another façade design feature that can affect natural ventilation performance of
buildings are balconies. Balconies are one of the main architectural features in
subtropical climates (Buys, Summerville, Bell, & Kennedy, 2008), being used as a
private outdoor space, while potentially providing benefits to indoor air flows.
There are some studies investigating the impact of the provision of balconies on
indoor airflow in low-rise buildings. Prianto and Depecker (2002) pointed out that
balconies have a significant influence on indoor air movement and they can result in
an increase in internal air velocity. Chand et al. (Chand, Bhargava, & Krishak, 1998)
conducted an experiment to investigate the effect of balcony provision on pressure
distribution on the building façade. They found wind pressure distribution alters on the
windward side but not significantly on the leeward side and provision of a balcony
resulted in wind pressure increases in most cases. While Chand et al’s study focused
on pressure distribution on the façade of a case model without openings, their
experimental data, later on, was used for CFD validation and evaluation of the effect
of balcony provision on indoor ventilation performance (Z. Ai, Mak, Niu, Li, & Zhou,
26 Chapter 2: Literature Review
2011) and thermal comfort (Z. Ai, Mak, Niu, & Li, 2011). They carried out simulations
for both single-sided and cross ventilation configurations utilising small and large
openings. The simulation results from these studies indicated that mass flow rate
increases and average velocity decreases in the case of single-sided ventilation and no
significant change was found in cross ventilation (Z. Ai, Mak, Niu, Li, et al., 2011).
Thermal comfort status, calculated using the extended Predicted Mean Vote (PMV),
was also reported with no change (Z. Ai, Mak, Niu, & Li, 2011).
While these studies have been concerned with the effect of balconies on natural
ventilation, they were all based on simple geometries, and the combined effect of
balcony features (i.e. balcony type and depth) with other determinant parameters such
as ventilation mode and incident wind direction has not been adequately investigated.
Wing walls are another architectural feature that affects indoor air flow and
natural ventilation by creating pressure differentials (Aynsley, 2007). Givoni (Givoni,
1962, 1968) investigated the effect of wing walls on natural ventilation in a wind
tunnel. A room with and without wing walls was tested under different wind speeds
and directions. The experimental results confirmed that the addition of wing walls to
single-sided ventilation would significantly improve the natural ventilation and indoor
air circulation. Building on Givoni’s experiment, Mak et al. (Mak, Niu, Lee, & Chan,
2007) used CFD to investigate the effect of wing walls on ventilation performance of
single-sided ventilation under different wind directions. Similarly to Givoni’s study, it
was found that wing walls improve the ventilation performance by improving indoor
average air velocity and air change rate. Among the tested incident wind directions,
application of wing walls at the 45˚ wind direction was reported as the best
performance.
2.2.5 Plan layout and internal obstacles
A building’s depth and plan layout also affect the effectiveness of natural
ventilation and needs to be considered in the building design.
A study (C. R. Chu, Chiu, & Wang, 2010) on partitioned buildings with cross
ventilation demonstrated that increases in internal porosity result in ventilation rate
increases. In addition, it was found that the ventilation rate of buildings without
partitions is always higher than that of buildings with partitions. Similar to Chu et al’s.
(2010) study, Chu and Chiang (2013) studied the influence of internal obstructions on
Chapter 2: Literature Review 27
ventilation rate and the external pressure of cross ventilation. Properties of internal
obstacles such as width, height, and the location were considered in their study. The
results show that size of obstacle does not have a significant effect on external pressure.
Furthermore, in the case of small openings (less than 3% of external wall) the effect
of internal obstacles on ventilation rate can be neglected due to the dominant resistance
effect of external walls. The effect of internal obstacles, however, should be considered
in buildings with larger opening areas. Subsequently, cross ventilation in long
buildings was assessed by Chiang and Chu (2014). It was pointed out that internal
fractions and the smaller pressure difference between the openings of long buildings
(aspect ratio L/H ≥ 2.5) result in ventilation rate reductions compared to short buildings
(L/H=1.25). Moreover, their results confirmed ventilation rate decreases as building
length increases. In addition to the studies with cross ventilated subjects, Gan (2000)
investigated the effective depth of a room with single-sided ventilation. Local mean
age of air, air flow pattern, and air temperature were used as criteria for defining the
effective room depth. The study concluded that the effective room depth is directly
affected by room internal heat gain and window size.
In conclusion, natural ventilation performance not only depends on outside
weather conditions, but also is affected by the parameters that can be addressed
through design. These parameters are not independent of each other, and combination
of them is what determines the overall ventilation performance.
2.3 NATURAL VENTILATION AND THERMAL COMFORT
One of the main purposes of natural ventilation is to provide building occupants
with a thermally comfortable environment. Thermal comfort, therefore, is an
appropriate criterion for assessment of natural ventilation effectiveness. ASHRAE
standard (ASHRAE, 2013) defines thermal comfort as “that condition of mind which
expresses satisfaction with the thermal environment and is assessed by subjective
evaluation”. Parameters that define thermal comfort can be divided into three main
categories of physical, physiological, and psychological factors. Physical parameters
that influence heat loss and heat gain are temperature, humidity, air speed, metabolic
rate and clothing insulation (Papakonstantinou et al., 2000). In addition to physical
parameters, physiological and psychological parameters also play an important role in
defining one’s thermal conditions (R. De Dear & Schiller Brager, 2001). Among
28 Chapter 2: Literature Review
physical parameters, air speed plays an important role in determining thermal comfort
conditions in hot-humid climates.
When outside temperature is lower than the inside, air movement lowers the
indoor temperature by replacing the warm air inside the space with the cooler air from
the outside. It also affects the radiant temperature by cooling the building’s structure
and removing the heat stored in the building’s mass. In addition, an elevated air speed
affects the human body’s thermal condition directly in two main ways (Givoni, 1969).
Firstly, air movement over the skin surface affects thermal sensation by accelerating
convective heat transfer (Marc Fountain, Arens, De Dear, Bauman, & Miura, 1994; M
Fountain & Arens, 1993). Secondly, it reduces the discomfort from skin wetness by
increasing the sweat evaporation rate (Givoni, 1969; Szokolay, 2004). Accordingly,
higher air speed extends the comfort zone and allows higher temperature tolerance.
2.3.1 Fanger’s PMV/PPD model
Much research has been conducted in the last few decades with the aim of
establishing thermal comfort models and indices that can predict the thermal
conditions. Fanger’s comfort model (Poul O Fanger, 1970) is perhaps one of the very
first developed prediction models. This model was developed based on the physiology
of the human’s body heat exchange with the environment supported by a series of
experiments on human subjects in a controlled environment of laboratory and climate
chamber. Air temperature, radiant temperature, humidity, air speed, clothing
insulation, and metabolic rate are incorporated in Fanger’s model and the result is an
index called PMV (Predicted Mean Vote). PMV is a seven-point physiological scale
ranging from -3 to 3 where each scale indicates a thermal sensation as below
(ASHRAE, 2013):
-3 -2 -1 0 1 2 3
Cold Cool Slightly
cool
Neutral Slightly
warm
Warm Hot
PMV predicts the average thermal sensation vote of a group of people in a given
environment. Using the experimental data, Fanger correlated a Predicted Percentage
of Dissatisfaction (PPD) index to the PMV. The PPD calculation using PMV is
presented in Eq 2.1 (Hazim B Awbi, 2003):
𝑃𝑃𝐷 = 100 − 95𝑒𝑥𝑝 − {0.03353(𝑃𝑀𝑉)4 + 0.2179(𝑃𝑀𝑉)2} Eq 2.1
Chapter 2: Literature Review 29
The relation between PMV and PPD is presented in Figure 2.11. Considering
PMV=0 as neutral thermal sensation, the chart forms a symmetrical curve around zero.
As can be seen, the minimum PPD is 5% corresponding to PMV=0, meaning even in
the neutral condition there are some dissatisfied individuals. ASHRAE standard
(ASHRAE, 2013) considers an environment within the comfort zone when the
percentage of dissatisfaction is less than 10% which is equivalent of 0.5>PMV>-0.5.
Figure 2.11. PPD as a function of PMV (Hazim B Awbi, 2003).
The PMV/PPD model has been the basis of numerous thermal comfort studies
since. It also has been adopted in a number of standards such as ASHRAE 55
(ASHRAE, 2013), and ISO 7730 (Standard, 1994) for thermal comfort assessment of
an environment. However, there are some deficiencies associated with the use of the
PMV model. Although acceptable for thermal comfort prediction of air-conditioned
buildings, the PMV model under predicts the thermal comfort condition in naturally
ventilated buildings (Brager & De Dear, 1998; Croome, Gan, & Awbi, 1993; R. De
Dear & Brager, 1998; Humphreys, 1978). This under prediction is due to the steady
state assumption of thermal comfort in the PMV model, as well as neglecting the
adaptation of humans to their environment.
2.3.2 Adaptive model
De Dear and Brager (1998) divide the human thermal adaptation into three main
categories: 1) behavioural, 2) physiological, and 3) psychological. Behavioural
adjustments include the conscious and unconscious actions a person takes to adjust to
their thermal environment when feeling uncomfortable such as opening windows and
removing a piece of clothing. Physiological adaptation is a human’s physiological
response change as a result of being exposed to thermal environmental parameters.
30 Chapter 2: Literature Review
Acclimatization is an example of physiological adaptation. Psychological adjustment
is the change in thermal perception based on the previous expectations and
experiences.
The traditional PMV model offers a very narrow range of temperature as an
acceptable condition while occupants of naturally ventilated buildings can tolerate a
wider range of temperature compared to occupants of centrally air-conditioned
buildings. Therefore, a thermal condition perceived as acceptable by the occupants of
free-running buildings may be considered as a very uncomfortable condition using the
PMV model. The Adaptive comfort model (R. De Dear & Brager, 1998), therefore,
was developed based on an extensive field study (RP, 1997) (R. J. De Dear, 1998) to
predict the thermal condition of naturally ventilated buildings. The adaptive model
complements the traditional PMV model by accounting for the adaptation of humans.
Contrary to the PMV model, rather than predicting thermal sensation votes, the
adaptive model is a regression equation that represents the acceptable indoor operative
temperature as a function of mean outdoor temperature for 80% and 90% acceptability
limits (Figure 2.12). Today, the adaptive comfort model is added to the ASHRAE 55
standard for evaluation of thermal conditions in naturally ventilated buildings
(ASHRAE, 2013).
Figure 2.12. Acceptable operative temperature range for naturally ventilated buildings (ASHRAE,
2013).
Chapter 2: Literature Review 31
Although considered in the model development process, there is no direct input
for the four environmental parameters, metabolic rate, and clothing insulation in the
adaptive comfort model. Therefore, it is not an appropriate model for the current study.
2.3.3 Extended PMV model
After it was proven unsuccessful for thermal prediction of free-running
buildings, Fanger and Toftum (2002) introduced an extension for the traditional PMV
model appropriate for non-air-conditioned buildings using the data from the RP-884
database (R. J. De Dear, 1998). The extended PMV adds two corrections to the
traditional PMV: the expectancy factor and reduced metabolic rate. The difference in
expectation of occupants of naturally ventilated buildings with that of the air-
conditioned buildings is addressed by the introduction of expectancy factor (e). The
expectancy factor varies between 0.5 and 1 and should be multiplied with the
traditional PMV. The expectancy factor value is defined based on the period of hot
weather and the dominance of building type in terms of cooling system (i.e. air-
conditioned or free-running). Table 2.2 presents low, moderate and high expectancy
factors based on the location and according period of warm weather.
Table 2.2. Expectancy factor based on the location and weather (P Ole Fanger & Toftum, 2002)
Expectation Classification of non-air-conditioned buildings Expectancy
factor, e
Location Warm periods
High In regions where air-conditioned
buildings are common
Occurring briefly
during the summer
season
0.9–1.0
Moderate In regions with some air-conditioned
buildings
Summer season 0.7–0.9
Low In regions with few air-conditioned
buildings
All seasons 0.5–0.7
The other parameter considered in the extended PMV model is the activity level.
People tend to reduce their activity level unconsciously when feeling warm (P Ole
Fanger & Toftum, 2002). This reduction is 6.7% by every scale unit increase in PMV
index above the neutral point. Therefore, for PMV values above zero, a new metabolic
rate needs to be obtained and considered in the recalculation of the traditional PMV.
Accordingly, PPD can be calculated based on the obtained extended PMV value.
32 Chapter 2: Literature Review
2.3.4 SET* index
Another index developed for the calculation of thermal sensation is Standard
Effective Temperature (SET*) (Gagge, Fobelets, & Berglund, 1986). ASHRAE 55
(ASHRAE, 2013) defines SET* as the temperature of an environment at 50% relative
humidity and average air speed of below 0.1 m/s, where air temperature and radiant
temperature are equal in which “the total heat loss from the skin of an imaginary
occupant with an activity level of 1.0 met and a clothing level of 0.6 clo is the same as
that from a person in the actual environment, with actual clothing and activity level”.
SET* accounts for the combined effect of temperature, humidity, air velocity,
metabolic rate, and clothing insulation on thermal comfort of the occupants (Gagge et
al., 1986) and is the ASHRAE-55’s (ASHRAE, 2013) recommended comfort model
for cases with the indoor air speed of greater than 0.2 m/s. Figure 2.13 represents the
acceptable range of operative temperature as a function of the elevated air speed for
0.5 and 1.0 clo clothing value for the situations with and without occupant control on
air speed. As can be seen from the Figure 2.13, increases in air speed extend the
boundaries of acceptable operative temperature by about 4˚C.
Figure 2.13. Acceptable range of operative temperature as a function of air speed (ASHRAE, 2013).
For studies that are concerned with the impact of different parameters on natural
ventilation, the comfort models that provide direct input for air velocity offer the most
promise. Since the focus of the current study is on design related parameters and their
consequent impact on natural ventilation performance, SET* and the extended PMV
are considered the most suitable for thermal comfort evaluation.
Chapter 2: Literature Review 33
2.4 LITERATURE REVIEW SUMMARY
The application of natural ventilation as a passive cooling system has numerous
advantages such as energy saving, improving indoor air quality and providing thermal
comfort. On the other hand, there are some limitations associated with it such as
limited control. Despite the limitations, it is a feasible solution especially for
residential buildings in cooling-dominant climates. However, there are very limited
regulations in the building codes and standards regarding the effective application of
natural ventilation in buildings. The available codes focus more on minimum
ventilation requirements with no specific requirements for high-rise residential
buildings.
In addition to the climatic driven forces such as wind, natural ventilation is
influenced by a range of design features including ventilation mode, building height,
the design of openings, balconies and projections, and internal obstacles. Natural
ventilation, therefore, can be improved by appropriate integration of these design
features. Among the design related parameters, opening design has been extensively
explored while ventilation mode and balconies can benefit from more in-depth
investigations. In addition, there is a lack of holistic models that connect different
design features in a single chart that can be used as a guideline for natural ventilation
design of buildings.
Given that one of the main purposes of natural ventilation is to provide cooling
effect for building occupants, thermal comfort is an appropriate criterion for
assessment of natural ventilation performance. Accordingly, a comfort model that
reasonably predicts thermal condition in naturally ventilated buildings and
incorporates air velocity as an input is needed for natural ventilation performance
evaluation. Among several developed comfort models, the extended PMV and SET*
indices are deemed suitable for this purpose.
34 Chapter 2: Literature Review
Chapter 3: Methodology 35
Chapter 3: Methodology
3.1 RESEARCH DESIGN
As explained in the introduction chapter and based on the gaps identified in the
literature review chapter, this thesis focuses on the effect that different design elements
have on natural ventilation, and the methods that can be used for evaluation of design
performance. These areas are the two key factors that were presented as objectives in
the first chapter. These two factors are complementary and necessary for a successful
natural ventilation design, meaning that a designer needs to know which design
elements assist in delivering efficient natural ventilation into an indoor environment,
and also needs to test the design outcome to evaluate if the combination of the elements
work as expected. Figure 3.1 represents the relation of these two factors. The process
of designing and testing continues until a satisfactory outcome is reached. The research
design of both parts presented in Figure 3.1 will be discussed in the following sections.
Figure 3.1. Relation of design elements and evaluation methods in natural ventilation design.
3.1.1 Design performance prediction and evaluation
To study the design evaluation and prediction section of Figure 3.1, the methods
used for natural ventilation studies were studied and analysed in Chapter 4. Their
advantages and limitations were discussed regarding a set of criteria crucial for the
design process. These methods, then, were gathered into a diagram that takes into
account the requirements of different design stages and advantages and limitations that
Design performance
prediction and
evaluation
Effect of different
parameters on ventilation performance
36 Chapter 3: Methodology
each method offers. This study and the proposed diagram are elaborated in Chapter 4.
The next part is to explore the proposed model to evaluate its appropriateness, hence,
some of the methods in the proposed model were applied in this study. The relation of
the natural ventilation design process model diagram to different chapters of this thesis
is presented in Figure 3.2. This diagram illustrates the application of the proposed
model in this study, however, the process behind development of the proposed design
process model is explained in detail in Chapter 4. Due to time and resource limitations,
only three main parts of the proposed model are explored in the current thesis. These
stages are: “Feasibility”, “Final design”, and “Construction”. These three stages were
chosen as they have different levels of complexity and accuracy. The “Feasibility”
stage is applied in Chapter 5, where an empirical model that predicts indoor air velocity
using meteorological data is proposed and explored. Chapter 6, and 7 examine the last
stage: “Construction”, where full-scale experimental measurements are carried out in
a case study building. The experimental data is then analysed and the natural
ventilation performance of the case study is evaluated. Additional modifications for
improvement of natural ventilation performance of the case study building are
discussed in Chapter 8, where the “Final design” stage is implemented. A number of
design related parameters are studied in that chapter using a coupled method of CFD
and full-scale experiment. The outcomes of these chapters assist in understanding the
appropriateness of the proposed design process model and provide designers with a
guideline for the application of the different methods at various design stages.
Chapter 3: Methodology 37
Figure 3.2. Relation of the design process model diagram with chapters of this thesis.
3.1.2 Effect of design related parameters on natural ventilation
The effect of different architectural elements and their impact on natural
ventilation is investigated in the current thesis at two stages. Firstly, by a detailed
review of the literature and extraction of the available knowledge about these elements.
Secondly, by studying some of the elements that are identified as gaps in the literature.
The first part is presented in Chapter 2, literature review. The second part is allocated
to investigation of a number of factors such as ventilation mode, balconies, and
building orientation. These factors are chosen since they were identified as gaps in
knowledge (refer to Chapter 2, pages 26, 32, and 33). Two major ventilation modes,
namely single-sided and cross ventilation are studied in Chapters 6 and 7, and
balconies and building orientation are explored in Chapter 8.
The outcomes of these studies in addition to the findings from survey of the
literature are captured in a flowchart that is presented in the Discussion chapter
38 Chapter 3: Methodology
(Chapter 9). This flowchart takes into account the effect of each design related
parameter and places them in a holistic way. This chart can be used as a guideline for
natural ventilation design of high-rise buildings.
3.2 EMPLOYED METHODS AND DATA SOURCES
To investigate the aforementioned objectives of the thesis, a series of methods
that are aligned with the research objectives and questions and serve the overall
research design and methodology need to be adopted. Firstly, a case study located in
the targeted context needs to be chosen. This case study should meet criteria that
accommodate the research objectives. It also needs to be representative of a broader
context rather than a single, specific case. Different sources of data relevant to the
research questions, then, need to be collected in the selected case study. Finally, the
collected data along with computer simulations should be analysed to answer the
research questions and to develop a logic that can be applied to similar but extended
situations.
The adopted research method in the current thesis is illustrated in Figure 3.3. As
can be seen, a case study in a specific context (i.e. hot-humid climate, and high-rise
residential) is selected. Data related to natural ventilation and thermal comfort, then,
are collected in the selected case study. The collected data in addition to computer
simulation results are analysed and research outcomes are concluded.
Chapter 3: Methodology 39
Figure 3.3. Diagram of the thesis methods.
Data collection and computer simulations, and their relation to the research
outcomes are illustrated in detail in Figure 3.4. As explained in Chapter 4, combined
CFD and full-scale experiments can yield accurate and detailed results. These
methods, therefore, were adopted in this study. Full-scale in-situ measurements
conducted in a high-rise building (the case study) in addition to the weather data were
used to investigate the correlation of wind speed and indoor air velocity by developing
an empirical model (Chapter 5). The same data were also used for investigation of the
effect of ventilation mode on indoor thermal comfort and ventilation performance
(Chapter 6 and 7). Finally, using the collected data, a CFD code was validated and
used in the simulation of different balcony characteristics and the impact of building
orientation on ventilation performance (Chapter 8).
40 Chapter 3: Methodology
Figure 3.4. Illustration of methods employed and their relation to the research outcome.
The following section of this chapter introduces the case study approach, case
study building selected for the full-scale experiments, the equipment used, test
configurations, climatic conditions of Brisbane, and the reference weather stations.
CFD simulation and validation process is presented in Chapter 8, therefore, it is not
repeated here.
3.2.1 Case study approach
This thesis uses a case study approach to investigate natural ventilation and
thermal comfort in high-rise residential buildings in hot-humid climates. Case study
research can be categorized into intrinsic, collective and instrumental case studies
(Stake, 2008). Intrinsic case study research improves understanding of a particular case
where the intention is to explore the specifications within that case rather than
generalise from it. A collective case study is suggested when exploration of the same
issue through different views from the different cases are intended (Stake, 2008).
Collective case study is also preferred when the main case study is built up from
smaller cases (Patton, 2002). Instrumental case studies assist in understanding an issue
outside the case by helping to explain an issue or phenomenon or by providing
additional vision into the problem (Stake, 2008). A case study strategy in architectural
research is defined as “an empirical inquiry that investigates a phenomenon or setting”
Chapter 3: Methodology 41
(Groat & Wang, 2013) which can mostly be considered as instrumental case study.
Case studies in architectural contexts can be identified by five main characteristics
(Groat & Wang, 2013):
1. Studying of one or several cases in their actual contexts.
2. The capability of explaining cause-effect association.
3. The need for development of a theory/theories at research design stage.
4. Relying on more than one source of evidence where data from these sources
converges.
5. The theory developed through case study research should have the power of
generalisation.
The case study approach involves analysis of a case in relation to its dynamics
within that context. Accordingly, the case study finds meaning within its context and
cannot be separated from it (Groat & Wang, 2013). Consequently, outcomes of case
study research in one context do not necessarily apply to another context (Ary, Jacobs,
Irvine, & Walker, 2013). In addition, the boundaries of the case study should be well-
defined, where these boundaries can be place, time, events, activities, and processes
(Crouch, 2012).
In case study research, focus and purpose of the research should be identified
prior to the selection of the case study as different case studies can result in different
outcomes (Crouch, 2012). Results of case study research can be generalized to other
cases with similar key attributes (Robert K. Yin, 2009).
Considering the above mentioned parameters, this study uses a case study in a
particular setting appropriate to the research question(s). The case study building used
in this study was carefully selected based on the following criteria:
1. Located in hot-humid climate.
2. Located in a high-rise residential building.
3. Have the possibility of operating in both single-sided and cross ventilation
modes.
4. To be vacant and accessible during the measurement period.
42 Chapter 3: Methodology
Where criteria one and two provide context for the case study and criteria three
and four are associated with the research design and research questions.
Outcomes of case study research can be generalised to other situations using
analytic generalisation, meaning developing a logic that can be applied to other
situations. This applies to both single-case and multiple-case studies regardless of size
of the case studies (Robert K Yin, 2012). In addition to the level of complexity
involved in obtaining access to the units that meet the study criteria, this lead this
research to investigate a single case study deeply from different aspects rather than
superficial analysis of multiple-cases.
Full-scale structure (case study building)
To serve the focus of the current study (i.e. high-rise residential buildings), a unit
in a 36-storey residential building located near the Brisbane Central Business District
(CBD), Australia, was chosen for the full-scale measurements. Figure 3.5 presents
photos from the case study building (right) and living area of the unit (left). The
building is adjacent to the Brisbane River (250m wide) on the southern side and a 25-
meter wide street from the northern side. The Brisbane CBD, dominated by high-rise
buildings, is on the western side. The height and density of adjacent buildings are
relatively low on the eastern area of the case study building. At the building’s southern
side next to the river there is a parkland, therefore, there is no major construction up
to 120m distance from the river, whereas at the northern side, there is a relatively high
building (approximately 35m high) across the street. The case study’s site plan and
schematic profiles of the surroundings is presented in Figure 3.6.
Chapter 3: Methodology 43
Figure 3.5. The case study building (right) and living area of the case study unit (left).
The building is oriented 35° west of north (equator) and the case study apartment
is located at the eastern end of the building. Situated on the fifth floor, the case study
unit is approximately 18m above the ground. The residence contains two bedrooms,
two bathrooms, living area and kitchen, and two balconies at the opposite sides of the
living area. The balconies are connected to the living area by two identical doors with
the openable area of 1.16m x 2.5m=2.9m2. Figure 3.7 shows the unit’s plan and its
location within the building.
44 Chapter 3: Methodology
Figure 3.6. Case study’s site plan (left), and schematic east-west section (top-right), and north-south
section (bottom-right).
Figure 3.7. Case study's plan (right) and its location within the building (left).
3.2.2 Brisbane climatic conditions
Brisbane is located at 27.4° S latitude and 153° E longitude. Brisbane’s climate
is subtropical with warm and humid summers and mild to cool winters. Monthly mean
maximum daily temperature lie between 20°C in July to 30°C in January, and mean
Chapter 3: Methodology 45
relative humidity ranges from 50% to 70%. The annual mean wind speed is 3.6 m/s
and it predominantly blows from south and south-west in the mornings and from east
and north-east in the afternoons (Australian Government Bureau of Meteorology,
2016). The graph below (Figure 3.8) shows mean monthly temperature, relative
humidity, and wind speed in Brisbane over a five-year period (2010-2015). Prevailing
wind direction at 9 am extracted from Australian Bureau of Meteorology is also
presented in Figure 3.9 (Australian Government Bureau of Meteorology, 2016).
Figure 3.8. Average temperature, wind speed, and relative humidity in Brisbane (2010-2015).
0
10
20
30
40
50
60
70
20
22
24
26
28
30
32
Jan
Feb
Mar
Ap
r
May Jun
Jul
Au
g
Sep
Oct
No
v
De
c
Win
d s
pe
ed
(km
/h)
Tem
pe
ratu
re (°C)
MonthMean temperature Mean wind speed Mean RH
Rel
ativ
e H
um
idit
y (%
)
46 Chapter 3: Methodology
Figure 3.9. Prevailing wind direction in Brisbane at 9 am (from (Australian Government Bureau of
Meteorology, 2016)).
3.2.3 Reference weather stations
Meteorological data from three different weather stations was obtained for this
study: Brisbane, Brisbane Airport and Archerfield stations, which are located
approximately 2 km, 9 km and 12 km from the case study building respectively (Figure
3.10). These stations were chosen as they were the closest stations to the case study
building. Table 3.1 presents the weather stations location, elevation, and distance to
the case study building.
Table 3.1. Weather stations information
Weather station Distance to case
study
Latitude Longitude Station
height
Brisbane Station ~2 km -27.4808 153.0389 8.13 m
Brisbane Airport ~9km -27.39 153.13 4.51 m
Archerfield ~12 km -27.5717 153.0078 12.5 m
Chapter 3: Methodology 47
The meteorological data was obtained from Australian Government Bureau of
Meteorology (Australian Government Bureau of Meteorology, 2016) and included 1-
minute data of wind speed (km/h), wind direction (˚), air temperature (˚C), and relative
humidity (%).
Figure 3.10. The weather station locations in relation to the case study building
The obtained data was used as reference data in the analyses for ventilation mode
investigations. In addition, along with the measured values in the case study the
meteorological data were used for investigation of the relation of air speed at the
openings with the measured values at weather stations.
3.2.4 Full-scale experiment
The full-scale experiments were conducted during 12 days in summer 2016 (13th
-25th January) using wind, temperature, and humidity sensors. The measurements were
intentionally carried out in summer to represent the worst case scenario in terms of
cooling demand. If natural ventilation can provide sufficient cooling in this period, the
chances that it would apply for the rest of the year is very high. This section explains
the equipment used and the test configurations.
48 Chapter 3: Methodology
Experimental equipment and data acquisition
Air velocity, temperature, and humidity were measured at different points using
six anemometers, six thermometers, and a hygrometer. Four out of the six
anemometers were ultrasonic anemometers (three 2D and one 3D) with the capability
of measuring air speed and direction to a high level of accuracy. The rest are
omnidirectional velocity transducers. Velocity transducers are highly accurate
especially for low air speeds, therefore, they were used in internal spaces where the air
velocity was expected to be lower than the balconies. All the sensors were installed at
1.2m height representing the height of a sitting adult human’s head.
The implemented sensors were logged locally using data loggers, three laptops,
and built-in loggers at different sampling rates. The 3D anemometer measured the air
velocity and direction at a sampling rate of 1Hz. The measured data was recorded using
WindView software provided by the sensor’s manufacturer (Gill instruments) installed
on a laptop. Similar to the 3D anemometer, two of the 2D anemometers logged the
data using a laptop and WindView software. The sampling interval for these 2D
anemometers was set to 4Hz. The last 2D anemometer had to be connected to a data
logger (Campbell Scientific) for recording the measured values. The sensor’s sampling
rate was 1Hz and it was averaged and recorded at 1 minute intervals. The velocity
transducers’ sampling intervals were set to 5Hz and their data was logged using a
National Instruments data logger and LabView (Manual, 1998) program. The
hygrometer and thermometers had built-in loggers and sampled temperature and
humidity at 1-minute intervals. The OneWireViewer (MAXIM, 2009) software was
used to download the temperature and relative humidity data. Instrumentations’
specifications are presented in Table 3.2.
Chapter 3: Methodology 49
Table 3.2. Instruments' specifications
Instrument
(manufacturer)
NO. Parameters Accuracy and
resolution
Picture of the instrument
3D anemometer
(Gill instruments)
1 U,V,W
vectors
Speed: <1.5%
RMS @12 m/s
Direction: 2°
@12m/s
2D anemometer
(Gill instruments)
3 Wind speed
and 2D
direction or U
and V vectors
Speed: 2% @12m/s
Direction: 3° @12
m/s
Velocity
Transducer (8475
series, TSI)
2 Air velocity 3% of reading from
20° to 26° C.
1% of selected full-
scale range (2.5
m/s)
Thermometers
(iButton, Maxim
integrated)
6 Temperature Resolution: 0.0625
50 Chapter 3: Methodology
Hygrometer
(iButton, Maxim
integrated)
1 Relative
humidity
Resolution: 0.04
Data logger
(National
Instruments)
1 N/A Resolution: 24-bit
Downloaded from National
Instruments web page (National
Instruments, 2017)
Data logger
(Campbell
Scientific-
CR200X)
1 N/A Resolution: 12-bit
Experiment configurations
Access to a vacant apartment in a residential tower was negotiated as the location
for the live measurements.
During a 12-day experiment, different configurations were tested and the
corresponding data was recorded. The test variables included different ventilation
modes and opening sizes. Each configuration was tested for a minimum of 12 hours
and maximum of 30 hours. The collected data served two main purposes, firstly, to
evaluate the ventilation performance, thermal comfort, and relation of reference wind
Chapter 3: Methodology 51
to indoor airflow, and secondly, to be used for CFD validation. Therefore, the duration
of the tests was considered sufficient for these purposes. In most of the tests bedroom
and bathroom doors were kept shut and only the air flow inside the living area and the
balconies was measured. In most of the settings, sensors were placed in positions that
could capture the mainstream of flow. In addition, some sensors were situated on the
balconies to record the airflow before entering the internal space. The case study unit
was unoccupied for the duration of the experiments and there were no fan-assisted or
mechanical ventilation systems operating. Figure 3.11 presents the sensor placements
in the case study for all the test configurations.
52 Chapter 3: Methodology
Chapter 3: Methodology 53
54 Chapter 3: Methodology
Figure 3.11. Sensors' placements for different test configurations
3.3 SUMMARY
This study investigates two main areas: 1-methods of prediction and evaluation
of natural ventilation, and 2-the effect of design related parameters on natural
ventilation performance. The specific context of this research is high-rise apartments
in hot-humid climates. According to the research objectives and requirements, a case
study approach was identified as an appropriate methodology for this thesis. A case
study situated in a targeted context was chosen and was used for experimental
measurements. Using the collected data, validated CFD simulations were also carried
out for investigation of some identified design related parameters.
This study only uses a single case study location, however, the data collection
and simulations are carried out so that they are applicable to a broader context rather
than being specific to the studied apartment. The extent of generalisations from the
results and quality assurance of the collected data are explained below.
As explained in the previous section, the experimental measurements were
conducted under both single-sided and cross ventilation configurations. The results
driven from analysis of the collected data, therefore, can be applicable to both single-
sided and cross ventilated buildings located within a context similar to the case study’s
context. Accordingly, the outcomes of this study are not specific to the studied unit
Chapter 3: Methodology 55
and similar results could be expected from units located in high-rise buildings
surrounded by similar obstructions.
In order to alter some influential design features and study their effect on natural
ventilation and thermal comfort of high-rise residential buildings, the experimental
measurements were coupled with CFD. This coupled method not only offers accurate
and reliable results, but also offers control over the features that cannot be modified
through experimental measurements (Chen, 2009). A variety of design options,
therefore, can be explored accurately.
To assure the quality of the collected data, highly accurate sensors were
employed in this study (specifications are provided in Table 3.2). Since the sensors
were factory calibrated and were used in this study for the first time, no additional
calibrations were deemed necessary. In addition, reference weather data of high
frequency (1-minute) was purchased from theAustralian Bureau of Meteorology
(Australian Government Bureau of Meteorology, 2016). In terms of reliability of
computational simulation results, the CFD simulations were tested and validated
against the experimental data, and results were not used unless acceptable agreement
between the experimental data and simulations was achieved. The validation of the
CFD process, as well as the simulation settings, are presented in detail in Chapter 8.
56 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Chapter 4: Natural ventilation in multi-
storey buildings: design process
and review of evaluation tools
Omrani, S., Garcia-Hansen, V., Capra, B., & Drogemuller, R. (2017). Natural
ventilation in multi-storey buildings: Design process and review of evaluation tools.
Building and Environment.
doi: http://dx.doi.org/10.1016/j.buildenv.2017.02.012
Statement of contribution of co-authors for thesis by published paper
The authors listed above have certified that:
1. they meet the criteria for authorship in that they have participated in the
conception, execution, or interpretation of (at least) that part of the
publication that lies within their field of expertise;
2. they take public responsibility for their part of the publication, while the
responsible author accepts overall responsibility for the publication;
3. there are no other authors of the publication;
4. potential conflicts of interest have been disclosed to (a) granting bodies, (b)
the editor or publisher of journals or other publications, and (c) the head of
the responsible academic unit; and
5. Consistent with any limitations set by publisher requirements, they agree to
the use of the publication in the student’s thesis, and its publication on the
QUT ePrints database.
The authors’ specific contributions are detailed in below:
Contributor Statement of contribution
Sara Omrani Conducted literature review and analysis,
produced the graphics, developed the
study, and wrote the manuscript.
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 57
Veronica Garcia-Hansen Assisted in developing the study, and
reviewed the manuscript.
Bianca Capra Assisted in developing the study, reviewed
and proof-read the manuscript.
Robin Drogemuller Assisted in developing the study, and
reviewed the manuscript.
Principal Supervisor Confirmation
I have sighted emails or other correspondence from all co-authors confirming their
certifying authorship.
__Veronica Garcia Hansen___ 28/04/2017_____
Name Signature Date
QUT Verified Signature
58 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Abstract
Energy demand in cooling-dominant climates can be reduced by implementation
of natural ventilation as a passive cooling strategy. Accordingly, suitable evaluation
and prediction tools are requirements for effectively introducing natural ventilation in
building design. In addition, the rapid emergence of multi-storey buildings can
accelerate energy consumption, especially in cases of inappropriate building design.
This study, therefore, proposes a process model for the better integration and
evaluation of natural ventilation design into the overall building design process for
multi-storey buildings. To achieve this, available methods of natural ventilation
evaluation were identified through a literature review and classified into three main
categories: analytical and empirical methods, computational simulations, and
experimental methods. Strengths and limitations of each method are then evaluated
with regards to accuracy of the results, cost, applicability to complex geometries,
resolution of results, and time cost. Finally, a process model was proposed based on
the methods’ advantages and limitations, as well as needs of each design stage and
recommends the most suitable integration of natural ventilation evaluation methods
into the overall design process.
Keywords: natural ventilation; multi-storey buildings; cooling-dominant climate;
prediction methods; design process
4.1 INTRODUCTION
The world’s energy use has increased by more than 50% over the last few
decades (1971-2013) (International Energy Agency, 2015; L. Wang & Hien, 2007)
due mainly to population and economic growth (Pachauri et al., 2014). This rapid
increase in non-renewable energy consumption is not only monetary expensive, but
also has several negative environmental impacts (greenhouse gasses (GHG) emission,
climate change, global warming, etc.). Buildings are one of the main energy
consumers, and their maintenance and operation are responsible for 20% to 40% of the
total energy use globally (Pérez-Lombard et al., 2008). The largest portion of energy
delivered to buildings is used by Heating, Ventilation, and Air-Conditioning (HVAC)
systems for space conditioning (Orme, 2001) and is expected to increase up to 64% in
2100 (Mat Santamouris, 2016). Such high levels of energy consumption, as well as
resultant GHG emissions, have made energy efficiency strategies a priority in building
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 59
regulations in many countries (Pérez-Lombard et al., 2008; Roetzel et al., 2010). One
such strategy, that has significant potential to reduce HVAC energy usage, is to
provide effective passive cooling and heating in cooling-dominant, and heating-
dominant climates respectively.
Economic development and population growth have resulted in the densification
of our urban settings through the increase in multi-storey towers (Cheung et al., 2005).
Passive strategies with regards to the local climatic conditions, however, have largely
been ignored in the design of these buildings, resulting in them being highly energy
intensive (Cheung et al., 2005; R. Kennedy et al., 2015). Adoption of effective passive
design strategies for space heating and cooling, rather than sole reliance on air-
conditioners, therefore offers significant potential for energy conservation in multi-
storey buildings.
Air exchange between outdoor and indoor environments without mechanical
assistance such as ventilation fans and cooling process is known as natural ventilation.
Natural ventilation can improve thermal comfort and provide a healthier indoor
environment by changing the used air inside a space with fresh air from outside (M.
W. Liddament, 1996). Effective provision of natural ventilation into the buildings can
save both energy and money compared to mechanical ventilation due to its low
maintenance cost and zero energy consumption (Aynsley, 2007). This is particularly
relevant to cooling-dominated climates where the air-conditioners are the main
determinants of the buildings’ energy usage (W. F. Miller & Nazari, 2013).
In building design, however, predicting the natural ventilation performance of
buildings can be challenging due to the complex physics involved and for optimal
results should be integrated into the early design stages. Appropriate methods,
therefore, should be utilised for evaluating a building’s ventilation performance during
the design process. Natural ventilation performance is measured primarily through
fluid dynamic parameters such as airflow pattern, average velocity, airflow rate,
pressure distribution, Mean Age of Air (MAA), volumetric flow rate and other
qualities that can be derived from these parameters. These flow features can also be
used to determine the broader characteristics of the internal environment of buildings
through parameters such as Indoor Air Quality (IAQ) and thermal comfort. There are
a number of methods available for prediction and evaluation of natural ventilation
performance, each having their own advantages and limitations. As such, the
60 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
appropriate method(s) need to be chosen based on the project resources, requirements
and design stages.
In summary, the ongoing emergence of multi-storey buildings, the need for
energy conservation, the energy saving potential of natural ventilation within cooling-
dominant climates, and the challenges associated with effective integration of natural
ventilation into the building design, highlight the necessity for application of
appropriate tools for evaluation and prediction of natural ventilation in design process
of multi-storey buildings in cooling-dominant regions.
The literature indicates that a number of methods have been used to evaluate
natural ventilation performance. However, available studies focus mainly on very
specific matters and, the link between suitability of these methods to different design
stages of multi-storey buildings is yet to be investigated. By considering the benefits
and limitations of each evaluation method, as well as the design resources and
requirements, a clear connection between the methods and different design stages can
facilitate the successful application of natural ventilation. Consequently, energy saving
associated with space conditioning can be reduced at no significant cost to occupant
thermal comfort.
This study proposes a model for the integration of natural ventilation analysis
tools into the overall design process of multi-storey buildings. First, a discussion on
the specific challenges associated with natural ventilation design of these buildings is
presented. Following this, the results of a detailed literature review identifying the
commonly employed methods for the analysis of natural ventilation is given. In an
effort to limit the number of the publications to only those relevant to this study in
order to provide an achievable data set, this review focuses only on literature involving
multi-storey buildings examining natural ventilation performance in cooling-dominant
climates. An investigation of the advantages and limitations of the methods currently
being used to quantify natural ventilation performance is then presented. A
comprehensive review of the methods’ advantages and limitations is not only one of
the main steps toward defining a meaningful relation between the methods and the
design process, but also contributes in a better choice of tools and will facilitate
possible adjustments. Lastly, this study proposes a natural ventilation design process
model based on the analysis methods’ advantages and limitations.
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 61
4.2 NATURAL VENTILATION DESIGN OF MULTI-STOREY
BUILDINGS CHALLENGES
The main driving forces of natural ventilation (wind and buoyancy) are the same
for low-rise and high-rise buildings. However, the main challenge associated with
natural ventilation design in multi-storey buildings comes from the greater pressure
differences created by both wind and buoyancy as a result of the higher heights
(Etheridge, 2011). Wind speed and wind pressure both increase with building height
(Günel & Ilgin, 2014), resulting in a building experiencing a wider pressure range
across the facade. Figure 4.1 illustrates a schematic of the atmospheric boundary layer
showing the correlation between wind speed and height. As is evident from this figure,
the wind pressure loading on a building varies significantly with height, with upper
levels experiencing higher wind pressure loads than lower levels. Accordingly, in
upper levels of high-rise buildings, the higher wind pressure introduces additional
challenges for a natural ventilation design in terms of the size and design of the
openings (Wood & Salib, 2013). Moreover, semi-open spaces such as balconies on
upper floors may be subjected to high velocity and draught (Irwin, Kilpatrick,
Robinson, & Frisque, 2008), becoming practically unusable for high velocity wind
instances. Alternatively, design strategies such as double skin façades have been
proposed to mitigate this problem (Gratia & De Herde, 2004a; P. Wong, Prasad, &
Behnia, 2008).
Figure 4.1. Schematic atmospheric boundary layer profile.
Buoyancy forces result when there is a temperature and height difference
between inlets and outlets (Wood & Salib, 2013). In the case of high floor to ceiling
spaces and chimney like structures, the space height is the main determinant of
62 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
buoyancy driven pressure differences. Etheridge (2011) divides natural ventilation
strategies of tall buildings into three categories (Figure 4.2). In type A (Figure 4.2- A),
where the whole floor area is covered and openings of each floor are not connected to
vertical voids, the pressure differential generated by buoyancy forces are not
problematic and would act similar to the buoyancy forces of low-rise buildings. In this
condition, wind is usually the main driving force of natural ventilation. In type B
(Figure 4.2- B), high-rise buildings with central voids and large internal openings, this
pressure differential becomes challenging. In such cases, the building would act as a
single-cell and the overall height of the buildings would determine the pressure
difference made by buoyancy forces. Accordingly, the units at the lower parts
experience a great pressure drop that may result in an unacceptable force requirement
for opening the windows. Segmentation (Figure 4.2- C) is proposed by Liu et al. (2012)
to overcome this excessive pressure differential resulting from buoyancy forces in
buildings with central voids. In this method, each segment is separated from the other
segments and as such, is analogous with a low-rise building.
Figure 4.2. Ventilation strategies in tall buildings: A) whole floor covered (isolated), B) connected
floors with central void, and C) segmentation (based on a figure by Etheridge (Etheridge, 2011)).
Whether the natural ventilation driving force is wind or buoyancy or both, there
are different tools for prediction and evaluation of their effect on ventilation
performance. These methods and their application to multi-storey buildings will be
explained in the following section.
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 63
4.3 EVALUATION METHODS FOR NATURAL VENTILATION
A number of methods for natural ventilation studies were identified in the
reviewed literature. These methods are divisible into three main categories: 1-
Analytical and empirical methods, 2- Computational simulation, and 3- Experimental
methods, with each category being further subdivided into a number of groups. The
most appropriate method for the evaluation of natural ventilation can be one, or a
combination of two or more, of these groups.
4.3.1 Analytical and empirical methods
Analytical and empirical methods work with fluid flow equations. They are very
similar in terms of capabilities; however, an analytical method is derived from
fundamental mathematical fluid dynamics and heat transfer theory while an empirical
method is developed from experimental measurements and observations. A number of
empirical models for prediction of single-sided natural ventilation that were developed
based on classical orifice equation (equation 4.1) can be found in the literature (Z. Ai,
Mak, & Cui, 2013; Caciolo, Cui, Stabat, & Marchio, 2013; De Gids & Phaff, 1982;
Warren, 1977).
𝑄 = 𝐶𝑑𝐴√2|∆𝑃|
𝜌 (4.1)
Where Q is ventilation rate (m.s-3),
Cd the dimensionless discharge coefficient,
A is the opening area (m2),
|∆P|=|P2-P1|the pressure difference (Pa) or (kg.m-1.s-2),
𝜌 is density (kg.m-3).
The main problem with analytical and empirical methods is the required amount
of assumptions, simplifications and approximations needed to produce a closed
equation, which may compromise the accuracy of the results. Furthermore, the
necessary simplifications required to create a solvable equation requires some higher
order fluid flow parameters to be overlooked, thus creating a further limitation of these
methods. Despite this, analytical and empirical correlations remain useful for a
designer in terms of providing an estimation on ventilation performance for simple
situations but are usually are not suitable for complex geometries.
64 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Ai and Mak (2014), conducted a study to investigate the applicability of existing
empirical models for determining ventilation rates in multi-storey buildings. They
compared results of existing empirical methods with data from on-site measurements
together with results from Computational Fluid Dynamic (CFD) simulations. Their
study indicated that empirical models developed for single-zone buildings are not valid
for multi-storey buildings due to their inability to calculate ventilation rate differences
in different zones of a building. Further development of analytical and empirical
methods for multi-storey buildings may therefore be desirable.
In terms of the accuracy of the results, validation results within the literature
indicate a 10 to 28 percent difference between analytical and empirical results with
experimental measurements (C.-R. Chu, Y. H. Chiu, Y.-T. Tsai, & S.-L. Wu, 2015;
Kotani, Narasaki, Sato, & Yamanaka, 1996; Larsen & Heiselberg, 2008). This
suggests that analytical and empirical methods, although simplified, are useful and
have been validated with experiments.
4.3.2 Computational simulation
A number of simulation approaches are identified within the literature including
1- CFD and 2- a combination of CFD with multi-zone models (network airflow model)
or Building Energy Simulation (BES) programs.
The majority of publications reviewed in this study have used CFD combined
with experimental measurements. This coupled method is used for three main
purposes: 1- to visualise the collected data from the experiments and to explore various
parameters, 2- to provide insight into the flow physics not easily achievable with
experiments and 3- to validate a CFD model in order to use it in further similar studies.
These coupled method studies are explained in this section rather than experimental
method section.
CFD
CFD solves the governing Navier-Stokes equations to directly solve for the fluid
dynamic properties governing airflow movement (Anderson, 1995). Although this
type of simulation is computationally expensive, it provides a detailed description of
airflow patterns in and around buildings. In particular, CFD can be used to provide
detailed information on the distribution of air velocity, temperature, pressure and
particle concentration within the area analysed, be that the internal or external
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 65
environment. The accuracy of CFD results, however, is dependent on the quality of
the grid used, the application of correct boundary conditions, and the appropriateness
of any assumptions applied to the model. CFD validity as a prediction tool has been
investigated in a number of studies showing a reliability of this method in different
cases (Jiang, Alexander, Jenkins, Arthur, & Chen, 2003; Jiang & Chen, 2003; Lo,
Banks, & Novoselac, 2012; Ramponi & Blocken, 2012; van Hooff, Blocken, Aanen,
& Bronsema, 2011; Z. J. Zhai, Zhang, Zhang, & Chen, 2007). A detailed review on
the application of CFD in wind-induced natural ventilation in buildings can be found
in (Jiru & Bitsuamlak, 2010).
CFD was used as the sole tool for natural ventilation investigation in a number
of studies (Hazim B Awbi, 1996; Chiang & Anh, 2012; Omrani, Capra, Garcia-
Hansen, & Drogemuller, 2015; Omrani, Drogemuller, Garcia-Hansen, & Capra, 2014;
Visagavel & Srinivasan, 2009; P. Wong et al., 2008) and is extensively being used for
airflow related analysis (Tsou, 2001). For example, Chiang and Anh (2012)
investigated natural ventilation in a courtyard of a multi-storey residential building in
the subtropical climate of Taiwan. Their results confirmed the suitability of this
passive design feature in providing an effective air circulation and natural ventilation
in the multi-storey buildings surrounding the courtyard in hot and humid climates. The
same method was employed in another study to explore natural ventilation heuristics
in subtropical climates (Omrani et al., 2014). From this study, it was concluded that
despite the effectiveness of rules of thumb to some extent, more sophisticated methods
should be used as a design develops. Wong et al. (2008) also used CFD to investigate
a new type of double-skin façade for high-rise office buildings in hot-humid climates.
CFD was used to analyse airflow effects and investigate the possibilities of applying
natural ventilation in a multi-storey office building. It was found that their proposed
double-skin façade can provide acceptable indoor thermal conditions. Provision of
balconies and their effect on natural ventilation as another façade design feature has
also been evaluated using CFD (Omrani et al., 2015). In this study, three case studies
were defined; without balcony, open and, semi-enclosed balconies, and compared in
terms of their effect on flow uniformity, average velocity and Air Change per Hour
(ACH). It was concluded that semi-enclosed balconies provide a more uniform airflow
pattern for internal spaces.
66 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Application of CFD in the reviewed publications indicates the capability of CFD
in simulating a diverse range of subject matters and design alternatives and operating
conditions such as wind speed and temperature changes. Also, a variety of flow
qualities and quantities can be obtained from CFD simulations. In fact, it is these
qualities that make CFD a powerful analysis and design tool to a designer.
CFD coupled with multi-zone and BES
A common approach for natural ventilation studies found in the literature is CFD
coupled with multi-zone (airflow network model) and BES. Multi-zone and BES are
similar in terms of simulating natural ventilation, hence, both coupling methods of
“CFD and multi-zone” and “CFD and BES” are explained here under one section.
Multi-zone can be classified as a macroscopic model and is based on mass,
chemical species, and energy conservation equations. It models natural ventilation by
assigning a node to each zone and flow paths between the zones. Conditions within
the zone, such as air velocity, temperature, and humidity, are then computed based on
the pressure difference between each defined zone and is commonly solved under
steady-state conditions (Johnson, Zhai, & Krarti, 2012). Use of multi-zone models
requires assumptions that the air within a zone is well-mixed with uniform
temperature, air velocity, contaminant concentration and relative humidity throughout
each zone. Multi-zone models are useful in prediction of ventilation performance in
an entire building as they provide bulk solutions, however, they cannot provide
detailed information about flow behaviour within each zone (Tan & Glicksman, 2005).
Furthermore, they cannot be used for external airflow simulations and can only be used
for indoor spaces. Several airflow network models have been developed and made
available for public use. COMIS (Helmut E Feustel & Smith, 1997) and CONTAM
(Walton & Dols, 2005) are the most popular multi-zone models for natural ventilation
studies (Johnson et al., 2012). Extensive background and theory of multi-zone models
can be found in (Axley, 2007).
BES models found in the literature involve the coupling of thermal models and
network airflow models (multi-zone models). Hence, BES analysis provides the same
results as multi-zone models for natural ventilation in addition to results on energy
flows in a building including lighting, heating, and cooling. Information on thermal
performance of a building, as well as its natural ventilation performance provided by
BES modelling can be further used for thermal comfort investigations. Johnson et al.
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 67
(Johnson et al., 2012) evaluated the performance of two network airflow models
(COMIS and CONTAM) and built-in airflow models within two of the most popular
building energy simulation models (EnergyPlus and ESP-r) in terms of natural
ventilation simulation. Their results confirmed that these programs performed the
same, as the underlying thermal physics are the same. Thus, it can be concluded that
the selection of program is less crucial than the application of correct modelling
assumptions.
Multi-zone and building energy simulation models often have been coupled with
other simulation tools such as CFD to increase the accuracy and resolution of results
for natural ventilation studies. Combining these tools, for example, can provide
detailed information for thermal comfort studies, the energy use required to maintain
comfort conditions, and provide a detailed assessment of local airflow patterns within
a space. A comprehensive exploration of BES programs and CFD coupling approaches
can be found in (Z. Zhai, Chen, Klems, & Haves, 2001).
Integration of CFD with multi-zone and BES was investigated in a number of
studies focusing on natural ventilation (Asfour & Gadi, 2007; Negrao, 1995; Schaelin,
Dorer, Maas, & Moser, 1993; Tan & Glicksman, 2005) and more specifically in multi-
storey buildings (Carrilho da Graça, Chen, Glicksman, & Norford, 2002; P.-C. Liu et
al., 2012; L. Wang & Hien, 2007; L. Wang, Hien, & Li, 2007; Yik & Lun, 2010).
Carrilho da Graça et al. (2002) investigated day and night cooling ventilation using
CFD coupled with building thermal analysis. CFD simulation was implemented to
predict airflow and the results were used for setting boundary conditions for the
building thermal analysis. Finally, both thermal analysis and CFD simulation results
were used to define the building’s thermal comfort using Fanger’s comfort model.
Figure 4.3 represents the method implemented by Carrilho da Graça et al (2002).
Figure 4.3. Diagram of the coupled strategy (Carrilho da Graça et al., 2002).
68 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Wang et al. (2007) conducted a parametric study to investigate the effect of
façade design on natural ventilation in high-rise residential buildings in Singapore. The
authors implemented a coupled method between BES and CFD and evaluated natural
ventilation based on thermal comfort criteria. Wang and Hien (2007) used the same
method to evaluate the impact of façade design and different ventilation strategies on
the indoor thermal comfort of naturally ventilated apartment units in Singapore. The
reason for using this coupled method is the inability of built-in network airflow models
in building energy simulations to provide detailed airflow velocity in interior spaces.
Hence, a data exchange interface was programmed to exchange the data between CFD
and building energy simulation software (ESP-r) to increase the accuracy of the result
and detail for the thermal comfort study.
Figure 4.4. Coupling process between Building simulation and CFD (L. Wang et al., 2007).
Coupling CFD and multi-zone models has also been used to explore the energy
saving potential of natural ventilation utilization into air-conditioned apartments (Yik
& Lun, 2010). Using CFD, pressure coefficients of the openings from the external
airflow impinging on the building were obtained which were then used to simulate
natural ventilation inside the selected units using a multi-zone model. Results from
this study highlighted that utilising natural ventilation to an air-conditioned building
can result in more than 20% energy saving.
Based on the reviewed publications, coupling CFD with BES and multi-zone
methods can be applied for exploration of a wide range of subjects related to natural
ventilation performance. Compared to only using CFD, this coupling approach can
improve the reliability of the simulation results, possibly decrease the computational
cost, and provide additional information on buildings’ thermal performance.
Furthermore, simulation of a buildings’ thermal performance in addition to airflow
simulation provides more realistic results in cases with high ceilings where buoyancy
operates (i.e. atriums, double skin facades, etc.).
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 69
CFD and Experimental Methods
CFD models often require approximation and simplifications of the flow
physics. Hence, when using CFD models, uncertainties are inevitable to some extent.
Validating CFD results against experimental data increases the accuracy and reliability
of the results through the minimisation of this approximations and simplifications.
Experimental methods, however, are also not without limitations, including cost (time
and money), equipment needed to increase spatial resolution and access to case study
buildings. A benefit CFD has over experimental methods is that once validated, CFD
analysis can provide detailed airflow information within the whole space compared to
point data available experimentally. Therefore, CFD and experimental methods
together can provide reliable and detailed information about ventilation performance.
The coupling of CFD and experiments has been used extensively for investigating the
effectiveness of various design related parameters including opening type (CF Gao &
Lee, 2011b), opening size and configuration (CF Gao & Lee, 2011a; Shetabivash,
2015), ventilation type (Evola & Popov, 2006; Jiang et al., 2003), building orientation
(C.-R. Chu, Y.-H. Chiu, Y.-T. Tsai, & S.-L. Wu, 2015; Hooff & Blocken, 2010; Horan
& Finn, 2008; Norton, Grant, Fallon, & Sun, 2009) and façade design (Aflaki,
Mahyuddin, Al-Cheikh Mahmoud, & Baharum, 2015; Ding, Hasemi, & Yamada,
2005).
Gao and Lee (2011a) investigated the impact of opening configuration on the
natural ventilation performance of multi-storey residential buildings in Hong Kong.
CFD was used as a simulation method while experimental data collected from tracer
gas decay was used to validate the CFD model. This method was also used by the same
authors (CF Gao & Lee, 2011b) to evaluate the effect of different window types on
natural ventilation performance of multi-storey buildings. Fung and Lee (2014)
implemented the same method to identify the most influential parameter amongst
window type, window to wall ratio, living room area, ventilation type, and orientation
on natural ventilation performance of high-rise residential buildings. Their results
show that ventilation type (Single-sided, cross ventilation, etc.) is the most influential
parameter affecting the MAA amongst the investigated factors. The last three studies
mentioned above used the same on-site measurement data collected from a site in
Hong Kong.
70 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
The influence of a ventilation shaft on natural ventilation enhancement and
thermal comfort improvement was evaluated by Prajongsan and Sharples (2012).
Using CFD, average velocity and temperature were numerically calculated and
compared to full-scale experimental measurements from a high-rise residential unit in
the hot-humid climate of Bangkok. Agreement of about 90% in directly measured and
simulated parameters was demonstrated before the numerical results were used to
examine the effect of the ventilation shaft on thermal comfort.
The surrounding environment can also affect the natural ventilation inside
buildings. Le et al. (Le, Li, & Su) conducted a study on wind environment
characteristics on a high-rise residential district in the subtropical climate of Changsha
city, China. They used measurement sensors to measure flow characteristics such as
air velocity, humidity, and temperature around the buildings. A CFD model was
developed based on this collected data and used for further studies to provide more
detailed information about flow characteristics around the buildings. Zhou et al. (C.
Zhou, Wang, Chen, Jiang, & Pei, 2014) proposed an optimized design strategy for the
high-rise residential buildings, where which then evaluated using CFD analysis
together with field measurements.
The reviewed literature demonstrates that CFD in combination with
experimental measurements is the most commonly used method for evaluating and
predicting the performance of natural ventilation. This combination is the most
common, and ideal, as it provides a validated numerical model from which further,
more detailed analysis can be derived to examine a number of design solutions
numerically with confidence in the accuracy of the results.
When using CFD for multi-storey buildings the following points need to be
considered. Firstly, an atmospheric boundary layer should be assigned to the inlet
boundary condition to account for the wind speed increase as a result of an increase in
height. This can be seen in most of the aforementioned studies that employed CFD (Z.
T. Ai & Mak, 2014; Carrilho da Graça et al., 2002; Chiang & Anh, 2012; CF Gao &
Lee, 2011a, 2011b; Omrani et al., 2015; Prajongsan & Sharples, 2012). Additionally,
the CFD computational domain is proportional to the building’s height, therefore, a
larger domain is required for simulation of high-rise buildings. The increase in domain
size results in greater computational time and computer resources requirements.
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 71
4.3.3 Experimental methods
In experimental methods, various measurement techniques can be used to
measure flow characteristics such as air velocity, temperature, pressure, and humidity.
These measurements can be conducted on both small and full-scale models. The
former uses a reduced scale model of the building while the latter is usually conducted
in the actual building or on a full-scale model of the building in a laboratory. A number
of techniques can be used in experimental methods such as tracer gas techniques
(Sherman, 1990) and air motion measurement (Sun & Zhang, 2007). A review of
experimental techniques for natural ventilation studies can be found in (Hitchin &
Wilson, 1967), and a more recent study investigating room air measurement in (Sun
& Zhang, 2007). Experimental methods found in the literature have mainly been used
to validate mathematical or computational methods (primarily CFD). The coupled
method of experiments and CFD is discussed in the Computational simulation section.
Full-scale experiments
Full-scale experiments can be conducted either in-situ or in laboratories (Chen,
2009). In the case of multi-storey buildings, however, laboratory full-scale
experiments are not possible due to the size of buildings. Accordingly, only in-situ
full-scale experiments with multi-storey subjects were found in the literature. Tracer
gas techniques are one of the most popular techniques for full-scale experiments
natural ventilation studies (Shao & Riffat, 1994; Shao, Sharples, & Ward, 1993).
Parameters, such as ventilation rate and Mean Age of Air (MAA), can be measured
using these techniques which can be used for both natural ventilation and Indoor Air
Quality (IAQ) studies. Ai et al. (2013) conducted on-site measurements implementing
the tracer gas decay method in residential units located in high-rise buildings in Hong
Kong to assess both ventilation performance and IAQ. The study concluded that in the
presence of adequate wind speed, single sided ventilation would provide enough ACH
to achieve acceptable IAQ. Air motion measurement techniques are also popular in
full-scale experimental studies. A study about the prediction of indoor air velocity
according to meteorological data is an example of a full-scale experiment using these
techniques (Omrani, Garcia-Hansen, Drogemuller, & Capra, 2016a). In this study,
ultrasonic anemometers were used for air velocity measurement at the openings of a
high-rise residential unit in Brisbane, Australia. It was found that there is a linear
72 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
relation between the wind speed recorded at weather station and air velocity at
building’s openings.
Small-scale experiments
Wind tunnel experiments are one of the most common methods of small-scale
experiments for natural ventilation studies (Lo et al., 2012) and are frequently used for
measurements of wind pressure on tall buildings’ structure (Dalgliesh, 1975). Wind
tunnels offer a good degree of control over experiments as well as repeatability and
replicability of the conducted tests. However, scale change can affect airflow and heat
transfer unless correct non-dimensional parameters are maintained between the scaled
models. Hence, in order to obtain realistic data from wind tunnel experiments, flow
characteristics should be modelled as in the full-scale building (Hitchin & Wilson,
1967).
Integration of a stack system into a high-rise residential building in order to
improve natural ventilation was investigated by Priyadarsini et al. (Priyadarsini,
Cheong, & Wong, 2004). A small-scale experiment using a wind tunnel was chosen
as the method for this study. Both passive and active stacks were investigated to
evaluate their effectiveness for natural ventilation enhancement in typical residential
buildings in Singapore. Although the study was concerned with high-rise buildings,
only a single floor (scale 1:5) was built and tested in the wind tunnel. Wind tunnel
experiment was also used to study courtyard buildings in Singapore. Four different
courtyard buildings were explored by Wong et al. (N. Wong, Feriadi, Tham, Sekhar,
& Cheong, 2000). Scaled model of buildings (1:200) were tested using a boundary
layer wind tunnel replicating urban areas wind profile. They compared the wind tunnel
results with full-scale measurements to identify the ventilation characteristics of the
courtyard buildings in tropical climates. Kotani et al. (1996) also conducted a small-
scale experimental measurement to investigate the stack ventilation in the courtyard of
high-rise buildings. Air temperature and velocity in the courtyard were measured using
a 1:100 scaled model of the building. They further developed a simple mathematical
model for ventilation prediction using the collected data. Their mathematical model
can predict air temperature and airflow rate of the courtyards.
As can be seen in the reviewed literature, the scale of the model used in small-
scale experiments might differ depending on the problem. For instance, studies
concerned with flow behaviour around high-rise buildings and those concerned with
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 73
bulk volumes such as courtyards may not need a large scale model (1:100- 1:200),
while indoor airflow of units of a high-rise building may require a sufficiently large
scaled model (1:5). This adds difficulty from a point of view of wind tunnel size and
the associated costs. Additionally, to account for the atmospheric boundary layer and
variation of wind magnitude at different heights, turbulent boundary layer wind tunnels
with floor roughness needs to be used.
A limited number of studies were found using experimental methods as the only
method for natural ventilation investigations. They were mainly coupled with
computational methods and were explained in Computational simulation section.
The reviewed literature shows analytical and empirical methods are the least
common methods of natural ventilation studies of multi-storey buildings in cooling-
dominant climates. Geometry size and complexity may be the reason, as analytical and
empirical models are more suitable for simple geometries. CFD coupled with
experiments, however, was the most frequently applied method due to the reasons
discussed in section “CFD and Experimental Methods”.
4.4 DISCUSSION
This section discusses two main areas based on the previously reviewed
literature. First, an evaluation of the advantages and limitations of the aforementioned
methods based on five criteria: results’ accuracy, cost, applicability for complex
geometries, resolution, and variety of results and the required time, is presented. Given
the results of this, a design process model proposed for natural ventilation design of
multi-storey buildings is then provided. This proposed model is based on different
design stages requirements and the analysis methods advantages and limitations.
4.4.1 Method Evaluation
Accuracy of results
Results’ accuracy can be referred to the results’ representation of reality. In terms
of natural ventilation, accurate results must be representative of flow behaviour, such
as velocity and temperature distribution to name two. Full-scale experiments generate
data closest to reality (Chen et al., 2010) under appropriate conditions of implementing
measurement devices with specifications aligned with the aim of the experiments.
Nonetheless, this does not mean there is no error associated with them. Errors can be
minimized to some extent, but not totally eliminated (Melikov, Popiolek, Silva, Care,
74 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
& Sefker, 2007). In small-scale experiments, one of the main difficulties in obtaining
accurate results is the flow similarity requirements which need to be maintained
between the reduced-scale model and full-scale building. While some parameters can
be easily controlled in wind tunnel experiments (e.g. model scale and material,
temperature, wind speed and wind direction), some are more difficult, requiring
appropriate design, both physical and experimental, to satisfy the flow similarity, such
as the Reynolds Number. A non-dimensional number, the Reynolds Number indicates
the importance of viscous effect (for example if a flow is laminar or turbulent) (Katz,
2010) to a given situation and is an important parameter that must be maintained to
ensure flow similitude between scale. Reproducing Reynolds number similitude in
wind-tunnels, however, is often complicated especially in studies concerned with high-
rise buildings, where the whole building needs to be studied, due to limitation of wind
tunnel size, smaller scale models of multi-storey buildings need to be used. This will
add difficulties in satisfying flow similarity. As a consequence, the results accuracy
would be affected to some extent (Hitchin & Wilson, 1967). Additionally,
measurement equipment can affect the airflow pattern; hence, compromising the
results accuracy (Jiang et al., 2003). Some studies have reported up to 20% higher
readings of wind speed in small-scale experiments compared to the full-scale
experiments (Kawamura, Kimoto, Fukushima, & Taniike, 1988; N. Wong et al., 2000).
CFD simulation has some uncertainties in reproducing complex turbulent flows.
Therefore, CFD models need to be verified and validated against experimental data,
or validated test cases, in order to provide accurate results (Blocken & Gualtieri, 2012).
In addition, errors associated with CFD can be reduced by an appropriate mesh size
(grid independence solution) and valid assumptions on boundary conditions. Apart
from the limited application of mathematical methods, the simplification and
approximations they use for representation of complex flow behaviour in natural
ventilation studies can make them the least accurate method compared to the other
methods (Caifeng Gao, 2011). Furthermore, they need to be adjusted for each case in
order to provide reliable results.
Cost
Experimental methods have the highest cost compared to the other available
methods. This includes both small-scale and full-scale experiments. Monetary cost can
vary depending on the measurement equipment, type and number, utilised. Sun and
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 75
Zhang (2007) provide a cost estimation for air motion measurement techniques. The
cost associated with full-scale experimental studies differs little for high-rise and low-
rise buildings, yet it is shown to be highly dependent on the quantity and quality of the
equipment employed. For small-scale experiments, however, larger scaled models
may be required for high-rise buildings which can affect the wind-tunnel test and
material cost. Experimental methods involve material and labour costs, while
computational simulation methods have computational costs. The former is
experiencing an upward trend while the latter is decreasing due to advances in
computers (Allocca et al., 2003). Mathematical methods, on the other hand, are neither
labour- nor computationally -intensive, which suggests they are the least expensive
method.
Complex geometries
Multi-zone models may be the most suitable method for large scale geometries
as they simulate the ventilation in each zone with simplified assumptions such as
uniform temperature. Hence, the computational time would decrease significantly
compared to CFD (Tan & Glicksman, 2005). Although CFD has the capability to
simulate large and complex geometries, the large amount of required computing time
can be an issue. Full-scale experiments in large buildings may introduce more
uncertainties and uncontrollable variables. Boundary conditions are very difficult to
control in natural ventilation studies using full-scale experimental methods unless an
environment chamber is used (Chen et al., 2010). However, it is not possible to use
environmental chambers for studies on large-size multi-storey buildings due to the
building size and construction costs. In fact, outside weather conditions would define
the boundary conditions in such cases. Small-scale experiments, on the other hand,
provide a high level of control for conducting various tests on complex geometries,
especially for wind load studies on high-rise buildings (Y. Zhou, Kijewski, & Kareem,
2003). For investigation of detailed indoor airflow, however, the required model size
can introduce additional challenges. Mathematical methods may be the least suitable
method for large scale and complex geometries due to the amount of approximation
they use and their limited applicability. To conclude, all the reviewed methods except
for analytical and empirical models can handle complex geometries to a good extent.
Given this, other requirements such as level of accuracy, time, cost, and data resolution
76 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
have a greater influence on their applicability to the analysis of natural ventilation
design.
Result resolution and variety
The amount and variation of detailed information that can be extracted from
CFD are far greater than other methods. CFD can provide information about various
flow characteristics at any point of a computational domain. Analytical and empirical
methods, on the other end, may provide the least detailed information on ventilation
performance. They are suitable for generating quick insight about natural ventilation
in general, but they are incapable of providing detailed information. Likewise, multi-
zone methods simulate the ventilation in the whole building and cannot provide
detailed information such as airflow pattern inside each zone. The type and variation
of information that experimental methods can provide depends on the measurement
equipment being used in the experiment. Higher resolution of information may need
more sophisticated devices and/or increase in the number of devices. This can have a
direct effect on the cost and the amount of required time for the experiment.
Time
Mathematical methods are most suitable for very early design stages when a
parametric study of various design configurations may be needed to evaluate a large
number of design options rapidly. Li and Delsante (2001) suggest using analytical
methods before numerical methods such as CFD due to their ability for providing
quick estimations ventilation performance. The amount of time required for the other
examined methods depends on the building scale, type, and amount of information
needed. With this in mind, comparing the computational methods, the required time
for multi-zone simulations is less than the required time for CFD simulations for the
same building. In CFD simulations, the computational domain is proportional to the
building’s height. Thus, for high-rise buildings, the domain size can be much larger
than a computational domain of low-rise buildings. This adds a significant
computational and time cost to the study. Furthermore, in the same case scenario,
experimental methods may be the most time-consuming methods, due to set up time,
and physical time required to log data that can span a week or more.
A summary, and quick reference guide, of the advantages and limitations
discussed above for each analysis methodology, is given in
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 77
Table 4.1 and
Table 4.2.
Table 4.1. Summary of Methods’ Features
Analytical
and
empirical
Network
airflow models
(multi-zone)
CFD Small-scale
experiments
Full-scale
experiments
GENERAL FEATURES
• Low
accuracy
• No cost
• Not suitable
for complex
geometries
• Bulk results
with limited
detail
• Not much
time
required
• Moderate
accuracy
• Low cost
(only
software
package cost)
• Suitable for
complex and
big scale
geometries
• Bulk results
with limited
details
• Moderate
time required
• High accuracy
in case of
applying
appropriate
settings
• Low cost (only
software
package cost)
• Suitable for
complex
geometries
• Time-
consuming
simulations
• High accuracy
• Costly
• Suitable for
complex
geometries
• The amount of
detail is
highly
dependent on
the number of
measurement
equipment
(direct
relation to
cost)
• Time
consuming
• The most
accurate method
• Costly
• The amount of
detail is highly
dependent on
the number of
measurement
equipment
(direct relation
to cost and more
limited
compared to
small-scale
experiments)
• Time
consuming
HIGH-RISE SPECIFIC
• Very
limited
application
• Greater
computationa
l time is
required
• Larger
computational
domain is
required
• Atmospheric
boundary layer
should be
considered in
determination
of boundary
conditions
• Greater
computational
time and
• Boundary
layer wind
tunnel with
floor
roughness
should be
used
• Larger size
scaled model
may be
required
which would
affect the cost
• Full-scale
laboratory
experiments are
not possible for
high-rise
buildings
•
78 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
resources are
required
Table 4.2. Methods’ Advantages and Limitations
Accuracy Cost Complex
geometries
Results
detail
time
Analytical
and empirical ×
✓ × ×
✓
Network
airflow
models
✓ ✓ ✓ ×
✓
CFD ✓ ✓ ✓ ✓
×
Small-scale
experiments ✓ ×
✓ × ×
Full-scale
experiments ✓
×
(after
construction)
× ×
It is important to note that the suitability of these methods and their application
is highly dependent on the case study and can vary case by case. Their suitability
presented here is based primarily on type of the building that is the focus of this study
(multi-storey buildings).
4.4.2 A design process model for integration of natural ventilation analysis into
overall building design
Building design evolves stage by stage, and each design stage has its own
particular needs and resource allocations. Besides, there are different methods for
natural ventilation design of buildings with certain advantages and limitations. This
section proposes a natural ventilation design process model, shown in Figure 4.5,
based on the analysis methods and their advantages and limitations discussed in the
previous section. It identifies five staged, four design and an one after construction,
together with the most suitable evaluation methods proposed for each phase. The
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 79
design phases are 1-feasibility, 2-concept, 3-detail, 4-final stage, 5-construction and
after construction respectively.
Feasibility
Empirical and analytical methods are suitable methods for preliminary natural
ventilation feasibility evaluation as they are quick, costless and easy to implement.
They can provide reasonable estimates about the ventilation performance of the
proposed design. Analytical and empirical models are suggested to be used for
parametric studies before using the fully numerical methods (such as CFD) (Li &
Delsante, 2001). Bulk quantities such as ventilation rate, volume flow rate and the
whole space temperature can be obtained using analytical and empirical methods.
However, as the results obtained from these methods are limited in their accuracy, they
are not recommended to be used as the main analysis tool in later design stages. As the
design evolves, the comprehension and accuracy of the results will become
increasingly important, hence more sophisticated methods are suggested to be used in
the later design stages.
Concept
For the next design stage -- concept -- either a multi-zone or a coarse-meshed
CFD analysis is recommended. Multi-zone requires less time to simulate the same
building compared to CFD. However, CFD can provide more detailed information
about the flow behaviour inside and outside the building. Bulk properties such as
temperature, air velocity, relative humidity, and contaminant concentration within
each zone can be obtained using multi-zone models. Simulation using CFD, on the
other hand, provides detailed information about the aforementioned properties at each
point of the domain. This analysis method can also be used for parametric studies
where variables can be changed in relation to each other in order to reach the most
suitable configuration. Examples of using CFD for conducting parametric studies can
be seen in (L. Wang & Hien, 2007; L. Wang et al., 2007).
Detail
As a design approaches its final stages, more detailed information will be
required. For the “detail” phase two coupled methods are suggested: 1- CFD and multi-
zone, 2- CFD and BES. As previously discussed in Section “CFD coupled with multi-
zone and BES”, the built-in network airflow models within BES models perform the
80 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
same as multi-zone models. Hence, the results they provide are the same in terms of
airflow and natural ventilation properties. BES models, however, also provide
information on a building’s energy consumption in addition to thermal performance
and can, therefore, be used for thermal performance studies. Obviously, simulating a
building’s thermal performance in addition to network airflow simulation increases the
computational time required.
Final stage
Highly accurate results are usually required at the final design stage.
Experimental methods can provide the most accurate results among the investigated
methods, however, they are costly and time-consuming, and hence, they are suggested
to be used at the final design stages. Small-scale experiments can be conducted at this
stage offering the best solution in terms of result accuracy. In experimental methods
increase in result resolution and variation generally increase cost (monetary and time)
associated with the analysis. Therefore, based on the information needed at the final
design stages, CFD coupled with small-scale experiments offer the best combination
for detailed and accurate results. Another benefit of this combined approach at the
“Final” design stage is that the results from numerical simulations provide a graphical,
and easily readable set of results on the performance of natural ventilation, in terms of
air movement and temperature distribution for designers and engineers. The results
detail and accuracy is particularly important at this stage since it is the last phase before
construction and any issue can be looked into before the construction starts. Any
changes to the design after the construction may become very costly and sometimes
impossible.
Construction and after construction (usage)
Data collected from full-scale experiments are both highly accurate and reliable.
However, in the case of multi-storey buildings, full-scale experiments cannot be
conducted prior to the construction due to the building size and model availability.
Hence, the final stage of the design process diagram (Figure 4.5) can be used to
evaluate natural ventilation and post occupancy research rather than as a prediction
tool.
In some cases, buildings do not perform in a way they were predicted. Such
discrepancies can be explained by a number of reasons including: 1- the accuracy of
prediction tools, 2- occupants interference with the building, 3- changes in building’s
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 81
surrounding environment due to new constructions, and 4- buildings not being
constructed as they designed. Research on buildings in their usage phase can help to
evaluate the intended design performance of the building after construction. Once the
actual performance has been quantified, various solutions can be offered to improve
the ventilation accordingly. Full-scale experiments are the most suitable methods for
design evaluation after construction due to acquisition of reliable data of on-site
situation.
The proposed methods for the “detail” stage usually provide results with an
acceptable level of accuracy. Therefore, in most cases, they can be used as the final
analysis method in the design process. In the case of crucial designs, the proposed
methods for “Final stage” should be used to mitigate the possible uncertainties
associated with the results obtained from the previous stages.
The proposed “natural ventilation design process” diagram is based on the
common design stages involved in the design of a multi-storey building. However,
each stage can be adjusted or bypassed based on the specific needs of each project.
For example, it is not common in design practice to use small-scale and full-scale
measurements in design of residential buildings, and this method of analysis can be
replaced with a combination of suitable alternatives. The last two proposed phases are
more common in research projects and for crucial studies of wind loads on high-rise
buildings.
82 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Figure 4.5. Natural ventilation design process model within the overall design process
4.5 CONCLUSION
This study has proposed a natural ventilation design process model to be used at
key design and construction phases for high-rise residential buildings. The design
model has been developed after identification, and critical review of, the analysis
methods used in the prediction of natural ventilation performance in such buildings.
Based on the literature reviewed, three main categories of evaluation methods
were identified where each of these categories was divided into a number of sub-
categories:
• Analytical and empirical methods
Analytical and
Empirical
Multi-zone CFD
CFD + BES
CFD + Multi-zone
CFD + Small-scale experiments
Small-scale
experiments
Full-scale experiments
Feasibility
Concept Design
Detail Design
Final Design
/Documentation
Construction
Design Stages Activities
After Construction
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 83
• Computational simulation methods: CFD, coupled method of CFD and BES
or network airflow models
• Experimental methods: full-scale experiments, and small-scale experiments
Analysis methods used for natural ventilation studies fall within one, or more, of
these above sub-categories.
It was found that analytical and empirical methods had a very limited
contribution to the methods used for natural ventilation studies, while CFD combined
with experimental methods was the most widely used. This can be explained by the
fact that analytical and empirical models are more case specific and have limited
application. On the other hand, CFD coupled with experimental methods can be
applied to various subjects. Additionally, this coupled method can provide detailed and
reliable information about natural ventilation performance.
The advantages and limitations of the implemented evaluation methods were
also investigated and reported. Each evaluation method was assessed against five
criteria, namely: accuracy, level of results’ resolution, cost, applicability to complex
geometries and the required time.
From this analysis, a natural ventilation design process model was proposed. The
proposed design process model is based on the different needs associated with each
design stage as well as the evaluation criteria used in the assessment of the identified
methods. Accordingly, quick and inexpensive methods were suggested to be used at
early design stages and more accurate methods to be employed as the design develops.
It also needs to be noted that although experimental methods can provide highly
accurate results compared to other methods, it is not common to use costly experiments
in natural ventilation design of regular multi-storey buildings. These methods are
mainly employed for research purposes and wind load studies of high-rise buildings
and skyscrapers where accuracy level matters most.
Acknowledgements
This research did not receive any specific grant from funding agencies in the
public, commercial, or not-for-profit sectors
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 85
4.6 APPENDIX
Table 4.3. Summary table
Reference Problem
focus
Building
Type
Ventilation
Type Methodology
Climate/
Region Criteria Comment
Kotani, Narasaki
et al. (1996)
(Kotani et al.,
1996)
Courtyards
High-rise
residential
Buoyancy-
driven
ventilation
Small-scale
experiments
Empirical model
Japan (hot-
humid)
Temperature
distribution
and air flow
rate
Airflow rate measurements device:
Omnidirectional temperature compensated
anemometer.
Temperature measurement devices: C-C
thermocouples.
Small-scale experiment was conducted in
scale of 1:100.
Wong, Feriadi et
al. 2000) (N.
Wong et al.,
2000).
Courtyards
Multi-
storey
(from three
to eighteen
storeys)
Full-scale and
small-scale
experiments
Singapore
(hot-humid)
Wind speed,
Wind speed
ratio
Small-scale experiments were conducted
using wind tunnels in scale of 1:200. Wind
tunnel results were compared to full-scale
measurement showing 20% higher velocity
readings.
Carrilho da
Graça, Chen et al.
Daytime
ventilation
Temperature,
Relative
k-ϵ turbulence model was used for CFD
simulations.
86 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Reference Problem
focus
Building
Type
Ventilation
Type Methodology
Climate/
Region Criteria Comment
(2002)(Carrilho
da Graça et al.,
2002).
and night
cooling
6-story
suburban
apartment
building
Wind-driven
ventilation
Coupled model of
building BES and
CFD
Beijing and
Shanghai
(hot-humid)
Humidity
(RH) and
average wind
speed
Priyadarsini,
Cheong et al.
(2004)
(Priyadarsini et
al., 2004)
Stack
systems
High-rise
residential
building
Buoyancy-
driven
ventilation
Small-scale
experiments
Singapore
(hot-humid)
Air speed and
airflow path
The 1:5 scale model of an apartment was
investigated in an open-circuit boundary
layer wind tunnel. Air velocity
measurement devices: Omnidirectional
velocity and temperature transducers. CFD
was used for illustration of velocity vectors.
Liping and Hien
(2007) (L. Wang
& Hien, 2007)
Different
ventilation
strategies and
façade design
18-story
residential
building
Wind-driven
ventilation
Coupled model of
building BES and
CFD
Singapore
(hot-humid)
Thermal
comfort
BES software: ESP-r
CFD software: FLUENT
BES and CFD exchanged data through an
interface.
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 87
Reference Problem
focus
Building
Type
Ventilation
Type Methodology
Climate/
Region Criteria Comment
Wang, Wong
Nyuk et al. (2007)
(L. Wang et al.,
2007)
Façade
design
optimisation
Multi-
storey
residential
apartment
Wind-driven
ventilation
Coupled model of
building BES and
CFD
Singapore
(hot-humid)
Thermal
comfort
The main focus in this study is thermal
comfort. CFD was coupled with BES to
provide more accurate information in terms
of ventilation.
BES software: ESP-r
CFD software: FLUENT
Wong, Prasad et
al. (2008) (P.
Wong et al.,
2008)
Double skin
façade
High-rise
office
building
(18 storey)
Wind and
buoyancy-
driven
ventilation
CFD simulations
Singapore
(hot-humid)
Thermal
comfort
CFD software: AIRPAK
k-ϵ turbulence model was used for CFD
simulations.
Yik and Lun
(2010) (Yik &
Lun, 2010)
Natural
ventilation
performance
and energy
saving
High-rise
residential
building
Wind-driven
ventilation
CFD simulations,
airflow network
model and building
heat transfer model
Hong Kong
(hot-humid)
Natural
ventilation
rate,
CFD software: FLUENT
Network airflow model: COMIS
Heat transfer model: HTB2
88 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Reference Problem
focus
Building
Type
Ventilation
Type Methodology
Climate/
Region Criteria Comment
evaluation
method
Indoor
temperature
Air-conditioner energy simulation program:
BECRES
Gao and Lee
(2011a) (CF Gao
& Lee, 2011a).
Openings
configuration
Multi-
storey
residential
building
Wind-driven
ventilation
CFD simulations
and full-scale
experiment
Hong Kong
(hot-humid)
MAA
Tracer-gas decay method was used for full
scale measurements.
CFD software: AIRPAK
RNG k-ϵ turbulence model was used for
CFD simulations.
Gao and Lee
(2011b) (CF Gao
& Lee, 2011b)
Window
types
Multi-
storey
residential
building
Wind-driven
ventilation
CFD simulations
and full-scale
experiment
Hong Kong
(hot-humid)
MAA
Tracer-gas decay method was used for full
scale measurements.
CFD software: AIRPAK
RNG k-ϵ turbulence model was used for
CFD simulations.
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 89
Reference Problem
focus
Building
Type
Ventilation
Type Methodology
Climate/
Region Criteria Comment
Chiang and Anh
(2012) (Chiang &
Anh, 2012)
Courtyards
Multi-
storey
apartment
(11 storey)
Buoyancy-
driven and
wind and
buoyancy-
driven
ventilation
CFD simulations
Yong-he-
Taipei,
Taiwan
(hot-humid)
Temperature
Velocity
k-ϵ turbulence model was used for CFD
simulations.
CFD software: PHOENICS-FLAIR
Prajongsan and
Sharples (2012)
(Prajongsan &
Sharples, 2012)
Ventilation
shafts
High-rise
residential
(25-storey)
Buoyancy-
driven
ventilation
using
ventilation
shafts
CFD simulations Bangkok
(hot-humid)
Average air
velocity and
air
temperature
Data collected from a full-scale experiment
was used for CFD validation. Hot-wire
anemometers were used for air velocity
measurements.
CFD software: DesignBuilder.
k-ϵ turbulence model was used for CFD
simulations.
Liu, Ford et al.
(2012) (P.-C. Liu
et al., 2012).
Segmentation
High-rise
office
building
Buoyancy
driven/ wind
and buoyancy
Single-cell EFM
model (semi-
empirical) and
Taipei,
Taiwan
(hot-humid)
Airflow rate
Flow pattern
Network air flow model
software: ESP-r
90 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Reference Problem
focus
Building
Type
Ventilation
Type Methodology
Climate/
Region Criteria Comment
driven
ventilation
network air flow
model
Ai, Mak et al.
(2013) (Z. Ai et
al., 2013).
Natural
ventilation
performance
and IAQ
High-rise
residential
building
Wind-driven
ventilation
Full-scale
experiment
Hong Kong
(hot-humid)
Ventilation
rate (m3/s),
ACH, RH,
temperature
Tracer gas decay method was used to define
ACH.
Wind velocity measurement device: Model
8475 air velocity transducer.
ACH, temperature and relative humidity
measurement device: Tracer gas CO2,
Telaire 7001 CO2 monitor.
Validity of some empirical models was also
investigated in this study.
Ai and Mak
(2014) (Z. T. Ai
& Mak, 2014).
Single-sided
ventilation
determination
methods
Multi-
storey
building
Wind-driven
ventilation
CFD simulations
and full-scale
experiment
Hong Kong
(hot-humid)
Ventilation
rate, ACH
Tracer gas decay method used for
determination of ventilation rate.
This paper investigates the applicability of
current empirical models for determination
of ventilation rate in multi-storey buildings.
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 91
Reference Problem
focus
Building
Type
Ventilation
Type Methodology
Climate/
Region Criteria Comment
RNG k-ϵ turbulence model was used for
CFD simulations.
CFD software: FLUENT
Fung and Lee
(2014) (Fung &
Lee, 2014).
Impact of
configuration
parameters
on natural
ventilation
Multi-
residential
building
Wind-driven
ventilation
CFD simulations
and full-scale
experiment
Hong Kong
(hot-humid) MAA
Tracer-gas decay method was used for full
scale measurements.
CFD software: AIRPAK
RNG k-ϵ turbulence model was used for
CFD simulations.
Zhou, Wang et al.
(2014) (C. Zhou
et al., 2014).
Natural
ventilation
design
optimisation
High-rise
residential
building
Wind driven
ventilation
CFD simulations
and full-scale
experiment
Chongqing
(humid-
subtropical)
Age of air,
Air change
rate
HOBO-U30 weather station was used to
obtain metrological data from rooftop of the
case study building. Telaire 7001 CO2 meter
was used for tracer gas decay method.
CFD software: FLUENT
Omrani,
Drogemuller et al.
Natural
ventilation
heuristics
High-rise
residential
buildings
Wind-driven
ventilation
CFD simulations
Brisbane,
Australia
(subtropical)
Velocity
magnitude
RNG k-ϵ turbulence model was used for
CFD simulations.
CFD software: FLUENT
92 Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools
Reference Problem
focus
Building
Type
Ventilation
Type Methodology
Climate/
Region Criteria Comment
(2014) (Omrani et
al., 2014)
Omrani, Capra et
al. (2015)
(Omrani et al.,
2015)
Provision of
balconies
High-rise
residential
building
Wind-driven
ventilation
CFD simulations
Brisbane,
Australia
(subtropical)
Average
velocity, flow
uniformity,
ACH
RNG k-ϵ turbulence model was used for
CFD simulations.
CFD software: FLUENT
Le, Li et al. (Le et
al.)
Wind
environment
High-rise
residential
district
CFD simulations
and full-scale
measurement
Changsha,
China
(subtropical)
Temperature,
RH and wind
speed
Omrani, Garcia-
Hansen et al.
(Omrani et al.,
2016a)
Relation of
meteorologic
al data with
High-rise
residential
building
(36-storey)
Wind-driven
ventilation
Full-scale
experiment
Brisbane,
Australia
(subtropical)
Air velocity 2D and 3D ultrasonic anemometers used for
the full-scale measurements
Chapter 4: Natural ventilation in multi-storey buildings: design process and review of evaluation tools 93
Reference Problem
focus
Building
Type
Ventilation
Type Methodology
Climate/
Region Criteria Comment
openings air
speed
Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings 95
Chapter 5: Predicting environmental
conditions at building site for
natural ventilation design:
Correlation of meteorological
data to air speed at building
openings
Omrani, S., Garcia-Hansen, V., Drogemuller, R., & Capra, B. (2016). Predicting
environmental conditions at building site for Natural ventilation design: Correlation of
meteorological data to air speed at building openings. 50th International Conference
of the Architectural Science Association 2016, Adelaide, Australia.
https://eprints.qut.edu.au/103498/
Statement of contribution of co-authors for thesis by published paper
The authors listed above have certified that:
1. they meet the criteria for authorship in that they have participated in the
conception, execution, or interpretation of (at least) that part of the
publication that lies within their field of expertise;
2. they take public responsibility for their part of the publication, while the
responsible author accepts overall responsibility for the publication;
3. there are no other authors of the publication;
4. potential conflicts of interest have been disclosed to (a) granting bodies, (b)
the editor or publisher of journals or other publications, and (c) the head of
the responsible academic unit; and
5. Consistent with any limitations set by publisher requirements, they agree to
the use of the publication in the student’s thesis, and its publication on the
QUT ePrints database.
96 Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings
The authors’ specific contributions are detailed below:
Contributor Statement of contribution
Sara Omrani Collected the experimental data, analysed
the data, produced the graphics, developed
the study, and wrote the manuscript.
Veronica Garcia-Hansen Assisted in developing the study, and
reviewed the manuscript.
Robin Drogemuller Assisted in developing the study, proof-
read and reviewed the manuscript.
Bianca Capra Assisted in developing the study.
Principal Supervisor Confirmation
I have sighted emails or other correspondence from all co-authors confirming their
certifying authorship.
__Veronica Garcia Hansen___ QUT Verified Signature______28/04/2017_____
Name Signature Date
Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings 97
Abstract
For the design of naturally ventilated buildings, information of air speed at the
openings of a building is important. However, the only data set usually available to
designers is meteorological data, such as wind speed and direction measured at
weather stations. This paper explores the ratio of air speed at building openings to the
wind speed measured at weather stations. Meteorological data from three weather
stations as well as air velocity that was obtained through full-scale physical
measurements were used in this study. The results showed that air speed at building
openings was about half of the wind speed recorded at the closest station to the case
study. This ratio reduced to approximately 30% when comparing to the weather
stations located in greater distance and more open areas. Given that air speed at the
openings has a direct relation to the ventilation rate, employing these ratios to the
available weather data when designing for natural ventilation, can provide more
realistic picture of natural ventilation performance.
Keywords: Natural ventilation; Air speed at openings; Meteorological data;
Full-scale experiment
5.1 INTRODUCTION
Due to the oil and energy crises, energy efficiency policies have experienced a
rapid growth in many countries around the globe over the last four decades (e.g.
Europe, Japan, the United States, Australia, etc.) (Geller, Harrington, Rosenfeld,
Tanishima, & Unander, 2006). This includes building energy regulations and standards
(IECC, 2012; Recast, 2010). Buildings, as one the main energy consumers, have a
great potential to contribute to energy savings by adopting passive and low cost
strategies. However, passive strategies and design based on climate are often
disregarded in rapidly growing high-rise buildings which makes them highly energy
intensive (Cheung et al., 2005; R. Kennedy et al., 2015).
Appropriate design of natural ventilation as a passive cooling strategy can
provide thermal comfort for the building’s occupants (M. Liddament, Axley,
Heiselberg, Li, & Stathopoulos, 2006) which can result in reduced use of air-
conditioners and hence, save energy (Luo, Zhao, Gao, & He, 2007). Implication of
natural ventilation is even more feasible in cooling-dominant climates. Furthermore,
98 Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings
natural ventilation can improve the Indoor Air Quality (IAQ) by replacing the stale air
with the fresh air from the outside (M. W. Liddament, 1996).
Natural ventilation mainly relies on outside wind and the resultant pressure
difference for conditioning the space. Since wind is intermittent in nature and the
process involved is rather complex, it is difficult to predict natural ventilation
performance (Allocca et al., 2003). A major problem for building designers is that
accurately predicting environmental conditions inside and around a proposed building
is difficult. Having said that, understanding the potential air speed at the openings as a
representor of ventilation rate can be a step toward an accurate natural ventilation
prediction and a successful design.
To investigate the potential air velocity at building openings in relation to
meteorological data, wind speed data at openings of an apartment in a high-rise
residential unit in Brisbane, Australia were collected. In addition, weather data from
three different weather stations were obtained. The chosen weather stations are situated
in locations with different terrain roughness and various distances from the case study,
which allows further comparison considering urban context.
5.2 BACKGROUND
The ventilation rate has a direct relation with the air speed at the buildings
openings. In its simplest form it can be expressed as:
Q=VA (5.1)
Where Q is ventilation rate (m3/s), A is the area of opening (m2) and V is the air
velocity through the openings (m2/s). Hence, the air velocity at the buildings openings
can be used as a good indication of ventilation rate.
The main data source available to architects and building designers is
meteorological data from the weather station nearest to the location of interest.
Weather stations are mostly located in open areas and the meteorological data from
them are likely to be different from the expected wind at dense urban settings (Truong,
2012). Furthermore, wind magnitude changes with height and meteorological data are
usually measured at the height of 10 m while building openings can be above or below
that height.
Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings 99
Wind speed at different heights in relation to the wind speed at a reference height
can be expressed by power law equation (Helmut E. Feustel, 1999):
𝑉𝑧 = 𝑉𝑟𝑒𝑓(𝑍
𝑍𝑟𝑒𝑓)𝛼 (5.2)
Where Vz is wind speed at height z (m/s), Vref is wind speed at the reference
height of zref (m/s) and α exponent represents the terrain roughness and varies from
0.15 to 0.35. A greater value indicates a rougher terrain (Hazim B Awbi, 2003).
However, wind speed at building façade and openings is always lower than the Vz
value obtained from equation (5.2). That is due to the positive and negative pressure
built up as a result of wind hitting an obstacle (e.g. building). In order to use
meteorological data for natural ventilation design, knowing the relation between the
reference wind speed and the wind speed at the openings is an important factor which
can help in a more realistic prediction of natural ventilation. The lack of such a relation
in the literature was motivation of the current study.
5.3 METHODOLOGY
In order to investigate the correlation of weather data and air velocity at building
openings, full-scale physical measurement of air velocity at openings of a residential
apartment was carried out. Analysis of the collected data in addition to the available
weather data helps to reveal any possible connections. Full-scale measurement was
chosen as it can yield more reliable information compared to the other available
methods (e.g. small-scale experiments, simulation software, etc.) (Chen et al., 2010).
Selected weather stations and the case study used for the full-scale measurements are
described in the following sections.
5.3.1 Case study
A 36-storey building located at Brisbane, Australia (latitude: -27.46, longitude:
153.03) was chosen as the case study for this research. This building is located near
the Brisbane Central Business District (CBD) in a relatively dense urban layout.
However, there are no major obstructions in the case study’s immediate surroundings.
The building is oriented 35° from North toward West and is next to the Brisbane River
on one side and adjacent to a street approximately 25 meters wide on the other side.
Figure 5.1 shows the location of the case study in relation to its surroundings. A
residential unit located on the eastern side of the building at the fifth floor, was used
100 Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings
for the data collection. This 2-bedroom apartment features two balconies at two
opposite sides of the living area and is situated about 18 m above the ground. The case
study unit was vacant when measurements were conducted and no mechanical or fan-
assisted ventilation was operating.
Figure 5.1. Case study’s site plan.
The instruments employed to measure wind velocity at the case study’s openings
were a Windmaster 3-axis ultrasonic anemometer (3D) and a 2D WindSonic
anemometer (2D), commercially produced by Gill Instruments. The sensors allow
accurate measurement of wind speed and direction at resolution of 0.01 m/s. Wind
speed accuracy is 1.5% for 3D and 2% for 2D. The 3D sensor was attached to the
exterior of the southern balcony’s parapet wall and the 2D anemometer was placed
inside the northern balcony. The parapet walls of balconies are 1.2m high thus the
sensors were installed at a height of 1.3 m from the unit’s balcony floor. Wind velocity
was measured for 30 hours at sampling rate of 1Hz starting at 1:00 pm and ending 7:00
pm the day after. The authors believe that due to fluctuating nature of wind and the
frequent sampling rate of data logging adopted in this study, 30 hours of air velocity
data is enough to represent various wind speed ranges. All the doors and openings -
except for balcony doors- were kept closed for the duration of the data collection. The
sensors were placed close to the openings which provides information on external
airflow near the openings. Plan and placement of the sensors is presented at Figure 5.2.
Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings 101
Figure 5.2. Case study’s plan (right) and photos of sensors (left).
5.3.2 Weather stations
Meteorological data for this study was obtained from three weather stations:
Brisbane station, Brisbane Airport and Archerfield stations which are located
approximately 2 km, 9 km and, 12 km from the case study building respectively
(Figure 5.3). Information on the weather stations is presented in Table 5.1.
Table 5.1. Weather stations information
Weather station Distance to
case study
Latitude Longitude Station height
Brisbane Station ~2 km -27.4808 153.0389 8.13 m
Brisbane Airport ~9km -27.39 153.13 4.51 m
Archerfield ~12 km -27.5717 153.0078 12.5 m
The wind speed and direction 30-minute data for the duration of the experiment
was downloaded from Australian Government Bureau of Meteorology website
(Australian Government Bureau of Meteorology, 2016). Wind speed data are averaged
over 10 minutes and are rounded values with no decimal places, and wind direction
data is presented in 16-compass points.
Wind speed and direction data acquired from the installed anemometers together
with weather station data were analysed to investigate objectives of this study.
102 Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings
Figure 5.3. The weather station locations in relation to the case study building.
5.4 RESULTS AND DISCUSSION
This section presents and discusses the analysis of weather stations data along
with the measurement results from the installed sensors.
5.4.1 Weather stations
Figure 5.4 represents wind speed (Figure 5.4-left), and wind directions (Figure
5.4-right) captured at the three weather stations: Brisbane, Brisbane Airport, and
Archerfield stations for the duration of the experiment. Both graphs show that despite
the long distances between the stations and the different urban layouts and contexts,
wind speed changes and the prevailing wind direction are very consistent at all three
stations, and the wind predominantly blows from NNE to ENE. The graph at the top
shows the lower wind speed recorded at Brisbane station compared to Brisbane airport
and Archerfield stations. Since Brisbane weather station is located close to the
Brisbane CBD in a predominantly residential suburb with higher density urban layout
compared to Brisbane Airport and Archerfield, the recorded wind speed, as expected.
is lower due to the adjacent obstructions. Brisbane Airport and Archerfield are both
located in open terrain, and thus present a similar range of wind speed changes and
about twice that of the Brisbane station. In addition, as Brisbane Airport is close to a
large body of water (ocean), wind magnitude recorded at this station would be affected.
Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings 103
This is a justification for the higher mean wind speed captured by this station compared
to Archerfield station, despite the higher elevation of Archerfield station.
Figure 5.4. Wind speed change (left) percentage of different wind directions (right) at Brisbane,
Brisbane Airport and Archerfield weather stations.
Figure 5.5 represents wind speed at Archerfield and Brisbane Airport stations in
relation to Brisbane station’s wind speed. Regression lines confirm a linear relation
between air speed changes at these weather stations. Again, it also confirms that
Brisbane station had the lowest readings of wind speed with values of about 56% and
65% of wind speed at Brisbane Airport and Archerfield stations respectively.
Figure 5.5. Regression lines between wind speeds recorded at Brisbane station expressed according to
Brisbane Airport and Archerfield stations wind speed.
VBr = 0.6474VAr - 0.5796
R² = 0.7621
VBr = 0.5596VAi - 0.356
R² = 0.7133
-1
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
0 2 4 6 8
Bri
sba
ne
sta
tio
n w
ind
sp
eed
(m
/s)
Wind speed (m/s)
Brisbane Airport Archerfield
104 Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings
The correlations between the three weather stations are expressed in Table 5.2
where VBr, VAr, and VAi refer to Brisbane station, Archerfield and Brisbane Airport
stations wind speed respectively.
In view of confidence in the regression equations, the fact that the available wind
data is averaged over 10 minutes and is presented in rounded values without decimal
places might have introduced some marginal errors. To this end, the acquired
regression lines are considered reasonably valid.
Table 5.2. Linear regression equations of Brisbane station wind speed (VBr) on wind speed for
Brisbane Airport (VAi) and Archerfield (VAr) stations
Weather stations Regression equation R2
Brisbane and Brisbane Airport
stations
VBr = 0.5596VAi - 0.356 0.71
Brisbane and Archerfield stations VBr = 0.6474VAr - 0.5796 0·76
To summarize, data analysis of the weather stations exhibit consistency in wind
direction collected at different stations located in areas with different terrain roughness
and urban context over 10 km apart. This consistency in the recorded wind direction
can be mainly due to the unobstructed immediate surroundings of the weather stations
with the minimum distance of 30 meters even in the case with the highest density
setting (Brisbane station). In addition, the effect of urban density was clear in the
recorded wind speeds, and as expected, Brisbane station represented the lowest speed
range among the three stations. Most importantly, the linear relation between the air
speed values at different stations shows that wind speed changes with similar patterns
in different locations. Therefore, in the following section, results will be presented and
discussed using Brisbane station data only. The selection was made as Brisbane station
is located in a similar urban context and is the closest station to the case study building.
5.4.2 Wind speed at building openings
To explore the potential wind speed outside of the case study’s openings in
relation to the meteorological data, the collected data by the installed anemometers
was averaged over 10 minute intervals to allow comparison with Brisbane station’s
Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings 105
data. Wind speed changes during the measurement period recorded by Brisbane
weather station, 2D and 3D sensors are presented in Figure 5.6.
Figure 5.6. Wind speed changes of Brisbane station, 2D and 3D for duration of the data collection.
The intermittent line representing air speed changes at the 3D sensor is due to
some glitches in the logging system, which resulted in loss of some of data. As can be
seen 2D and 3D change trends are similar to the Brisbane station at a lower speed.
To investigate the correlation of this trend, data recorded by 2D and 3D sensors
were plotted in relation to Brisbane station wind speed data (Figure 5.7). It is evident
that a decrease in Brisbane station wind speed results in wind speed decrease at both
measurement points (2D and 3D). The obtained R-squared values equal to 0.7 for 2D
and 0.78 for 3D versus Brisbane station, confirm that the acquired regression lines are
acceptably valid. It also demonstrates that the air speeds captured by the sensors are
very similar in values and are roughly half the values recorded at Brisbane station.
Figure 5.7.Variation of wind speed recorded at measurement points (2D and 3D) versus Brisbane
station wind speed.
106 Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings
In summary, comparing wind speed outside of the case study’s openings with
Brisbane station wind speed, the same variation pattern was observed at a much slower
speed range (about 50%). This percentage is obviously smaller when compared to the
Brisbane Airport and Archerfield weather stations (nearly 30%). This slower speed
range was expected, as the building acts as a large obstruction on the airflow path and
as a result, the positive pressure built on the windward side results in a lower air speed.
In addition, air speed values were very similar at both the inlet and outlet of the case
study which is not surprising since openings were the same size. This can be explained
by conservation of mass.
It needs to be considered that the case study was located at fifth floor (nearly
18m above the ground), and there were no major obstructions in the immediate
surroundings of it. Application of the acquired results to a broader context needs to be
further investigated. The measured values are expected to be lower in the case of
buildings at lower floors in contrast to the higher floors, which would be expected to
be higher. In addition, that having openings at two opposite sides might have
accelerated the air speed compared to a case with openings at only one side.
Air flow rate in cross ventilation is higher than that of the single-sided ventilation
(Jiang et al., 2003). Therefore, the resultant air speed is expected to be much lower in
a building with openings only at one side (single-sided ventilation).
5.5 CONCLUSION
Natural ventilation rate directly correlates to air velocity at building openings.
Understanding the potential values of wind speed at the openings can result in better
estimation of ventilation rate. Considering meteorological data is the main data source
available to the building designers, this study explored possible air velocity at building
openings in relation to the available meteorological data. To this end, air speed at
openings of a high-rise residential unit was measured using 2D and 3D anemometers.
Furthermore, wind speed data from three weather stations situated in locations with
varying terrain roughness and different distances from the case study was also
obtained.
Firstly, wind speed and direction from three different weather stations situated
in different urban context were compared. The results showed consistency in the
captured directions by all three stations. Also, wind speed analysis showed similar
Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings 107
wind speed fluctuation pattern. Scatter plots of wind speed at the weather stations
confirmed this relation. As expected, the lowest wind speed range was from the station
located in the denser urban layout.
Air speed measurement data for the unit was then compared to the
meteorological data from the closest weather station to the case study. Once more, a
similar change trend was evident between weather data and the measured values at the
case study. Interestingly, regression lines revealed that the wind speed through the
openings was approximately half of the wind speed measured at the weather station.
The reference weather station was in a similar urban setting as the case study and
closest to it (about 2km). Unsurprisingly, this ratio reduced to nearly 30% when
compared to the two other meteorological stations located in low rough terrains further
way from the case study building.
In conclusion, when using the meteorological data to design for natural
ventilation, similarity of urban context of the closest weather station to the site of
interest is an important parameter which can yield values close to those that can be
expected at building openings. Even in that case, the potential air velocity at building
openings may not exceed half of the wind speed at the reference weather station.
However, in a case that design site is in a dense urban setting, chances of existing any
weather station in similar urban context is very low as meteorological station are
usually located in open areas with minimum obstructions. If it was the case, lower ratio
of wind speed of the reference station should be expected at the openings (roughly
30%). Findings of this study can help in better use of meteorological data in natural
ventilation design.
5.6 FUTURE WORK
The current study provided building designer with ratios of potential air velocity
at building openings to the meteorological data, considering urban context and distance
to the building of interest. However, when using results of this research, the following
factors needs to be taken into consideration. Firstly, the measurements were conducted
in a cross-ventilated unit with two openings at two opposite sides. Airflow produced
by cross ventilation can be much higher than the airflow produced by single-sided
ventilation. Even in such an instance, wind speed at inlet is far less than that of at the
same height in the free atmosphere. Hence, lower values of air velocity should be
108 Chapter 5: Predicting environmental conditions at building site for natural ventilation design: Correlation of
meteorological data to air speed at building openings
expected at the openings of a building utilising single-sided ventilation. Having said
that, more studies are needed to investigate the amount of airflow at openings in the
case of single-sided ventilation. Secondly, the measurements of this study were done
at a case study located at fifth floor. As wind magnitude increases with height, higher
ratios at upper levels and lower rations at lower levels should be expected.
In future research, it would be desirable to validate the applicability of the results
of this research to a broader context and different heights.
5.7 EPILOGUE
In terms of the relation of this chapter to the whole thesis, it mainly examines
the first step of the natural ventilation design process model (Figure 4.5) that was
proposed in Chapter 4 by looking at the relation of wind speed and indoor airflow. A
simple empirical model is proposed in this chapter which calculates the approximate
airspeed at building openings using meteorological data. Using such a model would
allow quick estimation of ventilation performance of an intended design which would
suit the design requirements of early design stages. Consequently, empirical models
can be useful tools for the early design stages.
It needs to be noted that the intention of this study was mainly to explore the
possibility of making estimations using the most readily available sources of data
(meteorological data) rather than to propose a model that can widely be adopted in
natural ventilation design. Since only one building was tested, the proposed empirical
model needs further validation to confirm its applicability to a broader context.
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 109
Chapter 6: Effect of natural ventilation
mode on thermal comfort and
ventilation performance: Full-
scale measurement
Omrani, S., Garcia-Hansen, V., Capra, B., & Drogemuller, R. Effect of natural
ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement. Energy and Building
Statement of contribution of co-authors for thesis by published paper
The authors listed above have certified that:
1. they meet the criteria for authorship in that they have participated in the
conception, execution, or interpretation of (at least) that part of the
publication that lies within their field of expertise;
2. they take public responsibility for their part of the publication, while the
responsible author accepts overall responsibility for the publication;
3. there are no other authors of the publication;
4. potential conflicts of interest have been disclosed to (a) granting bodies, (b)
the editor or publisher of journals or other publications, and (c) the head of
the responsible academic unit; and
5. Consistent with any limitations set by publisher requirements, they agree to
the use of the publication in the student’s thesis, and its publication on the
QUT ePrints database.
The authors’ specific contributions are detailed below:
Contributor Statement of contribution
Sara Omrani Collected the experimental data, analysed
the data, conducted literature review,
produced the graphics, developed the
study, and wrote the manuscript.
110 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
Veronica Garcia-Hansen Assisted in developing the study, and
reviewed the manuscript.
Bianca Capra Assisted in developing the study and
analysis, proof-read and reviewed the
manuscript.
Robin Drogemuller Assisted in developing the study, proof-
read and reviewed the manuscript.
Principal Supervisor Confirmation
I have sighted emails or other correspondence from all co-authors confirming their
certifying authorship.
__Veronica Garcia Hansen___ ______28/04/2017_____
Name Signature Date
QUT Verified Signature
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 111
Abstract
Natural ventilation can provide building occupants with thermal comfort and a
healthy indoor environment. Among all the design related parameters that affect
ventilation performance, ventilation mode (i.e. single-sided and cross ventilation) is
perhaps the main one. The current study uses full-scale in-situ measurements to
investigate the effect of ventilation mode on thermal comfort and ventilation
performance of a high-rise case study unit. Two modes of natural ventilation, single-
sided and cross ventilation, are investigated. Air velocity, temperature and relative
humidity were measured for both ventilation modes on two consecutive days in
summer. Indoor thermal conditions were evaluated using the extended Predicted Mean
Vote (PMV) and Standard Effective Temperature (SET*) comfort models. In addition,
the relationship between reference wind speed and internal airflow, airflow
distribution, and the effect of wind direction on internal airflow were investigated for
both single-sided and cross ventilation. Finally, the implications of the research
outcomes on natural ventilation design are discussed. Indoor thermal conditions were
found to be within the comfort zone for more than 70% of the time under cross
ventilation operation while single-sided ventilation provided adequate thermal
conditions for only 1% of the time. Results from this study highlight a significantly
better performance of cross ventilation over single-sided ventilation.
Keywords: Natural ventilation; thermal comfort; full-scale experiment; high-
rise residential; single-sided ventilation; cross ventilation
Nomenclature
CBD Central Business District
D Wind direction
e Expectancy factor
PMV Predicted Mean Vote
PPD Predicted Percentage of Dissatisfaction
SET* Standard Effective Temperature (˚C)
T Temperature (˚C)
U Airspeed (m s-1)
112 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
Vref Reference wind speed (m s-1)
Vz Wind speed at height z (m s-1)
x Length (m)
Z Building height (m)
Zref Reference height (m)
α Terrain roughness exponent
6.1 INTRODUCTION
Indoor environmental quality can be defined by parameters such as indoor air
quality, thermal comfort, visual comfort, and acoustics. Among these parameters,
indoor environmental quality is mostly affected by thermal comfort (Cao et al., 2012;
Frontczak & Wargocki, 2011; Marino, Nucara, & Pietrafesa, 2012). Today, indoor
thermal comfort in buildings is increasingly being achieved by application of air-
conditioners (Mat Santamouris, 2016). This increased use of air-conditioners results
in an increase in energy consumption and consequent negative environmental effects.
In developed countries, the HVAC (Heating, Ventilation and Air-Conditioning)
systems of residential buildings consume more than two-thirds of the energy delivered
to the buildings (Orme, 2001). Such high levels of energy consumption, as well as the
consequent burdens on the environment, have made energy efficiency strategies a
priority in building regulations in many countries (Pérez-Lombard et al., 2008; Roetzel
et al., 2010). Because of the energy intensive nature of this, there is a significant
potential for reduction of energy usage in buildings (W. Miller & Buys, 2012),
particularly by utilising passive cooling and heating in cooling-dominant, and heating-
dominant climates respectively.
Natural ventilation is one of the most effective passive cooling strategies,
especially for cooling-dominant climates and can provide building occupants with a
comfortable thermal condition and a healthy indoor environment (Liping & Hien,
2007; Matheos Santamouris & Allard, 1998). Furthermore, 30% to 40% less energy
consumption is reported in naturally ventilated buildings compared to mechanically
ventilated buildings (Gratia & De Herde, 2004b; Kolokotroni & Aronis, 1999; Schulze
& Eicker, 2013). Natural ventilation design, however, can be challenging due to the
complex and turbulent flows in and around buildings (Chen, 2004; Hu, Ohba, &
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 113
Yoshie, 2008; Seifert, Li, Axley, & Rösler, 2006), especially in dense urban areas.
Accordingly, natural ventilation has been largely disregarded in the design of high-rise
buildings resulting in them being highly energy intensive due to the use of mechanical
ventilation systems (Cheung et al., 2005; R. Kennedy et al., 2015).
In Australia, construction of high-rise buildings is experiencing significant
growth. Today, the number of approvals for high-rise constructions are higher than
ever with an approximate increase of 300% over the last ten years ("Australian Bureau
of Statistics," ; Kusher, 2016). Considering this large volume of apartments, an
adoptation of passive strategies such as natural ventilation offers great potential for
energy conservation in such buildings, whereas, a sole reliance on air-conditioning
may impose an excessive burden on both energy suppliers and the environment.
Natural ventilation performance is influenced by a combination of different
design features such as ventilation mode (i.e. single-sided ventilation and cross
ventilation), window to wall ratio, opening type, and floor area. Among these
parameters, ventilation mode has the largest impact on the ventilation rate of a building
(Fung & Lee, 2014). A number of studies have investigated cross ventilation using
full-scale measurements (Larsen & Heiselberg, 2008; Lo & Novoselac, 2012; Park,
2013), small-scale experiments (Karava, Stathopoulos, & Athienitis, 2011) and the
combination of both (Katayama, Tsutsumi, & Ishii, 1992). Lo and Novoselac (Lo &
Novoselac, 2012) investigated the dynamic nature of cross ventilation by measuring
wind speed and direction, façade pressure, and tracer gas concentration in a single
storey multi-zone building with cross ventilation. They found that there is a linear
relation between wind velocity and cross ventilation flow rate. In addition, their study
showed that wind fluctuations affect the façade pressure, therefore, the steady
assumption may not be always appropriate (Lo & Novoselac, 2012). The effect of wind
fluctuation on airflow rate was further investigated by Park (Park, 2013) using
experimental data from the one-year measurement of wind properties, flow rate, and
the pressure difference between the openings of a cross ventilated mock-up building.
It was concluded that fluctuating components of wind strongly affect the openings
pressure coefficient. Cross ventilation characteristics were also studied by Karava et
al. (Karava et al., 2011) using a scaled model of a single-zone building in a wind
tunnel. Their study provided precise internal flow pattern for different opening
configurations. The indoor airflow pattern presented in their study was later used in
114 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
several research for computational simulation validation (C.-R. Chu & Chiang, 2013,
2014; J. I. Perén, T. van Hooff, B. C. C. Leite, & B. Blocken, 2015; Ramponi &
Blocken, 2012; Tong, Chen, Malkawi, Adamkiewicz, & Spengler, 2016).
In addition to the research related to cross ventilation, there is a number of
studies about single-sided ventilation and its characteristics. The parameters that affect
air change rate of single-sided ventilation were investigated using a full-scale wind
tunnel (Larsen & Heiselberg, 2008). It was found that incident wind angle affects air
change rate and the dominancy of the driving force (wind and temperature) of single-
sided ventilation. Air change rate in single-sided ventilation was also measured in a
number of high-rise residential units in Hong Kong (Z. Ai et al., 2013). The
measurement results show that at wind speed above 0.3 m/s, single-sided ventilation
can meet the required air change rate by ASHRAE 62 (A. ASHRAE, 2010). The
collected data from one of the subject units of this study (Z. Ai et al., 2013) along with
CFD simulations were later used for evaluation of characteristics differences of single-
sided ventilation in single-storey and multi-storey buildings (Z. T. Ai & Mak, 2014).
Their results indicate that the available empirical methods are not applicable for
prediction of single-sided ventilation in multi-storey buildings mainly due to the
differences in envelope flow patterns of multi-storey and single-storey buildings.
The extensive review of the literature reveals the following gaps with regards to
the studies about natural ventilation mode: 1- The majority of the studies available are
based on low-rise buildings and simple geometries, and very few studies with a focus
on high-rise buildings were found, 2- nearly all of the available studies focus only on
one ventilation mode (single-sided or cross ventilation) and detailed comparison of
these ventilation modes in real case studies is yet to be thoroughly investigated 3- the
effect of ventilation mode on indoor thermal comfort has not been investigated.
Considering the exponential growth of high-rise buildings construction, the land and
space arrangement restrictions for application of cross ventilation in such buildings,
and the crucial role of ventilation mode on determination of ventilation performance
and indoor thermal conditions, there is a need for in-depth investigation of the effect
of ventilation mode on ventilation performance and thermal comfort of high-rise
buildings.
This study aims, therefore, to 1- extend the existing knowledge about ventilation
characteristics of single-sided and cross ventilation under similar conditions which is
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 115
a prerequisite of a successful natural ventilation design, and 2- to investigate the effect
of natural ventilation mode on ventilation performance and indoor thermal comfort of
high-rise buildings. Accordingly, using a case study approach, this study purposely
sets out to collect data from an existing apartment that is exposed to real environment
conditions where wind speed and direction regularly change to quantify these effects
on thermal comfort and ventilation performance. A case study approach enables
monitoring a phenomenon in depth over a set time period by collecting information
from various data sources (Swanborn, 2010). Among the methods available for natural
ventilation investigation (e.g. CFD, small-scale experiment, and full-scale
experiment), a full-scale experiment performed in-situ can yield the most reliable
information since it measures real conditions and thus results that reflect reality (Chen,
2009; S. Omrani, V. Garcia-Hansen, B. Capra, & R. Drogemuller, 2017). To this end,
field measurements of air velocity, temperature and relative humidity were undertaken
in an apartment in a high-rise residential building in Brisbane under cross flow
ventilation and single-sided ventilation configurations. The same floor plan has been
used for each ventilation mode and measurements took place under very similar
weather conditions. The collected data was then analysed in terms of thermal comfort
using the extended PMV and SET* comfort models. Natural ventilation performance
was also evaluated by analyses of air velocity. Air velocity can be used as an indication
of ventilation performance as it linearly correlates to other measures such as Air
Change per Hour (ACH) and ventilation rate (Lo & Novoselac, 2012). Mean
ventilation rate can also be estimated from airspeed data at an opening by multiplying
by the opening area. Furthermore, air distribution and flow uniformity can be analysed
using air velocity data. Accordingly, detailed analyses of indoor air velocity in relation
to the reference wind, airflow distribution, and wind direction and resultant internal air
velocity were conducted. From these, design implications of the results for the
improvement of natural ventilation performance are discussed.
6.2 METHODOLOGY
This study utilizes a case study approach to investigate natural ventilation
modes. This approach involves in-situ measurements of airflow properties,
temperature, and relative humidity in a high-rise case study building. Accordingly, an
apartment, capable of being operated in both single-sided and cross-ventilation mode,
was selected for this study (explained in detail in section 6.2.2). Airspeed and
116 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
direction, temperature, and relative humidity were measured at different points of the
case study unit for both single-sided and cross flow ventilation (explained in detail in
section 6.2.3). Data was collected during two consecutive days in summer (January).
The hot weather conditions of January represent the worst case scenario in terms of
weather, and the time when cooling strategies are most needed. The gathered data, was
then used to calculate indoor thermal conditions within the case study unit. In addition,
meteorological data obtained from the nearest weather station along with the measured
parameters were used for evaluation of ventilation performance. Figure 6.1 illustrates
the research approach employed in this study, correlating methods with research
outcomes.
Figure 6.1. Illustration of the employed methods and the relation to the research outcome.
6.2.1 Climate Conditions
This study was performed in Brisbane, Australia’s third largest city. Situated at
27.4° south latitude, Brisbane has a subtropical climate characterised by warm humid
summers and mild to cool winters. Relative humidity ranges from 50% to 70% on
average and mean maximum daily temperature lies between 20°C in winter (July) and
30°C in summer (January). Wind speed average is 3.6 m/s dominantly blowing from
south and south-west in the morning and afternoon respectively (Australian
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 117
Government Bureau of Meteorology, 2016). Figure 6.2 shows Brisbane’s mean
temperature, wind speed, and relative humidity. Due to the hot and humid nature of
Brisbane’s climate, cooling is required for the majority of the year. In addition, wind
speed which is the main driving force of natural ventilation is higher in the months
with higher temperature. Natural ventilation, therefore, is a feasible passive strategy
for providing thermal comfort in Brisbane’s climate.
Figure 6.2. Average maximum daily temperature (A), wind speed (B), and relative humidity (C) in
Brisbane (2010-2015) (Australian Government Bureau of Meteorology, 2016).
6.2.2 Case study building
A unit in a 36-storey residential building located near the Brisbane Central
Business District (CBD), Australia, was chosen for the full-scale measurements.
Figure 6.3 shows the site plan for the case study building (left) and the local
topography through east-west section (top-right) and north-south section (bottom-
right). The case study building containing the selected unit is highlighted in black. As
118 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
can be seen in Figure 6.3-left, the building is adjacent to the Brisbane River (~250m
wide) from the southern side and a 25-meter wide street from the northern side.
Brisbane CBD, which is dominated by high-rise buildings, is located to the west. As
illustrated in Figure 6.3-right, the height and density of adjacent buildings are
relatively low on the eastern side. At the building’s southern side, next to the river,
there is a parkland and no major construction up to a 120m distance from the river.
The northern side, however, has a relatively high building (approximately 35m)
located across the street. The case study building is oriented 35° from north toward the
west.
The case study unit is located at the eastern end of the building's fifth floor, about
18m above ground level. The apartment layout contains two bedrooms, a living area
and two balconies at each end of the living area. Such a layout enabled measurement
for both single-sided and cross ventilation configurations. All the measurements took
place in the living area and the balconies. Accordingly, doors and windows to the
bedrooms remained shut for the duration of the experiments. The case study unit was
unoccupied and there were no heat sources or mechanical ventilation operating during
the data collection.
Figure 6.3. Case study’s site plan (left), and schematic east-west section (top-right), and north-south
section (bottom-right).
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 119
Figure 6.4 shows plan layout of the case study. Two tests for two ventilation
modes were carried out. Test-1 was configured for cross flow ventilation, where the
balcony doors at the two opposite ends of the living area were fully open for the
duration of the test. Test-2 measured airflow characteristics under single-sided
ventilation, where the balcony door at the northern side was closed while the southern
door was kept fully open. Configurations of the openings are presented in Figure 6.5.
The balcony doors were both sliding aluminium doors with glazing, 3m wide, 2.5m
high with an operable area of 1.16m x 2.5m=2.9m2. This corresponds to 8% of the
gross living floor area and represents 30% porosity (opening area divided by wall
area).
Figure 6.4. Plan layout of the case study
6.2.3 Experimental setup and instrumentation
Airspeed and direction, relative humidity, and temperature were measured
simultaneously at different points of the case study for both tests. Figure 6.5 shows the
test configurations and measurement points. Measurements were conducted using a
3D ultrasonic anemometer, three 2D ultrasonic wind sensors, two air velocity
transducers, six thermometers and one hygrometer. The sensors specifications are
presented in Table 6.1. The employed instruments were factory-calibrated and were
used in this study for the first time. The factory calibration, therefore, was considered
120 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
adequate for the purpose of this study. All sensors were installed at a height of 1.2m
above the floor, thus within the breathable zone, also represented the head level of a
sitting adult occupant. Temperature and air velocity were measured at the same
locations and heights. Five out of the six wind sensors were placed in a horizontal line
along the mainstream flow direction and one was placed at the corner of the living
area. The 3D anemometer (P1) was attached to the outside of the balcony’s parapet
wall to capture the airflow properties at the immediate entrance to the case study unit.
Table 6.1. Summary of the instrumentation
Instrument (model,
manufacturer)
NO. Parameters Accuracy and resolution
3D anemometer (WindMaster
3-axis ultrasonic anemometer,
model number 1590-PK-020,
Gill instruments)
1 U,V,W vectors
Speed: <1.5% RMS
@12 m/s
Direction: 2° @12m/s
2D anemometer (Windsonic
2-axis ultrasonic anemometer,
Option 1, Gill instruments)
2 Wind speed and 2D
direction or U and V
vectors
Speed: 2% @12m/s
Direction: 3° @12 m/s
2D anemometer (WindSonic
ultrasonic anemometer,
Option 4, Gill instruments)
1 Wind speed and 2D
direction or U and V
vectors
Speed: 2% @12m/s
Direction: 3° @12 m/s
Air velocity Transducer (8475
series, TSI)
2 Air velocity 3% of reading from 20°
to 26° C.
1% of selected full-scale
range (2.5 m/s)
Thermometers (DS1922T,
iButton, Maxim integrated)
6 Temperature Resolution: 0.0625
Hygrometer (DS1923,
iButton, Maxim integrated)
1 Relative humidity Resolution: 0.04
The duration of measurements for each test was 24 hours during summer (13-
15th of January, 2016). Air velocity, temperature, and relative humidity were measured
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 121
at a sampling rate of 5Hz to 1 minute. All measurements took place on days with clear
skies and no precipitation, allowing consistency of analysis.
To have comparative results for both ventilation modes (Test-1, Test-2) the
opening type, size and position, plan layout, building height, height of the case study
unit within the building and the configuration of the surrounding environment were
kept constant.
Figure 6.5. Openings’ configuration and measurement points for cross ventilation (Test1-left), and
single-sided ventilation (Test2- right).
6.2.4 Meteorological data
Outside weather properties were required in this study for two main purposes:
(1) to be used in analysis as the reference weather data, and (2) to consider their effect
on indoor thermal conditions. Since installation of a local on-site weather station was
not possible in this project, meteorological data from the nearest weather station
(Brisbane Station), collected and analysed by the Australian Bureau of Meteorology
(Australian Government Bureau of Meteorology, 2016), was used as the reference
122 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
external weather conditions. With the elevation of 8.13m, the Brisbane weather station
is located at latitude -27.48 °S, longitude of 153.04 °E, about two kilometres from the
case study. Data included wind speed, wind direction, air temperature, and relative
humidity reported at one-minute intervals.
Wind speed data from the weather station was adjusted to compensate for the
above ground height difference between the weather station (8.13m) and the
measurement points (19.2m). Equation 1 below was used to perform this adjustment
(Davenport, 1960):
𝑽𝒛 = 𝑽𝒓𝒆𝒇(𝒁
𝒁𝒓𝒆𝒇)𝜶 (6.1)
Where Vz (m/s) is the reference velocity at the height of the sensors (Z=19.2 m),
Vref is the wind speed recorded at the Brisbane station at the weather station height
(Zref=8.13 m) and α is the terrain roughness exponent which was set to 0.35
representing city centres. Calculated reference wind speed at the experimental
measurement height is used for data analysis presented in the “Results and discussion”
section.
6.2.5 Thermal comfort models
In last few decades, a number of comfort models and indices such as PMV,
SET*, and adaptive comfort models have been developed for prediction of thermal
comfort conditions within an environment. Fanger’s PMV model (Poul O Fanger,
1970) is perhaps one of the first models developed which is a six scale unit index
indicating a human’s thermal sensation. It ranges from -3 to +3 where -3 refers to cold,
0 is neutral, and +3 indicates a hot sensation. The PMV model accounts for the
combined effect of temperature, humidity, air velocity, metabolic rate and clothing
insulation on occupants’ thermal sensation. Predicted Percentage of Dissatisfaction
(PPD) can also be calculated using the PMV value. ASHRAE Standard-55 (ASHRAE,
2013) considers an environment within the comfort range when at least 80% of the
occupants are satisfied with the thermal conditions of that environment. Comfort zone
using the PMV model, therefore, can be defined by -0.5<PMV<0.5, and 20%>PPD.
The well-known PMV model, however, is proven to underpredict thermal sensation of
occupants in naturally ventilated buildings (Croome et al., 1993; R. De Dear & Brager,
1998). De Dear and Brager (R. De Dear & Brager, 1998) explain that not being fully
accountable for the adaptive behaviour of naturally ventilated building’s occupants is
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 123
the main reason for this underestimation of thermal sensation votes by PMV model.
Given this, Fanger and Toftum (P Ole Fanger & Toftum, 2002) proposed the extended
PMV model that can predict thermal comfort of non-air-conditioned buildings more
precisely. The extended PMV model considers the adaptive behaviour of occupants by
adding a reduction of metabolic rate and an expectancy factor (e) to the original PMV
model. According to Fanger and Toftum (P Ole Fanger & Toftum, 2002) people are
likely to reduce their activity level under warm conditions to adapt their metabolic rate
in response to their environment’s thermal condition. This activity level reduction is
6.7% for every unit above the neutral point in the PMV model (P Ole Fanger &
Toftum, 2002). The psychological adaptation was also taken into consideration by
multiplication of the expectancy factor to the PMV model. The assumption here is that
the thermal comfort expectation of free-running buildings’ occupants is lower than the
expectation of people who are used to air-conditioners. The expectancy factor can vary
between 0.5 and 1 and is defined based on the period of hot weather and dominancy
of building type in terms of cooling system (i.e. air-conditioned or free-running) (P
Ole Fanger & Toftum, 2002).
In addition to the extended PMV model, the SET* index also accounts for
thermal comfort prediction in naturally ventilated buildings. SET* is ASHRAE’s
(ASHRAE, 2013) recommended model for prediction of thermal comfort conditions
for cases with airspeeds greater than 0.2 m/s. Similar to the PMV model, SET*
accounts for the combined effect of temperature, humidity, airspeed, activity level, and
clothing insulation while the output is a temperature that humans perceive rather than
thermal sensation vote (Gagge et al., 1986).
Since the extended PMV and SET* are both suitable for naturally ventilated
buildings and they introduce different qualities as an outcome, both models were used
in this study to allow a comprehensive interpretation of thermal conditions under
different ventilation modes.
In the current study, PMV and SET* values were calculated using the WinComf
tool (ME Fountain & Huizenga, 1996). The following assumptions are made for the
calculations. Considering the subject building is residential, sedentary activity level
was assumed, and metabolic rate was set to 1.2 met (ASHRAE, 2013). With regards
to clothing insulation, light typical summer clothing with a value of 0.5 clo was used.
124 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
In the extended PMV calculations an expectancy factor of 0.9 was adopted for
Brisbane (P Ole Fanger & Toftum, 2002). Standard PMV values were first calculated
using the experimental data. Metabolic rate reduction then was applied to these values.
Finally, the extended PMV was obtained by multiplication of expectancy factor to the
recalculated PMV values.
6.3 RESULTS AND DISCUSSION
A local coordinate system was adopted for easier interpretation of data with the
north-south axis parallel to the case study’s length and perpendicular to the openings.
All the sensors and weather station wind direction data were adjusted accordingly. The
adopted north is tilted 35° from the true north toward the west as shown in Figure 6.6.
North direction in the provided results in this section refers to the adjusted north (𝑁′).
Figure 6.6. Local coordinate system (𝑵′) in relation to the true north
6.3.1 Measurements summary
Summary of measured values at each measurement point, (see Figure 6.5 for
measurement points locations), as well as reference weather station values, are
presented in Table 6.2. It needs to be noted that these values are averaged over each
test's timeframe and do not reflect transient information. Reference outdoor conditions
as a function of time for both tests are also presented in Figure 6.7.
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 125
Table 6.2. Measurement summary
Experiments Parameters
Measurement points
Reference
weather
station
P1 P2 P3 P4 S-P5* C-
P5** P6
Test-1
(cross
ventilation)
U (m/s) 2.51 1.55 ±
0.02
0.87±
0.017
0.72±
0.02
1.16±
0.03 N/A
1.35 ±
0.027
0.35 ±
0.007
D E NNW N N/A N/A N/A N SSW
T (°C) 26.24 28.27 27.24 27.18 27.4 N/A 27.15 27.32
Test-2
(single-sided
ventilation)
U (m/s) 2.99 0.57 ±
0.01
0.11±
0.002
0.11±
0.003
0.05±
0.001
0.23±
0.004 N/A
0.05±
0.001
D SE ENE NNE N/A N/A E N/A SSE
T (°C) 26.26 29.55 28.4 28.24 28.84 28.37 N/A 28.4
*Data at this point was only collected for the Test-2
**Data at this point was only collected for the Test-1
U= Airspeed (m/s)
D=Direction (16 compass point)
T= Temperature (°C)
As Table 6.2 and Figure 6.7 show, although measured on two separate days,
outside weather conditions were very similar for both tests which allowed a fair
comparison of results. Peak temperature reached 31˚C with an average temperature
over the day of about 26˚C for both cases. Average wind speed was between 2.5-3 m/s
and wind speed fluctuations followed a similar trend for both cases with a relatively
low speed during the night (up to 3 m/s) and higher wind speed from morning to the
evening (up to 6 m/s). As can be seen in Table 6.2, the measured average velocity for
the single-sided ventilation (0.18 m/s) is much lower than that of the cross ventilation
(1 m/s). Also, the temperature difference between outside and inside in Test-2 is on
average more than 1˚C higher (higher indoor temperature) compared to Test-1 which
highlights the effect of higher airspeed in cooling the space.
126 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
Figure 6.7. Outdoor weather conditions: temperature (A), relative humidity (B), and wind speed (C)
It can also be seen that there is more than 1˚C temperature difference between
P1 and C-P5 in Test-1, although both measurement points are on the balconies. The
reason for this discrepancy can be explained by the sensors location, building
orientation, and the effect of solar radiation. As presented in Figure 6.8, P1 was
attached to the balustrade of the southern balcony, therefore, there was no ceiling
above it, while, C-P5 was placed in the northern balcony under the balcony ceiling.
Accordingly, C-P5 was mostly shaded during the measurement, while P1 was exposed
to the solar radiation from morning to noon. The sun path on the measurement day
relative to the building location and orientation is presented in Figure 6.9.
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 127
Figure 6.8. Sensors P1 and CP-5 location
Figure 6.9. Sun path on the measurement day relative to the case study building and location.
128 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
6.3.2 Thermal comfort
Results of extended PMV and SET* for both tests are presented in Figure 6.10-
A and Figure 6.10-B respectively. The indoor thermal condition was assumed to be
comfortable when a maximum of 20% of the occupants are dissatisfied (-
0.5<PMV<0.5) (ASHRAE, 2013). Comfort conditions as defined by the extended
PMV are found to exceed the comfort range between 11:30 am to 4:00 pm in both
ventilation modes (Figure 6.7-A). Single-sided ventilation, however, is outside the
comfort range (PMV>0.5) over the whole period of the experiment and it represents
higher PMV values than cross ventilation over the hottest part of the day (11:30am-
4:00 pm). As can be seen in Figure 6.10-A, the cross ventilation configuration is within
the comfort zone for most of the time (more than 70%). Given that in a thermally
comfortable space mechanical cooling is not needed and vice versa, the use of air
conditioning is expected to be about 70% less in the case with cross ventilation. Taking
into account that the tests were conducted on hot summer days, similar conditions can
be assumed for hot months of the year. Therefore, a significant amount of energy could
potentially be saved by application of cross ventilation.
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 129
Figure 6.10. Extended PMV and PPD (A), and SET* (B) values for single-sided and cross ventilation
Results of SET* (Figure 6.10-B) also demonstrate that a constantly lower
perceived temperature in the cross ventilated case throughout the experiments is
achieved. Minimum SET* values for single-sided and cross ventilation are 26.1˚C and
22.2˚C, and the maximum values are 30.7˚C and 28˚C respectively. The SET*
difference (ΔSET*) between different configurations can be referred as cooling
potential on the human body (S. Omrani, V. Garcia-Hansen, B. R. Capra, & R.
Drogemuller, 2017). The average ΔSET* (SET*single-sided-SET*cross) is about 3˚C,
therefore, there is approximately 3˚C potential cooling effect in application of cross
ventilation compared to single-sided ventilation.
Although the outside weather condition was very similar for both tests (as shown
in Table 6.2 and Figure 6.7), there is a significant difference between the indoor
130 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
thermal conditions under single-sided and cross ventilation. Since all the controllable
variables were kept the same, this discrepancy can only be explained by ventilation
mode difference and the effectiveness of natural ventilation under cross ventilation.
Accordingly, a detailed analysis of each ventilation mode and their mechanism under
various wind speeds and directions was carried out in order to identify the effect of
different variables on ventilation performance. Results of these analyses are presented
in the following sections.
6.3.3 Reference wind speed and resulting airflow
Airspeed at building openings can be estimated using meteorological data as a
ratio of the reference wind speed (Omrani et al., 2016a). This ratio can vary based on
the similarity of the reference weather station and a building's urban settings, and
building's height (Omrani et al., 2016a). Furthermore, ventilation mode (single-sided
and cross ventilation) may also affect this ratio. To investigate the potential effect of
natural ventilation mode on the ratio of reference wind speed and the airspeed at
building’s openings, data collected from the external sensor (P1) was plotted against
the reference wind data for both cross ventilation and single-sided ventilation. These
results are shown in Figure 6.11-A, where the linear trend line relating cross ventilation
data to the reference wind is seen to be much steeper than that of the single-sided
ventilation suggesting a higher ratio of reference wind would enter the cross ventilated
case. Furthermore, results indicate this ratio is more than two times higher than that of
the single-sided ventilated case. The opening size was identical in both tests, therefore,
this result suggests that airflow rate in Test-1 is almost twice as much as airflow rate
at the Test-2. The higher airflow rate is mainly due to the greater dynamic pressure
caused by wind in the cross flow ventilation which drives the ventilation and results in
higher airspeed at the openings. A similar relation was also observed between airspeed
of internal measurement points. A linear relation was found between airspeed at every
two measurement points (Figure 6.11-B, C, D). Therefore, airspeed at different points
in a space is a ratio of airspeed at the openings. It needs to be noted that Up# in Figure
6.11 refers to the airspeed captured at different measurement points. As can be seen,
in both tests airspeed decreases as the distance from the openings increases due to the
conversion of dynamic pressure to static pressure.
In conclusion, the airspeed at a buildings’ opening is a ratio of the reference wind
which can be obtained from meteorological stations. For this case study, this ratio is
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 131
about 0.3 and 0.15 for cross ventilation and single-sided ventilation respectively. A
similar relation is also evident between airspeed at the openings and airspeed in the
internal spaces. This ratio varies based on ventilation mode and the distance from the
opening. Thus, this study experimentally demonstrated that the ratio between reference
wind speed and internal airspeed is lower in the case of single-sided ventilation
compared to cross ventilation mode. Furthermore, this ratio decreases as the distance
from the opening increases.
Figure 6.11. Scatter plot of airspeed at P1 and reference wind speed (A), P1 and P2 (B), P2 and P3(C),
and P3 and P4 (D)
6.3.4 Air flow distribution
Indoor air distribution is one of the basic airflow aspects that is not only
important for developing natural ventilation analysis models (Karava et al., 2011), but
132 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
also directly affects the thermal sensation of occupants. Internal airspeed and
distribution varies in different ventilation modes and directly affects the natural
ventilation effectiveness. Therefore, airflow distribution along the case study’s length,
subject to single-sided and cross ventilation, is analysed.
Figure 6.12 represents the mean wind speed ratio of both test cases as a function
of the non-dimensionalised space length (x/D). Mean wind speed ratio is the average
indoor air velocity normalised by the reference velocity at the measurement height
(U/Uref). The sensor at the P1 measurement point installed at 0.2 m distance of the case
study’s façade is taken as the starting point of the chart with all other measurement
locations presented accordingly. Average U/Uref is at its highest just before entering
the building (P1) in both test cases. In the single-sided ventilation, this value keeps
decreasing gradually as the distance from the opening increases eventually reaching
near stagnation values. In cross ventilation, however, it is seen that U/Uref decreases
until halfway of the case study’s length, from which it increases as flow moves towards
the opposite opening where it reaches its initial velocity (U/Uref ~1). The airflow
distribution pattern through the cross ventilated case is similar to that presented in
Karava et al. (Karava et al., 2011). The wind speed ratio values presented in Figure
6.12, however, are higher than that of reported in Karava et al.’s work (Karava et al.,
2011) by a factor of 1.5 on average. This discrepancy can be a result of greater opening
porosity in the current study. The opening porosity in this study is 30%, while the
airflow distribution presented in Karava et al. (Karava et al., 2011) is from a case with
10% opening porosity. The variation bars show that U/Uref in Test-1 changes in a wider
range compared to the Test-2 and this variation increases as U/Uref increases. Despite
the higher deviation, minimum value in the cross ventilation (0.5 m/s) is still higher
than the maximum value in the single-sided ventilation (0.3 m/s). In addition, a much
higher value of average wind speed ratio is evident in the cross ventilation compared
to that of the single-sided ventilation (approximately seven times higher). In cross
ventilation typically, each opening acts as either an inlet or an outlet, therefore, wind
action upon the openings results in positive and negative pressure zones at the
openings that drives the airflow through space. In single-sided ventilation, however,
the single opening acts as both an inlet and an outlet. Being located at one pressure
zone (positive or negative), pressure difference generated as a result of wind is much
lower compared to the case with cross ventilation configuration resulting in
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 133
significantly lower airflow. In fact, wind-driven single-sided ventilation can be mainly
dominated by wind turbulence and penetration of eddies (Jiang et al., 2003).
Figure 6.12. Mean wind speed ratio in single-sided and cross ventilation in relation to the space
length.
6.3.5 Reference wind direction and internal air flow direction
Due to the fluctuating nature of wind, wind speed and direction vary over time.
Changes in wind direction do not necessarily depend on wind speed changes or vice
versa (Park, 2013). Having said that, changes in each parameter influence the air flow
behaviour inside naturally ventilated buildings. Accordingly, the effect of reference
wind direction on internal airflow was investigated.
Figure 6.13 demonstrates the wind direction frequency recorded by the reference
weather station and air direction at the installed sensors for the duration of the tests. It
should be noted that results from the sensors with airspeed below their direction
calculation limit are not presented in these graphs (P2 and P6 in Test-2). In Test-1, the
prevailing direction captured by the internal and external sensors is from the north
(inlet) to south (outlet) and is consistent at all measurement points except for P6. This
sensor was located near the corner of the living area and not in the mainstream flow
path. As such, it is likely to have experienced air recirculation and possible stagnant
zones, thus recording a different flow directional pattern from the prevailing
conditions. Results show that the recorded internal prevailing air direction is constantly
from the north to south, despite the reference wind was mainly blowing from the east
and thus parallel to the openings. This can be explained in the physical mode of
operation of the case study. In cross ventilation there are two openings, each on
134 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
opposite sides. One predominantly acts as an inlet while the other needs to be an outlet.
Therefore, regardless of outdoor prevailing direction, internal airflow direction is
constantly perpendicular to the openings. In Test-2, however, external sensors (P1
attached to the balcony balustrade and S-P5 on the other balcony) exhibit a diverse
direction distribution in almost all directions and no clear relation can be seen between
the reference wind direction and the prevailing direction captured by the sensors. This
confirms the existence of bidirectional airflow. In a building with only one opening,
the opening needs to act as both inlet and outlet, therefore, flow does not move in one
dominant direction. It needs to be noted that factors such as presence of furniture and
internal layout also affect the airflow direction and distribution. In this study, however,
these parameters were the same for both tests, therefore, their effects were not taken
into account. From the observations explained above it can be concluded that the
internal airflow direction is mainly affected by the ventilation mode rather than the
reference wind direction although wind direction remains important as will be
discussed in the following section.
Figure 6.13. Frequency of wind direction at reference weather station and measurement points for
Test-1 (left) and Test-2 (right).
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 135
6.3.6 Wind direction and internal air flow
Wind direction and its effect on internal air velocity can be translated to the
building orientation towards the prevailing wind and its effect on ventilation
effectiveness in building design. Therefore, four main directions of north, east, south
and west were chosen to investigate the effect of the external wind direction on internal
airspeed. The instances when wind was blowing from the four main directions and the
corresponding experimental measurements were extracted from the collected data. The
wind speed ratio corresponding to each reference direction, then was calculated by
dividing the measured air velocity value at each point by the reference wind speed.
These calculations were carried out for the duration of each experiment.
Reference direction and internal air flow in cross ventilation
Average wind speed ratio for each reference direction was calculated by dividing
the average measured values inside the case study by the average wind speed
corresponding to that direction and is presented in Table 6.3. The highest value belongs
to the southerly wind followed by northerly, easterly, and westerly winds. In an ideal
environment with no flow obstruction, the same wind speed ratios for wind
perpendicular to the openings (north and south) could be expected. The same
correlation could also apply to easterly and westerly winds. However, there is a
discrepancy in the obtained values for south and north, as well as the corresponding
values to east and west reference directions. The case study’s environment and
surroundings can explain this discrepancy. As detailed in Section 6.2.2, there is a
relatively tall construction, with an approximate height of 35m, 25m to the north of the
case study while at the southern side, there is no major obstruction up to 320m from
the building. The north neighbouring obstruction is approximately 15 m taller than the
case study unit and as such blocks part of the wind approaching from the north. This
flow obstruction is considered the cause of the difference in the values that correspond
to the southerly and northerly winds. At the western side of the case study, Brisbane
city’s major high-rise buildings form a major obstruction on wind path. Wind blockage
from the eastern side, however, is comparatively less due to the lower height and
density of the neighbouring buildings. These differences in flow blockage are the
primary cause for the lower wind speed ratio corresponding to the western and eastern
winds. Another possible explanation is the building shape. The case study unit is
located at the eastern end of the building and its adjacent unit’s wall is extended 1.4m
136 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
at northern side and 1.2m at southern side further from the balcony’s western walls
(can be seen in Figure 6.4). This building form further acts to redirect the approaching
wind from east to the internal spaces. Therefore, the building's surroundings and its
form influence the local wind penetration into the space, resulting in higher wind speed
ratio in the occurrence of easterly winds compared to the westerly winds. These
findings highlight the importance of building form, projections, and the surrounding
environment.
Table 6.3. Average wind speed ratio corresponding to the four main directions for Test-1.
Direction North East South West
Average U/Uref 0.678±0.013 0.585±0.011 0.812±0.016 0.481±0.01
To investigate the effect of wind direction change on internal airflow in more
detail, mean wind speed ratio at each measurement point corresponding to the external
wind directions was plotted in relation to the dimensionless space length (Figure 6.14).
Accordingly, wind speed ratio is at its highest under incident south reference wind then
it decreases as direction changes to the north, east, and west respectively. In addition,
U/Uref variations according to direction change are very similar at all the measurement
points and changes in the reference direction affect every point of the space to a similar
extent. In other words, direction change affects wind speed ratio value, however, its
impact is independent of the distance from the openings. For example, when
comparing the wind speed ratio changes as a result of direction change from north to
east, there is approximately 16% reduction at 0.3, 0.5 and 0.9 of the case study length.
A similar correlation applies to changes from one reference direction to another. In
another word, trend lines representing measurements results referring to each reference
direction are parallel curves with different starting points. Based on the conclusions
drawn from section 6.3.5, internal airflow would be redirected from inlet to the outlet,
hence, regardless of reference wind direction, it is predominantly normal to the
openings in the case of cross ventilation. Furthermore, according to section 6.3.3
findings, airspeed at each point of a space is a ratio of wind speed at the openings.
Accordingly, the correlation between internal airflow and airspeed at the openings
remains the same for all the reference directions, hence, they will be affected to a
similar extent.
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 137
Figure 6.14. Average wind speed ratio corresponding to the four main directions along the case study
for the cross ventilation test.
Reference direction and internal air flow in single-sided ventilation
The calculated mean wind speed ratios corresponding to the external wind
directions in the single-sided ventilation test are presented in Table 6.4. The highest
wind speed ratio corresponds to the southerly wind, followed by easterly, northerly
and westerly winds respectively. Similar to cross ventilation configuration, the highest
value of wind speed ratio is obtained for the instances when the approaching wind is
perpendicular to the opening of the apartment. This is also consistent with findings of
Aflaki et al. (Aflaki, Mahyuddin, & Baharum, 2016). In contrast, the lowest value
relates to the westerly winds. Under these conditions, the wind is parallel to the
opening and the time-averaged pressure difference between indoor and outdoor is
approximately zero (C. R. Chu, Chen, & Chen, 2011), therefore, the lowest value could
be expected. Wind speed ratios resulting from the easterly wind are slightly higher
compared to that of the west incident wind. This difference can be attributed to the
building shape (extended western balcony wall) and the nearby obstructions as
previously elaborated.
Table 6.4. Average wind speed ratio corresponding to the four main directions for Test-2.
Direction North East South West
Average U/Uref 0.079±0.002 0.082±0.002 0.095±0.003 0.071±0.002
0
0.2
0.4
0.6
0.8
1
-0.1 0.1 0.3 0.5 0.7 0.9 1.1
U/Uref
x/D
East
North
South
West
138 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
A similar trend as the overall average is also evident at different measurement
points throughout the case study’s length as shown in Figure 6.15. This figure shows
larger variations between values of different directions at the opening where the flow
reduces with the distance until becoming almost negligible at the furthest point. In
other words, the effect of direction change on wind speed ratio inside the building
decreases as the distance from the opening increases. In that case, the captured air has
either lost momentum before reaching the end of the unit or changed direction and
exited through the opening.
Figure 6.15. Average wind speed ratio corresponding to the four main directions along the case study
for the single-sided ventilation test.
Reference direction and internal air flow in single-sided and cross ventilation
As discussed in the previous sections, a change of incident wind direction affects
the airspeed inside the building. In this section, the impact of direction change for both
ventilation modes is discussed. Wind speed ratio (U/Uref) value change resulting from
the direction change was calculated and together with the highest and lowest values
associated with each test are shown in Figure 6.16. Although the change in direction
affects the internal airflow, the lowest U/Uref values in cross ventilation (west
direction) are still about twice the highest values in the single-sided ventilation (south
direction).
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
-0.1 0.1 0.3 0.5 0.7 0.9 1.1
U/Uref
x/D
South
East
North
West
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 139
Figure 6.16. Highest and lowest values of average wind ratio with regards to the reference direction
for Test-1 and Test-2
In Test-1, the effect of direction change on internal airspeed is almost equal at
different points between inlet and outlet, whereas in Test-2 the effect of direction
change on internal airspeed decreases as the distance from the opening increases. The
reason behind minimal changes of airspeed influenced by the direction change far from
the opening in single-sided ventilation is due to the loss of momentum in those areas.
Wind speed ratio is at its minimum far from the opening which shows that a small
portion of wind forces travels to the end of the space. Therefore, not only airspeed is
at its minimum amount, but also the effect of wind direction change is the lowest.
The results of this study also yields design implications for the improvement of
natural ventilation design of high-rise buildings. The effect of wind direction on indoor
air velocity can be translated into building orientation in building design. Accordingly,
the ventilation performance can be improved when the building openings are oriented
toward the prevailing wind. An analysis of wind direction revealed that extended walls
and projections can help the wind parallel to the openings to penetrate into internal
spaces by redirecting it. Therefore, implication of wing walls and projections can
improve the ventilation performance for the instances that wind is not normal to the
openings. Finally, a building’s neighbouring and surrounding environment were found
to play an important role in natural ventilation performance of buildings, hence, for
realistic performance expectations, size and approximation of neighbouring
obstruction need to be considered in natural ventilation design.
0
0.2
0.4
0.6
0.8
1
1.2
-0.1 0.4 0.9
U/Uref
x/D
South-Test 1
West-Test 1
South-Test 2
West-Test 2
140 Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement
6.4 CONCLUSION
While it is well understood that cross ventilation is preferable to single-side
ventilation, currently there is no study that quantifies the magnitude of the performance
of the ventilation modes for high-rise residential buildings. This study has
demonstrated that cross ventilation outperforms single-sided ventilation with respect
to achieving a comfortable thermal environment. This was shown through an analysis
of the extended PMV where it was seen that cross ventilation maintains comfortable
thermal conditions 70% of the time, while the single-sided case was consistently
deemed hot (PMV>0.5). In addition, SET* analysis has shown that indoor condition
in single-sided ventilation was on average 3˚C hotter than cross ventilation. The
findings on thermal comfort performance led to a detailed analysis of ventilation
performance of each ventilation mode. The correlation between meteorological data
and internal airflow indicated that indoor airspeed is a ratio of the reference wind,
hence, can be calculated using weather data. This ratio for single-sided ventilation was
found to be nearly half of the cross ventilation. In terms of indoor airflow distribution,
wind speed ratio inside the case study was two to four times higher for the cross
ventilation case indicating a significantly better performance compared to single-sided
ventilation. The relation between reference wind direction and internal airflow
direction indicated that wind direction had minimal effect on indoor dominant air
direction and it was mainly affected by ventilation mode. Additionally, indoor air
velocity was found to be influenced by reference wind direction where highest airspeed
was associated with the wind normal to the openings while it was at its minimum when
the wind was parallel to the openings.
It needs to be mentioned that the design considerations can be more crucial when
designing for single-sided ventilation since the air flow rate is already low and
unfavourable orientation and design features may result in insufficient ventilation.
6.5 LIMITATIONS AND FUTURE WORK
All the measurements in this study were conducted at height 1.2m from the floor.
Given that sitting is perhaps the main posture in a living area, measurements of
variables at the head level of a sitting occupant is deemed satisfactory for this study.
For future studies, however, measurements at different heights can be useful in the
determination of thermal conditions for sleeping and standing occupants. Furthermore,
Chapter 6: Effect of natural ventilation mode on thermal comfort and ventilation performance: Full-scale
measurement 141
the correlation between meteorological data and air velocity at different points of the
case study was only measured for one apartment. It is expected that similar
experiments with the subject case studies of various heights at different locations in
relation to the reference weather stations can lead to empirical ratios for expected
airspeed in buildings. These ratios could allow for prediction of airspeed and
ventilation performance at building openings and internal spaces of a high-rise
building using meteorological data.
It should be noted that the results of this study are based on the data captured in
a high-rise residential unit with rectangular floor layout. The results and conclusions
drawn from this study, therefore, may not be applicable to the buildings with
significantly different layout and/or opening configurations. This also applies to the
climates that greatly differ from Brisbane’s climate.
Acknowledgements
This research did not receive any specific grant from funding agencies in the
public, commercial, or not-for-profit sectors.
Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential building
143
Chapter 7: Thermal comfort evaluation of
natural ventilation mode: case
study of a high-rise residential
building
Omrani, S., Garcia-Hansen, V., Drogemuller, R., & Capra, B. (2016). Thermal
comfort evaluation of natural ventilation mode: case study of a high-rise residential
building. 50th International Conference of the Architectural Science Association 2016,
Adelaide, Australia.
https://eprints.qut.edu.au/103494/
Statement of contribution of co-authors for thesis by published paper
The authors listed above have certified that:
1. they meet the criteria for authorship in that they have participated in the
conception, execution, or interpretation of (at least) that part of the
publication that lies within their field of expertise;
2. they take public responsibility for their part of the publication, while the
responsible author accepts overall responsibility for the publication;
3. there are no other authors of the publication;
4. potential conflicts of interest have been disclosed to (a) granting bodies, (b)
the editor or publisher of journals or other publications, and (c) the head of
the responsible academic unit; and
5. Consistent with any limitations set by publisher requirements, they agree to
the use of the publication in the student’s thesis, and its publication on the
QUT ePrints database.
144 Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential
building
The authors’ specific contributions are detailed below:
Contributor Statement of contribution
Sara Omrani Collected the experimental data, analysed
the data, produced the graphics, developed
the study, and wrote the manuscript.
Veronica Garcia-Hansen Assisted in developing the study, and
reviewed the manuscript.
Robin Drogemuller Assisted in developing the study, proof-
read and reviewed the manuscript.
Bianca Capra Assisted in developing the study.
Principal Supervisor Confirmation
I have sighted emails or other correspondence from all co-authors confirming their
certifying authorship.
__Veronica Garcia Hansen___ ____________________________28/04/2017_____
Name Signature Date
Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential building
145
Abstract
Natural ventilation can be used as a low-cost alternative to mechanical
ventilation. Bearing in mind that ventilation mode plays an important role in natural
ventilation performance, the current study investigates the effectiveness of two major
natural ventilation modes (i.e. single-sided and cross ventilation) in providing thermal
comfort for occupants of high-rise residential buildings in cooling dominant climates.
Measurements of air velocity, temperature and relative humidity were carried out in a
unit located in a high-rise residential building in Brisbane, Australia. Both single-sided
and cross ventilation settings were examined in two consecutive days in summer. The
extended Predicted Mean Vote (PMV) and Predicted Percentage of Dissatisfaction
(PPD) were calculated and results showed a considerably better performance of cross
ventilation over single-sided ventilation. Cross ventilation could provide thermal
comfort in a typical hot summer day for most of the day (greater than 70% of the time),
while, for single-sided ventilation the thermal conditions of internal spaces was
comfortable for only 1% of the time.
Keywords: Natural ventilation; ventilation mode; thermal comfort; high-rise
residential.
7.1 INTRODUCTION
In cooling dominant climates, weather conditions mostly lie outside the comfort
range, especially during summer. Therefore, air-conditioners are widely used for space
cooling and providing a thermally comfortable environment. Air-conditioners are
energy intensive and consume a large portion of the energy delivered to buildings
(Pérez-Lombard et al., 2008). Natural ventilation as a passive cooling strategy, on the
other hand, is a low-cost alternative to air conditioners. Natural ventilation not only
contributes to thermal comfort but also can improve indoor air quality.
There are a number of parameters that affect natural ventilation performance and
can be addressed through building design such as building orientation, shape and size
of openings, and ventilation mode. Among these design related parameters, ventilation
mode has the most impact on ventilation performance (Fung & Lee, 2014).
There are two major ventilation modes namely: single-sided ventilation and
cross ventilation (Jiang & Chen, 2001). In single-sided ventilation, air enters and exits
from openings at one side of the space while in cross ventilation, air flow enters and
146 Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential
building
leaves through separate openings at different sides of the space (M. W. Liddament,
1996). Air movement in single-sided ventilation is mainly due to temperature
difference between inside and outside and the consequent buoyancy forces and
pressure difference (P. F. Linden, 1999). In cross ventilated spaces, on the other hand,
the pressure difference produced by the wind at inlet and outlet is the main driving
force (M. W. Liddament, 1996). As far as pressure difference goes, wind produces a
much larger force compared to buoyancy and temperature difference. Therefore, a
space with cross ventilation normally experiences a higher airspeed and ventilation
rate (Evola & Popov, 2006).
Although cross ventilation performs better than single-sided ventilation, it is not
always possible to design buildings with cross ventilation. Sometimes site restrictions
dictate single-sided ventilation as the only possible option especially in high-rise
buildings in dense urban areas. Despite the importance of this subject matter,
effectiveness of ventilation modes in providing a thermally comfortable environment
is yet to be thoroughly investigated.
The current study investigates the effectiveness of the two major ventilation
modes (single-sided and cross ventilation) in providing thermal comfort for a high-rise
residential building in a cooling dominant climate. Air velocity, temperature and
Relative Humidity (RH) data were collected for two hot summer days in a residential
apartment in a high-rise building located in Brisbane, Australia. The collected data
were used in calculating a thermal comfort index applicable to naturally ventilated
buildings. Finally, thermal conditions inside the case study for both cases of cross
ventilation and single-sided ventilation were evaluated and compared.
7.1.1 Climate condition of Brisbane
Brisbane is located in 27.4° S latitude and 153° E longitude. Brisbane’s climate
is subtropical with warm and humid summers and mild to cool winters. Monthly
minimum and maximum mean temperature ranges from 10°C in July to 30°C in
January and mean relative humidity is relatively high most of the time, laying in the
range of 50% to 70% on average. The annual mean wind speed is 3.6 m/s and is
predominantly blowing from south and south-west in the mornings and from east and
north-east in the afternoons (Australian Government Bureau of Meteorology, 2016).
The graph below shows mean monthly temperature and wind speed in Brisbane.
Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential building
147
Figure 7.1: Brisbane’s mean monthly temperature and wind speed
7.2 METHODOLOGY
This study investigates the effectiveness of single-sided and cross ventilation in
proving thermal comfort for building occupants using full-scale on-site measurements.
Air velocity, temperature and RH were measured in a high-rise residential apartment
for both single-sided and cross ventilation. The collected data was used for thermal
comfort evaluation by adopting extended PMV (Predicted Mean Vote) and PPD
(Predicted Percentage of Dissatisfaction) as criteria.
7.2.1 Full-scale measurements
Data collection for the current study was carried out in a residential unit located
at level five of a 36-storey residential building situated in Brisbane, Australia. The case
study’s layout with two balconies at two opposite sides of the living area allowed
measurements for both single-sided and cross ventilation. Both balcony doors were
kept fully open (1.16m*2.5m=2.9 m2 operable area each) for the cross ventilation
setting. For the single-sided ventilation setting, the northern balcony door was shut and
the southern door was kept fully open for the duration of the experiment. Figure 7.2
represents the location of the case study within the whole building (left) and the
measurement point on the case study’s plan (right). As can be seen, the case study is a
two-bedroom apartment; however, all the measurements were only carried out in the
148 Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential
building
living area. Therefore, doors and windows to the bedrooms were kept closed for the
duration of the data collection.
The data collection was conducted in summer (January 13th and 14th) to examine
the possible worst case scenario. In Brisbane, January is the hottest month of year
(Figure 7.1) and the most critical time in terms of cooling energy requirements.
Therefore, if a naturally conditioned building is thermally comfortable in the hottest
time of year, it may not need mechanical cooling for the rest of the year.
Temperature, RH and air velocity were measured inside the living area of the
case study (Figure 7.2) for single-sided and cross ventilation during 24 hours for each
setting. Considering the fluctuating nature of wind, temperature change and solar
radiation pattern, 24 hours might be long enough to cover typical weather condition
variations. Measurements were carried out on days with clear sky when no
precipitation occurred.
Instrumentations that were used in the data collection included a velocity
transducer (8475 series, TSI), and temperature and RH sensors (iBotton, Maxim
integrated). The velocity transducer logged air speed at a sampling rate of 5 Hz,
temperature and RH data were recorded at one-minute intervals. All the sensors were
installed at a height of 1.2m which represents the head level of a sitting occupant.
Figure 7.2. Case study location within the building (left) and plan and measurement point (right)
Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential building
149
7.2.2 Evaluation criteria
One of the main purposes of natural ventilation is to provide occupants with a
thermally comfortable environment. To this end, thermal comfort was chosen as the
criteria for assessment of ventilation modes. Hence, an appropriate comfort model
needed to be adopted for this study. In the last few decades, a number of comfort
models have been developed with the aim of predicting an environment’s thermal
condition for its occupants.
One of the first comfort models was the PMV developed by Fanger (Poul O
Fanger, 1970). PMV is an index for human body thermal sensation and ranges from -
3 to +3 where -3 refers to cold, 0 shows neutrality and +3 indicates hot sensation of
the environment. ASHRAE standard (ASHRAE, 2013) considers an environment
thermally comfortable when at least 80% of its occupants are satisfied with the thermal
condition of their environment which can be translated to -0.5<PMV>0.5. Parameters
such as air temperature, radiant temperature, air velocity, RH, metabolic rate and
clothing are taken into consideration in PMV calculations. PPD can also be calculated
based on PMV. The PMV model is proven to underestimate thermal comfort for
naturally ventilated buildings (Croome et al., 1993). De Dear and Brager (1998)
explain this shortcoming with regards to the steady-state assumption of thermal
comfort in the PMV model, as well as neglecting physiological (acclimatisation),
psychological and, behavioural effects. The adaptive comfort model, therefore, was
developed by De Dear and Brager (1998) based on an extensive field study to predict
thermal comfort in naturally ventilated buildings. The adaptive model represents the
acceptable limits of indoor operative temperature as a function of mean outdoor
temperature. Although considered in the model development process, there is no direct
input for air velocity in the adaptive comfort model. Therefore, it was not a suitable
model for the current study. Subsequently, Fanger and Toftum (2002) introduced the
extended PMV model by adding two correction factors to the traditional PMV model.
One is expectancy factor (e) which should be multiplied by the traditional PMV. The
expectancy factor considers thermal expectation of occupants based on their
experience and varies between 0.5 and 1. The other parameter considered in the
extended PMV model is the activity level. People tend to reduce their activity level
unconsciously when feeling warm. This reduction is 6.7% by every scale unit increase
in PMV index above the neutral point. Therefore, for PMV values above zero, a new
150 Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential
building
metabolic rate needs to be obtained and considered in recalculation of the traditional
PMV. Accordingly, PPD can be calculated based on the obtained extended PMV
value. The extended PMV model could predict thermal sensation votes for free-
running buildings in warm climates reasonably well (P Ole Fanger & Toftum, 2002).
The extended PMV model, therefore, was chosen for thermal comfort evaluation in
the current study.
The source code of the CBE thermal comfort tool (Hoyt, Schiavon, Piccioli,
Moon, & Steinfeld, 2013) provided by the developers were used for calculating PMV
using the R statistical software (Team, 2014). The expectancy factor and adjusted
activity level were then applied to the obtained PMV values and extended PMV was
calculated.
To assess thermal comfort performance of single-sided and cross ventilation
using the extended PMV model some assumptions needed to be made. Occupants were
assumed to be involved in sedentary activities. Metabolic rate therefore, was set to 1.2
met. Considering measurements were carried out in summer, typical light clothing
insulation value equal to 0.5 clo was taken for PMV calculations. The expectancy
factor for Brisbane was set to 0.9 based on Fanger and Toftum’s (2002) suggestion.
Activity level reduction was also taken into consideration.
7.3 RESULTS AND DISCUSSION
7.3.1 Cross ventilation
The experimental measurements for cross ventilation setting were carried out on
January 13th for 24 hours. A summary of external weather conditions and measured
values are presented in Table 7.1. A narrower temperature range is evident inside the
case study compared to the external weather temperature while the internal average
temperature is slightly higher yet very close to the external weather mean temperature
(∆Tmean=0.6).
Table 7.1: Weather condition and measured values summary for the cross ventilation setting.
External weather condition Internal measured values
Mean Maximum Minimum Mean Maximum Minimum
Temperature (°C) 26.25 31 20.3 26.85 29.4 25.1
RH (%) 65.8 92 47 63.5 72 54.5
Wind speed (m/s) 1.8 5 0 0.64 2 0
Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential building
151
The extended PMV values and corresponding PPD for the experiment duration
are plotted against the time of day in Figure 7.3. The lowest and highest values for
PMV are -0.64 and 0.98 respectively. Average PMV and PPD are 0.23 and 8.9%
correspondingly demonstrating a predominantly comfortable environment for the
cross ventilation setting. PMV exceeds ASHRAE upper limit (0.5) for 28% of the
experiment time and it is mainly from around 11:30 am to 4 pm when the outside
temperature is high.
Figure 7.3. Extended PMV and PPD results for the cross ventilation setting
7.3.2 Single-sided ventilation
Physical measurements were conducted on January 14th in the same case study
building with single-sided ventilation setting. All the opening conditions were kept the
same as cross ventilation setting except that the northern balcony door was fully closed
during the measurements. Outside weather and internal conditions presented in Table
7.2 show higher temperatures inside the case study with average value difference of
about 2 °C (∆Tmean=2.02). In addition, internal temperature changes in a relatively
limited range compared to the outside temperature variations.
Table 7.2. Weather condition and measured values summary for the single-sided ventilation setting
External weather condition Internal measured values
Mean Maximum Minimum Mean Maximum Minimum Temperature (°C)
26.28 31.7 21 28.3 30.2 26.1
RH (%) 66 84 46 62.2 68.3 54.5 Wind speed (m/s)
2.14 7 0 0.1 0.5 0
152 Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential
building
PMV and PPD were calculated and rendered in Figure 7.4. Average PMV of 1
and average PPD of 28% highlight a dominant warm internal thermal condition. PMV
results also confirm an uncomfortable internal condition for the single-sided
ventilation setting as PMV exceeds the 0.5 limit for 99% of a time. PMV reaches its
highest range (1.2-1.6) from around 11 am to 4:30 pm which can be due to high
external temperature and solar radiation.
Figure 7.4. Extended PMV and PPD results for the single-sided ventilation setting
7.3.3 Discussion
The experimental measurements for single-sided and cross ventilation cases
were carried out on two consecutive days in summer under relatively similar weather
conditions to allow fair comparison of ventilation mode performance and its effect on
thermal comfort in a hot summer day when cooling is needed most. All the influential
and controllable variables such as size of the openings, sensors height and location
were kept the same in both measurement settings. Results reported in section 7.3.1 and
7.3.2 revealed a significant difference between single-sided and cross ventilation
performance in terms of thermal comfort. PMV values from both settings are also
displayed in Figure 7.5 for better interpretation and comparison between the two cases.
Single-sided ventilation failed to provide thermal comfort in a hot summer day
since PMV value was within the comfort zone for only 1% of time. On the other hand,
cross ventilation could provide a comfortable thermal environment for more than 70%
of time. Average PMV values for single-sided ventilation was more than four times
higher than that of the cross ventilated case. The difference between these two
ventilation modes becomes even more apparent when considering that in the cross
Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential building
153
ventilation setting, the PMV values were under the lower limit of thermal comfort (-
0.5) representing cool thermal sensation for about 1% of time which happened around
midnight. Given that occupants have control on the openings, the cool sensation that
would result from high airspeed can be eliminated by the occupants in such instances.
Looking at Figure 7.5, both cases have experienced their highest PMV range
from around noon to 4:30 pm which should be related to temperature rise as a result
of solar radiation. In addition, both graphs follow a consistent trend while more
fluctuations of PMV values are evident in the cross ventilation graph. This can be
explained by the fluctuating nature of wind and the fact that the cross ventilation case
has experienced higher indoor airspeeds.
Figure 7.5. Extended PMV results for the single-sided ventilation setting
In summary, cross ventilation performed considerably better than single-sided
ventilation in terms of thermal comfort as could be expected. However, the major
result is the significant difference which puts the two ventilation modes almost at two
ends of the spectrum. While single-sided ventilation totally failed in providing thermal
comfort, cross ventilation offered desirable thermal conditions for more than 70% of
time. Considering all the influential parameters except for ventilation mode were
similar in both cases, this extreme difference can be explained by natural ventilation
driving forces in each case.
The potential reduction in air conditioning equipment cost versus the possible
increased cost of designing for cross ventilation needs to be studied.
154 Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential
building
7.4 CONCLUSION
This study evaluated the performance of two major ventilation modes, namely
single-sided and cross ventilation, in providing thermal comfort for occupants of a
high-rise residential building situated in Brisbane, Australia. Full-scale measurements
of airspeed, temperature and RH were carried out in a residential unit of the building.
Measurements were conducted in summer to allow assessment for the expected worst
case scenarios. Two experimental arrangements of single-sided and cross ventilation
were examined during two consecutive days in the same case study unit. Extended
PMV and PPD were adopted as thermal comfort assessment criteria. It was found that
cross ventilation could provide thermal comfort for more than 70% of the day while in
the case with single-sided ventilation thermal comfort was achieved for only 1% of
time. This suggests that in case of applying cross ventilation the need for air
conditioning for space cooling can be reduced significantly.
It needs to be noted that this study was conducted at a case study unit at fifth
floor. Considering that wind magnitude increases with the increase in height, higher
airspeeds can be expected at upper floors and vice versa. Therefore, higher floors could
potentially experience acceptable thermal conditions for longer periods compared to
the tested case study. Finally, regardless of building’s height, natural cross ventilation
is a much more effective solution than single-sided ventilation in providing thermal
comfort.
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 155
Chapter 8: On the effect of provision of
balconies on natural ventilation
and thermal comfort in high-rise
residential buildings
Omrani, S., Garcia-Hansen, V., Drogemuller, R., & Capra, B. R. (2017). On the effect
of provision of balconies on natural ventilation and thermal comfort of high-rise
residential buildings. Building and Environment
doi: http://dx.doi.org/10.1016/j.buildenv.2017.07.016
Statement of contribution of co-authors for thesis by published paper
The authors listed above have certified that:
1. they meet the criteria for authorship in that they have participated in the
conception, execution, or interpretation of (at least) that part of the
publication that lies within their field of expertise;
2. they take public responsibility for their part of the publication, while the
responsible author accepts overall responsibility for the publication;
3. there are no other authors of the publication;
4. potential conflicts of interest have been disclosed to (a) granting bodies, (b)
the editor or publisher of journals or other publications, and (c) the head of
the responsible academic unit; and
5. Consistent with any limitations set by publisher requirements, they agree to
the use of the publication in the student’s thesis, and its publication on the
QUT ePrints database.
The authors’ specific contributions are detailed below:
Contributor Statement of contribution
Sara Omrani Collected the experimental data, carried
out CFD simulations, analysed the data,
conducted literature review, produced the
156 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
graphics, developed the study, and wrote
the manuscript.
Veronica Garcia-Hansen Assisted in developing the study, and
reviewed the manuscript.
Bianca Capra Assisted in developing the study and
analysis, proof-read and reviewed the
manuscript.
Robin Drogemuller Assisted in developing the study.
Principal Supervisor Confirmation
I have sighted emails or other correspondence from all co-authors confirming their
certifying authorship.
__Veronica Garcia Hansen___ _QUT Verified Signature_____28/04/2017_____
Name Signature Date
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 157
Abstract
Natural ventilation and balconies are two of the most desirable features of a
living space in subtropical climates. The aim of this paper is to investigate the effect
of balconies on natural ventilation performance and thermal comfort of residential
buildings. To this end, in-situ full-scale measurements were carried out for
Computational Fluid Dynamics (CFD) model validation and further analysis. A
number of parameters such as balcony type, balcony depth, ventilation mode, and wind
angle were used in developing case studies. Once validated, the CFD model was used
for investigation of air movement inside each case study. Combined and separate
effects of the defined parameters on natural ventilation performance were evaluated
using air velocity and Standard Effective Temperature (SET*) as criteria. The results
indicate that the addition of a balcony to a building with single-sided ventilation can
improve the ventilation performance. In contrast, indoor air velocity was reduced as a
result of balcony addition when the case study was operated in cross ventilation mode.
Furthermore, ventilation performance of single-sided ventilation was found to be more
sensitive to the change of parameters compared to that of the cross ventilation. It has
also been found that among the investigated parameters, incident wind angle affects
the ventilation performance most for both natural ventilation modes.
Keywords: Natural ventilation; CFD; balcony; thermal comfort; single-sided
ventilation; cross ventilation
8.1 INTRODUCTION
Natural ventilation is proven to be an effective low-cost solution for space
conditioning, especially in cooling dominant climates (Liping & Hien, 2007; Matheos
Santamouris & Allard, 1998). Being a passive solution, building energy consumption
and associated negative environmental effects can be reduced by implementation of
natural ventilation. Furthermore, building occupants in subtropical climates have a
tendency to live in naturally ventilated buildings rather than fully air-conditioned
spaces (R. Kennedy et al., 2015).
In addition to the external weather conditions as the main driving force,
architectural design features play an important role in natural ventilation performance
and indoor airflow behaviour. Design parameters that alter the internal airflow include
type, size and placement of the openings, internal layout, height and orientation of a
158 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
building, and façade features such as balconies (Aflaki et al., 2015; C.-R. Chu &
Chiang, 2013; CF Gao & Lee, 2011a; Mak et al., 2007; Omrani et al., 2015).
Private outdoor spaces such as balconies are perceived as one the most desired
features in subtropical climates that can be used for a different range of activities (Buys
et al., 2008; R. Kennedy et al., 2015; R. J. Kennedy & Buys, 2010). Balconies act as
buffer spaces between indoor and outdoor that not only reduce the occupants’ exposure
to the pollutions (Niu, 2004) but also result in significant heating and cooling load
reduction (Song & Choi, 2012). In addition, balconies can reduce noise level –the
commonly stated limitation of natural ventilation- by acting as an acoustic protection
device (Mohsen & Oldham, 1977).
From a natural ventilation point of view, the addition of a balcony alters the
pressure distribution on a building façade and consequently affects the ventilative
forces (Chand et al., 1998). Chand et al. (Chand et al., 1998) carried out a wind tunnel
experiment on a five-storey building with mounted balconies to study this impact.
Their results demonstrated an alteration in pressure distribution on the windward side
and no significant change on the leeward side. While Chand et al’s study focused on
pressure distribution on the façade of a case model without openings, their
experimental data was later used for CFD validation and subsequent evaluation of the
effect of balcony provision on indoor ventilation performance (Z. Ai, Mak, Niu, Li, et
al., 2011), and thermal comfort (Z. Ai, Mak, Niu, & Li, 2011). The results indicated
that mass flow rate increases and average velocity decreases in the case of single-sided
ventilation, while no significant change was observed under cross ventilation mode (Z.
Ai, Mak, Niu, Li, et al., 2011). Thermal comfort status was also reported with no
change (Z. Ai, Mak, Niu, & Li, 2011). Prianto and Depecker (E Prianto & Depecker,
2002; E. Prianto & Depecker, 2003) adopted a numerical method to investigate the
effect of balcony, internal divisions, and openings on indoor velocity and thermal
comfort in a two-storey dwelling. They found that both balconies and openings play
an important role in the modification of indoor velocity and thermal comfort condition.
While these studies have been concerned with the effect of balconies on natural
ventilation, they were all based on simple geometries, and the combined effect of
balcony features (i.e. balcony type and depth) with other determinant parameters such
as ventilation mode and incident wind direction are not adequately investigated. The
objective of this study, therefore, is to investigate the impact of these parameters on
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 159
natural ventilation and indoor thermal conditions. Accordingly, full-scale
measurements were carried out in a residential unit located in a high-rise residential
building in Brisbane, Australia. The collected data was then used for validation of a
CFD model and good agreement between the data and the simulation results were
obtained. Two ventilation modes (single-sided and cross ventilation), two balcony
types (semi-enclosed and open balcony), four balcony depths (10%, 20%, 30%, and
40%), and four wind directions (0˚, 45˚, 90˚, and 180˚) were defined as variables. From
that 70 case studies were formulated to investigate the separate and combined effect
of these variables. The validated CFD model was then used for calculation of air
velocity in the case studies.
Average velocity was used as a criterion to evaluate the effect of the variables
on overall ventilation performance. Average velocity is linearly correlated with
qualities such as airflow rate and air change per hour (Lo & Novoselac, 2012), and is
also a determinant in thermal comfort calculations. Therefore, average velocity can be
used as a good indicator of ventilation performance. Acquired average velocity along
with typical meteorological data for Brisbane were further used in calculations of
SET* index for thermal comfort evaluation of the occupied zone.
8.2 METHOD OF ANALYSIS
8.2.1 Field measurement
Field measurements were carried out in a unit located on the fifth floor of a 36-
storey residential building located at Brisbane Central Business District (CBD),
Australia. The building is oriented 35° toward the west and the case study unit is
located at the eastern end of the building. Figure 8.1 shows the case study building and
its surroundings where the case study building is indicated in red boundary.
160 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
Figure 8.1. Case study building (right) and case study surroundings (left). The case study building and
the case study unit are indicated with red boundary.
The case study layout consists of two balconies at two opposite sides of the living
area which allowed measurements for both single-sided and cross ventilation
configurations (Figure 8.2). Balcony doors were identical with the operable area of
1.16m x 2.5m=2.9m2.
Figure 8.2. Case study building plan layout (Omrani, Garcia-Hansen, Drogemuller, & Capra,
2016b).
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 161
Air velocity was measured at six different locations within the living area and
balconies using a 3D anemometer (WindMaster ultrasonic anemometer, Gill
instrument), two air velocity transducers (8475 series, TSI), and three 2D anemometers
(WindSonic ultrasonic anemometer, Gill instrument). Sensors where not re-calibrated
in-situ as factory calibration was considered acceptable for this study. Specifications
of each sensor are given in Table 8.1. Sampling rates of 1-5 Hz was used (Table 8.1),
with collected data time averaged over 1-minute intervals for the detailed analysis.
Sensors were installed 1.2m above the floor level representing the height of a seated
human head. Two sets of measurements for single-sided and cross ventilation were
carried out for 24 hours for each configuration during summer (13th and 14th January
2016). Measurements were recorded on both the balconies and in the living area. Doors
to both the bedrooms and bathrooms were kept shut during all measurements to
minimise the impact of these on the ventilation in the main living area. Figure 8.3
shows the positioning of all sensors for each configuration mode.
Figure 8.3. Case study plan and sampling location for cross ventilation (left) and single-sided
ventilation (right).
Meteorological data from the Australian Government Bureau of Meteorology
(Australian Government Bureau of Meteorology, 2016) from the closest weather
station to the case study building (Brisbane station) was used as reference weather
162 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
conditions. The Brisbane weather station height is 8.13m and is located approximately
two kilometres from the case study building in an area with a similar urban setting to
the case study. The obtained weather data included wind speed and direction,
temperature and relative humidity that were further used in the simulation validation
and thermal comfort calculations.
Table 8.1. Sensors' specifications.
Instrument (manufacturer) NO. Parameters Accuracy and
resolution
Sampling rate
3D anemometer
(WindMaster ultrasonic
anemometer, model
number 1590-PK-020, Gill
instruments)
1 U,V,W vectors
Speed: <1.5%
RMS @12 m/s
Direction: 2°
@12m/s
1Hz
2D anemometer
(WindSonic ultrasonic
anemometer, Option 1,
Serial Numbers 15170156
and 15170157, Gill
instruments)
2 Wind speed and
2D direction or U
and V vectors
Speed: 2%
@12m/s
Direction: 3° @12
m/s
4Hz
2D anemometer
(WindSonic ultrasonic
anemometer, Option 4,
Serial Number13390047,
Gill instruments)
1 Wind speed and
2D direction or U
and V vectors
Speed: 2%
@12m/s
Direction: 3° @12
m/s
4Hz
Velocity Transducer
(8475-075-1, and 8475-
150-1, TSI)
2 Air velocity 3% of reading
from 20° to 26° C.
1% of selected full
scale range (2.5
m/s)
5Hz
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 163
8.2.2 Numerical method
CFD model and settings
CFD has been extensively applied for simulation of natural ventilation
simulation in buildings (Chen, 2009; Omrani et al., 2014; S. Omrani, V. Garcia-
Hansen, B. Capra, et al., 2017; Ramponi & Blocken, 2012). The current study applied
the 3D steady-state Reynold-Averaged Navier-Stokes (RANS) model. RANS
calculates the flow related parameters by solving time-averaged governing equations.
Despite some deficiencies (Hu et al., 2008; Lakehal & Rodi, 1997), RANS models are
proven to be capable of simulating natural ventilation reasonably well for both simple
geometries (Evola & Popov, 2006; J. Perén, T. Van Hooff, B. Leite, & B. Blocken,
2015; Ramponi & Blocken, 2012) and buildings with detailed façade elements such as
balconies and double skin facades (Montazeri & Blocken, 2013; Pasut & De Carli,
2012). Among the RANS models available, renormalisation group (RNG) k-ε
performs better for ventilation simulations compared to the others (Chen, 2009). The
RNG κ-ε turbulence model has been successfully used for simulation of both indoor
and outdoor airflows (Z. T. Ai & Mak, 2014; Bangalee, Lin, & Miau, 2012; Evola &
Popov, 2006; Jin et al., 2015) and has therefore been used in the present study. The
RNG κ-ε model is similar to the standard κ-ε model with a number of additional
refinements that makes it more reliable for a different range of flows (Fluent). A
comprehensive description of RNG κ-ε model can be found in (Fluent; Orszag et al.,
1993). Enhanced wall treatment was also implemented in this study to improve
accuracy in the near-wall regions. ANSYS Fluent (Fluent, 2016) combines enhanced
wall function with the two-layer model at near-wall regions which results in a near-
wall modelling approach that can accurately calculate the flow in near-wall regions
with relatively coarser meshes (Fluent). Wall functions allow for coarser grids in the
near-wall region, thus saving computational time (Blocken, Defraeye, Derome, &
Carmeliet, 2009), and have been used in the literatures for similar flows (Nikas,
Nikolopoulos, & Nikolopoulos, 2010; Papakonstantinou et al., 2000).
The Archimedes number, Ar, (Equation 8.1) (Z. Ai et al., 2013) was used to
determine the relative dominancy of wind and buoyancy forces in both the single-sided
and cross ventilation cases:
𝐴𝑟 =𝑔𝐻∆𝑇
𝑇𝑈2 (8.1)
164 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
Where g is the gravitational acceleration (m/s2), H is the opening height (m), ΔT
is the temperature difference between inside and outside (K), T is the average
temperature (K), and U is the average wind speed (m/s) at building height. Numbers
less than 1 indicate dominancy of forced convection over buoyancy induced natural
convection, thus, buoyancy forces can be neglected with no significant loss of accuracy
(Malkawi & Augenbroe, 2004).
For this study, the Archimedes number for single-sided and cross ventilation are
0.018 and 0.015 respectively, indicating that the flow within the space is dominated
by forced convection, and the effects of buoyancy driven flow can be ignored. As such,
only mass and momentum equations have been solved and the energy equation was
not calculated.
A CFD commercial code, ANSYS Fluent 17.0 (Fluent, 2016) was employed to
perform the simulations. The SIMPLEC algorithm was adopted for pressure-velocity
coupling, and the spatial discretization was set to second-order upwind. Convergence
was assumed to be achieved when all the residuals reached the convergence criteria of
10-4. This figure is commonly used in the literature.
Numerical grids and computational domain
A full-scale 3D model of the case study building was placed in a calculation
domain. The domain size was defined according to the case study building height (H)
and has dimensions L × W × H= 1800 m × 600 m × 600 m. Upstream and downstream
lengths of the domain were 3H and 15H respectively, lateral sides were 6H, and the
domain height was equal to 6H (Figure 8.4). The domain dimensions were defined
based on the recommended values by the best practice guideline (Franke, Hellsten,
Schlünzen, & Carissimo, 2007).
Tong et al. (Tong, Chen, & Malkawi, 2016) suggest three layers of obstructions
in the CFD model to capture street canyon effects, however, this requirement can be
reduced for buildings at height while still capturing the influence of surrounding
obstructions (Malkawi & Augenbroe, 2004). Inclusion of surrounding obstructions,
however, also leads to more complicated flow patterns, and thus increased
computational time. For this study, no surrounding buildings were included. This is
considered an adequate simplification of the external environment as there are no
major obstructions with 250m on the southern side, and the primary focus of the study
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 165
was the detailed comparison of the internal flow environment in relation to different
balcony parameters and ventilation modes. The authors recognise that the exclusion of
surrounding buildings will impact the airflow. However, this effect will be the same
among all the case studies, and hence for a comparative study does not detract from
the results.
An unstructured mesh with tetrahedral volume was created using ICEM CFD
("ICEM CFD," 2016). Two layers of 1.5 and 2.5 mesh density were applied to a radius
of 40m and 100m of the case study building respectively. Grid refinements were
applied to the openings and building surfaces. Opening meshes were the most refined
areas in the domain and consisted of grids with maximum sizes of 2e-3m. Maximum
mesh size for the building surfaces ranged from 0.5m on surfaces distant from the
openings, 0.1m at surfaces adjacent to the case study unit, 4e-2m at the case study unit
surfaces, 9e-3m at the wall adjacent to the openings, to 6e-3m at the openings frames.
Grid sensitivity tests were performed by creating three sets of coarse, medium,
and fine meshes with 8, 13, and 24 million elements respectively. Air velocity at key
locations corresponding to experimental data points was used to assess the quality of
the mesh. A 4% difference in results between the coarse and medium mesh and a 1.6%
difference between medium and fine mesh were obtained. Given this, the medium
mesh with 13 million elements was considered to be able to provide grid independent
solutions and was used for the simulations.
Boundary conditions
A power law equation (Equation 8.2) was used to calculate the wind boundary
layer profile and the acquired data was applied to the inlet boundary condition.
𝑽𝒛 = 𝑽𝒓𝒆𝒇(𝒁
𝒁𝒓𝒆𝒇)𝜶 (8.2)
Where Vz (m/s) is wind speed at height z (m), Vref (m/s) is the reference velocity
at the reference height zref (m), and α is a component that is representative of terrain
roughness. Considering the case study building is located in Brisbane CBD, α was set
to 0.35 corresponding to city centre terrains (Davenport, 1960).
166 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
Figure 8.4. CFD domain size
Before the parametric study was performed the CFD model was validated against
the experimental data (explained in CFD model validation section). For the validation
model, the inlet boundary condition was defined according to the wind condition at the
time of the experiment. Reference velocity values were extracted from the reference
weather station wind data associated with the time of the experiment. For the
parametric study, however, the annual average wind speed (2.8 m/s) was used as the
reference velocity at the reference height of 8.13m (weather station’s height).
Turbulence intensity was set to 5% representing a medium intensity and turbulent
viscosity ratio was 10.
For all simulations, the outlet boundary condition was set to outflow, top and
lateral boundaries were set to symmetry, and wall boundary conditions were applied
to ground and building’s surfaces as suggested by the best practice guideline (Franke
et al., 2007). In addition, all the wall boundary conditions were no-slip wall with no
additional surface roughness.
CFD model validation
The results from CFD simulations were compared to the experimental data for
both single-sided and cross flow configurations. This was performed by considering
the incident wind direction perpendicular to the openings from the experimental data
set. For this validation, internal data corresponding to an incident southerly wind were
extracted, time averaged over 1-minute intervals and used to validate the
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 167
computational results. The extracted external wind measurements (from the reference
weather station) where used as the inlet condition in the computational model. A
number of 186 and 203 1-minute data values at each measurement point and reference
weather station were attained for single-sided and cross ventilation respectively. The
obtained data were averaged and measured values of air velocity inside the case study
were compared to the simulation results of the same coordinates. Figure 8.5 shows the
discrepancy between the simulation results and experimental data for cross ventilation
(Figure 8.5-A) and single-sided ventilation (Figure 8.5-B) as well as measurement
uncertainties. As can be seen, CFD results slightly overestimate air velocity for both
cases (1-11%). The discrepancy between the CFD results and the experimental data
can be the resulted from steady-state assumption used in the simulations. Natural
ventilation is unsteady in nature and usually involves wind fluctuations (Jiang & Chen,
2002; Lo & Novoselac, 2012). The steady-state assumption is likely to underpredict
the fluctuation effect of wind (Lo et al., 2012), hence, the discrepancy between the
simulation results and the experimental data is considered likely to be a result of the
steady-state assumption. These errors, however, are considered acceptable since
significantly higher errors have been reported in some similar studies (Z. Ai, Mak,
Niu, & Li, 2011; CF Gao & Lee, 2011a; X. Liu, Niu, Perino, & Heiselberg, 2008).
Figure 8.5. Comparison of measurement and simulation results for A) cross ventilation, and B) single-
side ventilation
8.2.3 Tests configurations (case studies)
Balconies can be identified by two main features: depth and type. Therefore, to
evaluate the effect of a balcony on natural ventilation and indoor air flow, simulation
cases were generated with different balcony depths and types. Four different balcony
168 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
depths expressed as a percentage of the living area’s length (10%, 20%, 30% and 40%),
as well as two balcony types (open balcony and semi-enclosed balcony), were defined.
Figure 8.6 illustrates these variables. The defined balcony variations were tested with
both single-sided and cross ventilation configurations. In addition, to consider the
effect balconies have under different incident winds, the case studies were tested under
four wind directions (0˚, 45˚, 90˚, and 180˚). Including the cases without a balcony, a
total of 72 combinations of the aforementioned variables were formulated and tested.
A summary of the configuration parameters is presented in Table 8.2. It needs to be
noted that except for the varied parameters, all the other parameters such as opening
size and space length were kept constant.
Table 8.2. Configuration parameters.
Variable Variations
Balcony type Open Balcony (OB), Semi-enclosed Balcony (SB)
Ventilation mode Cross ventilation (CV), Single-sided Ventilation (SSV)
Balcony depth 0% (without balcony), 10%, 20%, 30%, 40%
Wind direction 0˚,45˚, 90˚, 180˚
Figure 8.6. Balcony types, open balcony (left) and semi-enclosed balcony (right)
8.2.4 Thermal comfort model
Comfort zone boundaries can be extended by elevating indoor air velocity
(Arens, Gonzalez, & Berglund, 1986; Ernest, Bauman, & Arens, 1992). The
recommended model for prediction of thermal conditions for cases with the indoor air
speed of greater than 0.2 m/s is SET* (ASHRAE, 2013). ASHRAE standard-55
ɵ
d
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 169
(ASHRAE, 2013) defines SET* as the temperature of an environment at 50% relative
humidity and average air speed of below 0.1 m/s, where air temperature and radiant
temperature are equal in which “the total heat loss from the skin of an imaginary
occupant with an activity level of 1.0 met and a clothing level of 0.6 clo is the same as
that from a person in the actual environment, with actual clothing and activity level”.
In the current study, nearly all the studied cases had represented air velocity of greater
than 0.2 m/s, therefore, the SET* model was adopted for thermal comfort evaluation.
The SET* model accounts for the combined effect of temperature, humidity, air
velocity, metabolic rate, and clothing insulation on thermal comfort of the occupants
(Gagge et al., 1986). Since changes in air velocity as a result of balcony provision is
of interest of this study, the remaining components were set to constant. Temperature
and humidity values for January was extracted from the acquired meteorological data.
The occupants were assumed to be involved in a sedentary activity, metabolic rate,
therefore, was set to 1.2 met. Light-weighted clothing corresponding to summer
condition was also assumed, hence, 0.5 clo clothing value was adopted. The climatic
conditions, obtained air velocity, metabolic rate, and clothing insulation values were
then used as inputs for SET* calculations using WinComf program (ME Fountain &
Huizenga, 1996).
8.3 RESULTS AND DISCUSSION
8.3.1 Results summary
Indoor air flow of the case studies was obtained using the validated CFD model.
Average velocity in the living area volume was extracted from the results and
corresponding SET* values were calculated. Results are presented using average
velocity and thermal comfort. The obtained velocity results are summarised in Figure
8.7 and are categorised based on the investigated variables. The presented results for
each variable (x-axis) is cumulative results from all the simulated cases where the
parameter of interest is looked at independently. For instance, the OB results are
extracted from all the cases with open balcony with different ventilation modes,
depths, and wind angles.
What stands out in Figure 8.7-A is that there is a significant difference between
the average velocity range when cross ventilation operates compared to that of the
single-sided ventilation (about 7 times higher). To reveal the effects of the parameters
170 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
on air velocity of each ventilation mode, cross ventilation and single-sided ventilation
results are plotted separately on Figure 8.7-B and Figure 8.7-C respectively. As can be
seen, both ventilation modes respond similarly to the change of variables in most cases:
average velocity decreases as balcony depth increases, in terms of wind angle, the
highest velocity is achieved when the wind is normal to the openings whereas the
lowest velocity is associated with the wind parallel to the openings (90˚). An open
balcony indicates a better performance compared to the semi-enclosed balcony for
both ventilation modes. However, for single-sided ventilation, the addition of an open
balcony can result in an increase of air velocity compared to the cases without balcony
(0%), while in cross ventilation both balcony types result in a lower velocity than the
cases without balcony. Therefore, there is a potential for improvement of single-sided
ventilation through the addition of a balcony.
To look into the results in more detail, the average velocity results for all the
simulated cases are further presented in the following sections for single-sided and
cross ventilation separately.
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 171
Figure 8.7. Results summary for both ventilation modes (A), cross ventilation (B), and single-sided
ventilation(C)
Single-sided ventilation
The average velocity of the cases with single-sided ventilation are presented in
Figure 8.8. The results for different balcony depths are categorised based on the
172 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
balcony types and the prevailing wind directions. As can be seen in Figure 8.8, in
overall, cases with semi-enclosed balcony present lower air velocities than the cases
with open balcony for all the prevailing wind directions. Furthermore, the addition of
semi-enclosed balconies have resulted in lower average velocities than the cases
without a balcony. In contrast, provision of open balconies have improved the average
velocity in most cases. An increase in balcony depths have resulted in decrease of
average velocity in the cases with semi-enclosed balconies. This also applies to most
of the cases with open balconies except for the instances of 90˚ incident winds. In such
cases, the average velocity increases with an increase in balcony depth of up to 30%
and decreases for 40% depth. In terms of prevailing wind direction, highest average
velocity is achieved when the wind is perpendicular to the openings (0˚), followed by
45˚, 180˚, and 90˚ incident winds respectively.
To summarise, the addition of a semi-enclosed balcony decreases the indoor
average velocity, whereas, provision of an open balcony can improve the ventilation
performance in most cases. The open balcony can improve the average velocity
significantly (up to 6 times) for the most unfavourable prevailing wind direction (90˚).
The results also highlight the importance of building orientation in ventilation
performance of single-sided ventilation, where placing the openings toward the
prevailing wind direction captures at least twice air velocity as in other orientations.
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 173
Figure 8.8. Indoor average velocity for single-sided ventilation subject to various balcony type, depths
and prevailing wind direction.
Cross ventilation
Average velocity results for cross ventilation are presented in Figure 8.9 based
on different balcony types, depths, and prevailing wind directions. Similar to single-
sided ventilation, in cross ventilated cases semi-enclosed balconies present a lower
velocity average than open balconies. Both balcony types, however, result in a
reduction of average velocity compared to the cases without balcony. The increase in
balcony depth decreases the average velocity in most cases. The discrepancy resulted
by depth increase in cross ventilation, however, is up to 9.5% on average which is
174 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
noticeably lower than that of the single sided ventilation (31% on average). The
prevailing wind direction also affects the average velocity in cross ventilation with the
highest velocity corresponding to the wind perpendicular to the openings (0˚),
following by oblique (45˚) and parallel (90˚) wind directions.
It needs to be mentioned that despite the potential improvement of natural
ventilation in single-sided ventilation by the addition of an open balcony, cross
ventilated cases still perform significantly better (at least twice) than the improved
single-sided ventilation cases.
Figure 8.9. Indoor average velocity for cross ventilation subject to various balcony type, depths and
prevailing wind direction.
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 175
8.3.2 Sensitivity analyses
To identify the separate and combined impact of the balcony type, depth and the
wind angle on ventilation performance, a sensitivity analysis for each ventilation mode
was conducted.
To this end, a base case for each ventilation mode was selected and the other
cases were compared to the base case. The configurations with the lowest average
velocity were chosen as the baseline cases. Natural ventilation sensitivity was
expressed as a percentage of increase in average velocity as a result of altering the
investigated variables comparative to the baseline case (Equation 8.3).
SA=Va-Vb
Vb× 100 (8.3)
Where SA is the sensitivity percentage, Va (m/s) is average velocity of
configuration a, and Vb (m/s) is average velocity of the baseline case. Varying
parameters for the analysis are balcony type (BT), balcony depth (D), wind angle (W),
and the combination of these independent variables (BT+D, BT+W, D+W, and
BT+D+W). The baseline cases for single-sided ventilation and cross ventilation were
the case with 10% length semi-enclosed balcony under 90˚ wind direction (SB10-90),
and the case with 40% length semi-enclosed balcony under 90˚ wind direction (SB40-
90) respectively. Figure 8.10 shows the distribution of air velocity inside and around
the baseline cases of single-sided and cross ventilation configurations. Figure 8.10-A
represents the air velocity distribution at the unit’s height around the case study
building for single-sided ventilation (left) and cross ventilation (right). As can be seen,
the obstruction caused by the extended wall at the right side of the building have
induced airflow from right side to the left in cross ventilation case. It needs to be noted
that Figure 8.10-C (single-sided baseline case) only represents the airflow inside the
living area and excludes the balcony. Due to a relatively higher air velocity inside the
balcony, it was not possible to capture air movement in both balcony and living area
using the same scale for velocity.
The sensitivity analysis was conducted for single-sided and cross ventilation
separately and results are discussed below.
176 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
Figure 8.10. A) Velocity magnitude around the building for baseline cases of single-sided ventilation
(left) and cross ventilation (right), B) Velocity magnitude plan at 1.2m (top) and section A-A (bottom)
for cross ventilation baseline case, and C) Velocity magnitude plan at 1.2m (top) and section A-A
(bottom) for single-sided ventilation baseline case.
Single-sided ventilation
Sensitivity percentage of average velocity for the investigated variables for
single-sided ventilation is presented in Figure 8.11. As can be seen, indoor average
velocity is mainly affected by the incident wind angle (W) followed by the balcony
type (BT) and balcony depth (BD) respectively. Among the two varying parameters,
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 177
air velocity is least sensitive to the combination of balcony type and balcony depth
(BT+D) while reasonable improvements can be achieved by a combination of balcony
type with wind angle (BT+W). In addition, change in all the parameters simultaneously
improves the ventilation performance but is not as effective as the change in balcony
type and wind angle. This highlights the high sensitivity of indoor air velocity to the
approaching wind direction and the minimal effect of balcony depth.
Figure 8.11. Sensitivity percentage of average air speed to different variables for single-sided
ventilation
Cross ventilation
It can be seen in Figure 8.12 that indoor air velocity is most sensitive to the
change of incident wind angle (W) followed closely by balcony depth (D) and is least
sensitive to the balcony type (BT). Additionally, varying two parameters together does
not significantly change the SA compared to one varying parameter and there is a small
difference between SA of single and two parameters (6% between the best and the
worst configurations). However, the most improvement can be achieved by changing
all the variables simultaneously.
Comparing the sensitivity analysis of cross ventilation and single-sided
ventilation shows that single-sided ventilation is much more sensitive to the change of
the investigated parameters compared to cross ventilation. Altering different
parameters in single-sided ventilation results in approximately 300% improvement on
average while this number is about 50% for the cross ventilation for the same variables.
Besides, the mean SA discrepancy between different variable configurations for the
cross ventilation is around 10% while in single-sided ventilation this number is about
50 times higher (~ 500%). The suction effect caused by the pressure difference
0
200
400
600
800
1000
1200
1400
1600
1800
BT D w BT+D BT+W D+W BT+D+W
SA (
%)
Variables configurations
Median
178 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
between inlet and outlet of the cross ventilated cases is very dominant making this case
less sensitive to the change of different variables compared to the single-sided
ventilation. The average velocity in single-sided ventilation is very low that even small
configuration changes result in a notable change of average velocity compared to the
baseline case. It can also be seen that among the independent parameters, both
ventilation modes are most sensitive to the wind direction change. Although wind
direction cannot be controlled by building designers, buildings can be oriented in a
way to take the most advantage of outside wind conditions.
Figure 8.12. Sensitivity analyses of average air speed to different variables for cross ventilation
configuration
It needs to be noted that the results presented in Figure 8.11 and Figure 8.12 are
calculated using average velocity at the 1.2m plane. The same trends were also found
using average velocity at 0.6m and 1.8m planes as well as the living area’s volume.
8.3.3 Thermal comfort analyses
The results of average velocity in the living area of the case studies were used in
the SET* index calculation. As could be expected, SET* values in the cases with cross
ventilation were significantly lower than that of the single-sided ventilation (3.4˚C on
average). Differences in SET* values of various cases (ΔSET*) can be interpreted as
the cooling effect on the human body. The potential cooling effect of the investigated
variables for the single-sided and cross ventilation configurations are presented in
Figure 8.13. It is evident from these results that the SET* values in the cases with
single-sided ventilation respond more to the change of the variables compared to that
of the cases utilised with cross ventilation. In addition, regardless of ventilation mode,
0
20
40
60
80
100
120
140
BT D w BT+D BT+W D+W BT+D+W
SA(%
)
Variables configuration
Median
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 179
the approaching wind angle has the most influence on potential cooling effect on the
occupants compared to the balcony depth and balcony type. It follows by balcony type
in single-sided ventilated cases and balcony depth for the cross ventilated cases. These
results are in accord with the results from sensitivity analyses of the average velocity.
Figure 8.13. Investigated parameters potential cooling effect
In addition, ΔSET* of minimum and maximum values for single-sided and cross
ventilation are 3.2˚C and 1.1˚C respectively. This indicates that single-sided
ventilation responds more to the change of variables compared to the cross ventilation
highlighting a higher potential for improvement.
It needs to be noted that cross ventilation can provide adequate ventilation rate
and thermal conditions independent of changing variables, thus is more likely to create
year-round comfort compared to single-sided ventilation which is heavily dependent
on variables particularly wind direction and building orientation.
8.4 CONCLUSION
In-situ full-scale measurements of air velocity were conducted in a high-rise
residential apartment. The collected data was used to validate a CFD model from
which a detailed investigation of the separate and combined effect of the balcony type
and depth, ventilation mode, and the wind angle on indoor ventilation was performed.
Various case studies were formulated based on two balcony types, four balcony depths,
two ventilation modes and four wind angles. Average velocity and SET* index were
used as criteria and the following results were found:
0 0.5 1 1.5 2 2.5 3 3.5
W
D
BT
SET* (˚C)
Var
iab
les
Cross ventilation Single-sided ventilation
180 Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings
• An open balcony results in a higher indoor velocity compared to the semi-
enclosed balcony and natural ventilation performance of single-sided
ventilation can be improved (up to 80%) by provision of open balcony.
• The increase in balcony depth leads to decrease in air velocity.
• Among the tested incident wind directions, highest indoor air velocity
corresponded to when the wind is normal to the openings, and is lowest
when the prevailing wind is parallel to the openings. This highlights the
importance of orientating building toward the prevailing wind direction for
natural ventilation improvement.
• Sensitivity analyses revealed that among wind angle, balcony depth, and
balcony type, both ventilation modes are most sensitive to the change of
wind direction. It was also found that the effect of altering the investigated
parameters on natural ventilation performance is much greater in single-
sided ventilation (300% on average) compared to the cross ventilation (50%
on average). This emphasises on the importance of appropriate design in
the case of single-sided ventilation.
• Among the two varying parameters, the most improvement was achieved by
changing wind direction and balcony type in the case of single-sided
ventilation. This highlights the significant effect of wind direction which
can be translated to the building orientation in building design. In cross
ventilation, however, the most improvement was associated with changing
all the parameters together.
• The cooling effect on the human body (ΔSET*) shows changing the
parameters in single-sided ventilation has the maximum potential of 3.2 ˚C
cooling effect improvement, while, this number was found to be 1.1˚C for
cross ventilation.
Comparing single-sided and cross ventilation under the same circumstances
shows a significantly better natural ventilation performance in the case of cross
ventilation. Analyses of different parameters, however, reveals that single-sided
ventilation is much more sensitive to the parameters alteration compared to the cross
ventilation. Therefore, considering lower performance of single-sided ventilation,
additional care much be given to its design to assure the ventilation effectiveness.
Chapter 8: On the effect of provision of balconies on natural ventilation and thermal comfort in high-rise
residential buildings 181
Since implementation of cross ventilation is not always a possible solution, especially
in dense urban areas, findings of this study provide solutions for improvement of
single-sided ventilation design through appropriate choice of balconies. Having said
that, the provided results can also be used for the improvement of cross ventilation
through design.
8.4.1 Limitations and future work
The main aim of this study is to provide comparative results about the effect of
different balcony attributes on natural ventilation performance of high-rise residential
units. Similar to any other study, this study has some limitations that need to be
addressed in future research. The current study is carried out with an isothermal
assumption and buoyancy-driven ventilation is neglected due to dominant effect of
wind. In future studies, buoyancy-driven ventilation can be considered to evaluate the
effect of temperature gradient and buoyancy forces on ventilation performance of the
balcony mounted residential units.
Acknowledgements
Computational and data visualisation resources used in this work were provided
by the HPC and Research Support Group, Queensland University of Technology,
Brisbane, Australia.
This research did not receive any specific grant from funding agencies in the
public, commercial, or not-for-profit sectors.
182 Chapter 9: Discussion
Chapter 9: Discussion
This chapter provides a discussion about the outcomes of this thesis. As
explained in Chapter 3, this study focuses on methods for evaluation and prediction of
natural ventilation as well as the effect of design related parameters on natural
ventilation design. Accordingly, this chapter is divided into two sections: first,
methods of analysis, and second, design related parameters.
9.1 METHODS OF ANALYSIS
Different methods used in natural ventilation studies were reviewed and
analysed in Chapter 4, and a model for the application of these methods during the
design process was proposed. This model is also presented in this chapter in Figure 9.1
to assist interpretation of the discussion. The model starts with less complicated
methods at early stages of design. These methods (e.g. empirical and analytical) are
rapid and they provide reasonable estimations of ventilation performance, therefore,
they suit early design stages. As design evolves more reliable results are needed,
therefore, more accurate and complicated methods are suggested. Three main parts of
this model, namely, “feasibility”, “detailed design”, and “after construction” were
applied in the remaining chapters of this thesis using a case study building. Chapters
5 and 6 explored the first step of the model (“feasibility”) by proposing empirical
models for prediction of air velocity at the openings and inside of an intended high-
rise building design. Two separate models were proposed for single-sided and cross
ventilation as their performance differs significantly. It was found that for the cross
ventilated case, air speeds at the openings of the case study were about 30% of the
wind captured at the reference weather station, while this number decreases to around
15% for the single-sided case. Although it is desirable to test models in similar
buildings for a firm outcome, it was concluded that empirical models can provide a
reasonable estimation of natural ventilation performance in buildings within a short
time-frame. It is also seen that the model is not perfectly accurate. However, it meets
the requirements for the concept stage of the design where an overall estimation of
Chapter 9: Discussion 183
ventilation performance seems satisfactory, hence, these models were considered to be
appropriate for early design stages.
Figure 9.1. Natural ventilation design process model within the overall design process
The last stage of the natural ventilation design process model (“construction and
after construction”) was applied and tested in Chapters 6 and 7. Full scale experimental
data collection was carried out in the selected case study. The measured values were
then analysed, and natural ventilation and thermal comfort performance were
evaluated. These studies provided a detailed analysis of natural airflow behaviour
inside high-rise buildings. It was also proved that cross ventilation performs
significantly better than single-sided ventilation. The experimental method allowed for
a detailed analysis and evaluation of ventilation performance by providing realistic
Analytical and
Empirical
Multi-zone CFD
CFD + BES
CFD + Multi-zone
CFD + Small-scale experiments
Small-scale
experiments
Full-scale experiments
Feasibility
Concept Design
Detail Design
Final Design
/Documentation
Construction
Design Stages Activities
After Construction
184 Chapter 9: Discussion
data that reflected the reality in a constructed building and led to improved
understanding of natural ventilation mechanisms under different weather conditions.
Accordingly, these analyses identified that there is room for improvement of natural
ventilation in the studied unit, especially for the case of single-sided ventilation.
Among the possible design alterations, it was decided to investigate balconies as one
of the most commonly used design features in hot-humid climates. Hence, methods
suggested for the “final design/documentation” stage were employed to examine
different configurations of balconies and their effect on natural ventilation and thermal
comfort (Chapter 8). However, the recommended small-scale experiment was
substituted by data acquired from the full-scale experiment. The collected data along
with reference weather data were used for validation of a CFD analysis. After reaching
an acceptable agreement, the CFD analysis was used in the simulation of 70 different
cases with various balcony configurations for both single-sided and cross ventilation.
It was found that an open balcony contributes to a higher air velocity for both
ventilation modes and integration of an open balcony can improve the ventilation
performance of single sided ventilation by up to 80%. The coupled method suggested
for “final design”, hence, was considered to be appropriate for finding design solutions
that improve natural ventilation.
Middle design stages (“concept design” and “detail design”) in the natural
ventilation design process model were bypassed in this study since they were mainly
suggested at these stages due to their decreased time and computational costs.
Accordingly, only the three main methods that noticeably differ in time, resources, and
accuracy were chosen and tested. This study showed that the proposed natural
ventilation design process model serves its purpose and can be practically used in the
natural ventilation design of high-rise buildings.
9.2 DESIGN RELATED PARAMETERES
This section of the discussion chapter is allocated to the effect of design related
parameters on natural ventilation performance of high-rise buildings. Findings from
the previous chapters as well as information extracted from the review of the literature
form the body of this section. All the available studies about the impact of different
architectural features on natural ventilation were reviewed (Chapter 2) and the extent
of their impact was identified. In addition, some design related parameters such as
ventilation mode, balconies, and orientation of openings were investigated in Chapters
Chapter 9: Discussion 185
6-8 of this thesis. All these findings and the extent of their effect were gathered and
combined in a flowchart that can be used for natural ventilation design of building.
This chart is presented in Figure 9.2. The diagram takes into account four main areas
including ventilation mode, building orientation, design of openings, and other
relevant design elements. Results of Chapter 6, 7, and 8 of this thesis as well as some
previous research studies (Fung & Lee, 2014) confirm that ventilation mode is the
most influential design parameter in natural ventilation performance and that cross
ventilation outperforms single-sided ventilation significantly (CF Gao & Lee, 2011a).
Accordingly, the flowchart starts with ventilation mode and suggests the implications
of cross ventilation rather than single-sided ventilation where possible. The next
section of the diagram is allocated to the building orientation (based on findings of
Chapter 8). Results of building orientation studies reveal that orienting the building
toward the prevailing wind offers the best orientation in terms of natural ventilation
and it is followed by 45° and 90° orientations respectively (Aflaki et al., 2016; S
Omrani, V Garcia-Hansen, B R. Capra, & R Drogemuller, 2017b; S. Omrani, V.
Garcia-Hansen, B. R. Capra, et al., 2017c). Therefore, 45° following by 90° are the
suggested orientations where 0° is not possible. Building orientation of 90° is not
included in the cross ventilation section of the diagram as it offers the least airflow
among the investigated orientations. For single-sided ventilation, however, the worst
orientation is 180° (results of Chapter 6), therefore, 90° is included in the diagram as
a better option. Since only these major orientations have been investigated in the
literature and the current thesis, and no information about the other orientations were
found in the literature, other possible orientations are not included in the diagram. In
terms of design of openings, side-hung or full-end slider openings are suggested to
perform best for cross ventilation (CF Gao & Lee, 2011b). Therefore, these window
types are suggested for cross ventilation. However, side-hung windows are proved to
perform best under 0° and 45° orientations for single-sided ventilation (CF Gao & Lee,
2011b), and casement windows for 90° orientation (O'Sullivan & Kolokotroni, 2017;
H. Wang et al., 2015). Accordingly, these window types are suggested based on the
chosen orientation for single-sided ventilation on the previous section of the chart. As
the chart moves on other recommendations are made if further improvement of natural
ventilation performance is required. For cross ventilation, increased Window to Wall
Ratio (WWR) is suggested (Fung & Lee, 2014) followed by same level of inlet and
outlet (W. Yin et al., 2010), and use of rectangular windows with width to height ratio
186 Chapter 9: Discussion
of 0.5 (Derakhshan & Shaker, 2017) respectively. For single-sided ventilation,
separate inlets and outlets are suggested (Matheos Santamouris & Allard, 1998),
followed by increase of WWR (Fung & Lee, 2014), and placing inlet and outlet far
apart (Hassan et al., 2007; Matheos Santamouris & Allard, 1998). In Chapter 8, it was
proved that provision of an open balcony for single-sided ventilation improves the
ventilation performance. Therefore, the next step of the chart suggests addition of an
open balcony. Additional suggestions about specific building elements were also
found in the literature regarding the enhancement of single-sided ventilation. These
suggestions include utilisation of a ventilation shaft (Prajongsan & Sharples, 2012)
and courtyard (Chiang & Anh, 2012) which are included in the chart as the last stages
after opening design and provision of balcony.
One point to be noted is that suggested modifications for cross ventilation are
less than those for single-sided ventilation. The reason for that may be the efficiency
of cross ventilation as is. As explored is Chapters 6, and 7, cross ventilation can
provide thermal comfort for the majority time of the year whereas single-sided is much
less efficient, hence, there is significant room for improvement for single-sided
ventilation.
It also needs to be highlighted that the part in the chart referred as “need
improvement?” can be tested using the “natural ventilation design process model”
presented in Chapter 4 and explained in the previous section. As previously mentioned,
these methods can be chosen based on the stage of design, available time, and
resources.
The natural ventilation design flowchart presented in this section is to be used
as a guideline for natural ventilation design of high-rise buildings. Not all of the
recommendations may be applicable for all designs, therefore, the building designers
need to make adjustments in accordance with the design requirements. Having said
that, this chart has gathered all the available recommendations available in the
literature and has put them in a holistic order that can be very helpful for natural
ventilation design of buildings.
Both diagrams presented in this section support building designers for natural
ventilation design of buildings. The first mainly contributes to the evaluation part of
the design and the second is related to the design elements and their effect on natural
Chapter 9: Discussion 187
ventilation. These two diagrams complement each other as the first is to be used for
testing the effectiveness of the design elements suggested in the second.
188 Chapter 9: Discussion
Figure 9.2. Natural ventilation design flowchart.
Use cross ventilation
Use single-sided ventilation
Add side-hung or full-end
slider windows
Use separate inlet and outlet
Utilize natural ventilation into building design
process
Finish
Increase WWR
No Yes
Yes
Add ventilation shaft
No
Yes
Add courtyard No Yes
No No
Yes
No
Yes
Orient toward the prevailing
wind
Yes
No Orient 45° of the prevailing
wind
Use same level inlet and outlet
No
Yes
Use rectangular window shape with width to height ratio of
0.5
No
Orient toward the prevailing
wind
Yes
No
Orient 45° of the prevailing
wind
Yes
No Orient 90° of the prevailing
wind
Add side-hung windows
Yes
Yes
No
Increase WWR No
Yes
Place the openings far
apart
No
Add casement windows
Building orientation
Yes
Openings
Is it possible to orient toward
prevailing wind?
Is it possible to orient toward
prevailing wind?
Is it possible to orient 45°of
prevailing wind?
Need improvement?
Are inlet and outlet placed separately?
Have you increased
WWR?
Have you increased
WWR?
Have you placed
openings far apart?
Have you used same
level inlet and outlet?
Have you used
rectangular windows?
Have you added
ventilation shaft?
Have you added
courtyard?
Need
improvement?
Add an open balcony
No Have you
added open balcony?
Yes
Is cross ventilation possible?
Chapter 10: Conclusion 189
Chapter 10: Conclusion
Aligned with energy efficiency policies and sustainable building design,
integration of natural ventilation into building design as a passive cooling system has
regained attention in the last few decades. The comprehensive survey of the literature
showed that there is a rich body of knowledge about natural ventilation in buildings.
It was also found that most of the available studies focus on low-rise and simple
geometries and very few are based on high-rise subjects. Insufficient information
related to high-rise buildings motivated the current study. The aim of this study,
therefore, is to improve natural ventilation design of high-rise residential buildings in
hot-humid climates. Natural ventilation performance evaluation and ample
information about the effect of different design features on natural ventilation are
prerequisites of a successful ventilation design. Accordingly, objectives of this thesis
were defined at two levels. Firstly, facilitating the process of natural ventilation
prediction and evaluation for designers. Secondly, producing a design guideline that
accommodates different design related parameters and puts them in a holistic way
based on the extent of their effect on natural ventilation performance.
The first objective was achieved in two steps. Firstly, a detailed analysis of the
available natural ventilation evaluation tools with regards to high-rise building
characteristics was provided. From that, a model for integration of these methods into
different design stages of high-rise buildings was proposed. Secondly, proposed
methods for critical design stages were employed and the proposed model and its
appropriateness was explored.
The second objective was also accomplished at two stages. Firstly, by
comprehensive review of the literature and extraction of the effect of currently studied
architectural features on natural ventilation performance and identification of the gaps
in the knowledge. Secondly, by investigation of the effect of two major design related
parameters, that were identified as gaps, on natural ventilation performance, namely
ventilation mode, and balconies.
This study is developed using a case study approach as a methodology.
Accordingly, a case study that could satisfy the selection criteria and accommodate the
190 Chapter 10: Conclusion
research objectives was selected. Full-scale in-situ measurements and CFD were
chosen as methods. The full-scale experiment was carried out in the case study unit
where air velocity, temperature and relative humidity were measured at different points
throughout the space. The effect of natural ventilation mode on ventilation
performance, and correlation of air velocity inside the apartment with meteorological
data were investigated using the collected experimental data. The full-scale
measurements data was then used for validation of a CFD model which was further
employed for investigation of the effect of the provision of balconies on natural
ventilation. The outcomes of this study were presented in a form of peer-reviewed
publications in five chapters. In addition, a design flowchart based on different design
related features and their influence is presented in the Discussion chapter. In this
chapter, a summary of the key findings, design implications, research limitations, and
direction for future works are presented.
10.1 SUMMARY OF KEY FINDINGS
In addition to the major contributions of the current thesis that were presented in
Chapter 9, key findings of the current study are as follows:
1. The available natural ventilation evaluation methods were critically
analysed according to the characteristics of high-rise buildings and the
advantages and limitations of these tools were discussed. From this,
designers can make insightful decisions on selecting desired tool(s) based
on their project needs and resources. Additionally, a design process model
was proposed considering the common design stages, their constraints, and
requirements. The least expensive methods (both time and monetary) were
suggested to be used in early design stages where different design
alternatives need to be tested. Given that accuracy matters more as the
design evolves, more accurate tools were suggested for the later design
stages. This study has practical implications with regards to integration of
natural ventilation evaluation tools into the overall building design process
which facilitates the evaluation of natural ventilation performance.
2. The correlation between wind speed at different weather stations and air
velocity at building openings and indoor spaces were investigated for single-
sided and cross ventilation. This investigation revealed the existence of a
Chapter 10: Conclusion 191
linear relationship. The expected air velocity in a building, therefore, can be
considered as a fraction of wind speed. This fraction for single-sided
ventilation was found to be approximately half of the cross ventilation. A
similar relation was also found to rule among different weather stations
within a region with different terrain roughness. Additionally, it was found
that the urban context of a weather station and its similarity to the building
site affects this ratio. Therefore, data from a meteorological station with
similar terrain roughness to the design site was suggested to be used for
natural ventilation prediction. These correlations can be used for prediction
of air velocity at building openings and at indoor spaces using
meteorological data. Findings of this study are of the benefit of building
designers for a quick ventilation performance estimation at the design stage.
3. The effect of natural ventilation mode (single-sided and cross ventilation)
on indoor thermal conditions and ventilation performance of high-rise
residential buildings was investigated using the full-scale in-situ
experimental method. Indoor thermal conditions, airflow distribution, and
the effect of wind direction on internal airflow were investigated. Regarding
indoor thermal environments, SET* analysis revealed that under similar
weather conditions, the indoor environment in cross ventilation was on
average 3˚C cooler than single-sided ventilation. Furthermore, evaluated
using the extended PMV model, the indoor thermal environment was found
to be acceptable for about 70% of the time when cross ventilation operated,
whereas, single-sided ventilation failed to provide adequate thermal comfort
(provided only 1% of the time). If we assume that building occupants turn
on the air-conditioners when they are subjected to thermal discomfort, a
significant amount of cooling energy can be saved by application of cross
ventilation compared to single-sided ventilation. In terms of average air
velocity, the performance of cross ventilation was two to four times higher
than single-sided ventilation. It was also found that reference wind direction
affects the ventilation rate inside the building where incident wind
perpendicular to the openings resulted in highest air velocity, whereas,
lowest air speed was associated with wind direction parallel to the openings.
To summarise, evaluated by different criteria, cross ventilation was proved
192 Chapter 10: Conclusion
to be much more effective than single-sided ventilation. Considering that
the ventilation mode is one of the main determinants of ventilation
performance, the outcomes of this study allow architects to make more
informed decisions in terms of natural ventilation design. Additionally, the
full-scale measurement data gathered for this study was used for a CFD
model validation that allowed simulation of another design related
parameter (balcony) under various weather conditions.
4. The effect of various balcony characteristics on natural ventilation
performance of high-rise buildings was investigated using CFD simulations.
Different balcony features such as balcony type (open balcony and semi-
enclosed balcony) and balcony depth (10%, 20%, 30%, and 40%) were
altered and simulated under various wind directions (0˚, 45˚, 90˚, and 180˚)
for single-sided and cross ventilation. The results showed that overall, open
balconies performed better than semi-enclosed balconies. It was also found
that for single-side ventilation, the addition of an open balcony mostly
improves the ventilation performance, while, in cross ventilation, both
balcony types worsened the ventilation performance. Although, it needs to
be noted that the worst cross ventilated cases still perform much better than
the improved single-sided cases. Additionally, increase in balcony depth
mostly resulted in air velocity reduction for both ventilation modes. With
regards to the effect of wind direction, incident winds perpendicular and
parallel to the openings resulted in highest and lowest indoor average
velocities respectively which is in agreement with the experimental results.
The sensitivity analysis showed that among the investigated parameters,
ventilation performance is most sensitive to the change of wind direction
which highlights the importance of building orientation in relation to the
prevailing wind direction. Furthermore, the analysis revealed that single-
sided ventilation is more sensitive to the change of variables than cross
ventilation. Thus, extra care needs to be taken when designing for single-
sided ventilation. Since balconies play an important role in the architecture
of hot-humid climates, effective integration of them into the building design
results in a better ventilation performance. Findings of this study, therefore,
Chapter 10: Conclusion 193
provide information for better integration of this element into the building
design.
The following design implication can also be concluded from the
abovementioned key findings:
Under similar conditions, cross ventilation is proved to perform
significantly better than single-sided ventilation. Cross ventilation, therefore, is
the ventilation mode of choice. Being less sensitive to the change of variables,
cross ventilation is more likely to provide adequate ventilation even under poor
choices of design related parameters. However, it does not eliminate the need for
consideration of the influential design features. Single-sided ventilation, on the
other hand, is very sensitive to the change of variables such as wind direction
which highlights the importance of design decisions in its effectiveness. Both
ventilation modes were found to perform their best when the wind is
perpendicular to the openings. Orienting buildings toward the prevailing wind,
therefore, plays a crucial role in improving the ventilation performance. In terms
of balcony design, the open balcony is a better choice than semi-enclosed
balcony in most of the cases. In fact, addition of the open balcony to single-sided
ventilation can result in improvement of ventilation performance. Other
parameters that were found to affect ventilation performance are building
surroundings and obstructions. Although these parameters cannot usually be
modified by designers, they need to be considered in the design process and
ventilation performance evaluations.
In conclusion, the current study findings contribute to better natural ventilation
design of high-rise buildings by facilitating the ventilation performance evaluation,
and by providing information on the effect of different design features on ventilation
performance.
10.2 LIMITATIONS AND FUTURE WORK
Similar to any other research, there were some limitations associated with this
study as stated below.
The field measurements for this study were conducted only at one unit of a high-
rise building since it was not possible to gain access to any other apartment. Given that
wind speed changes with height, different air velocity magnitudes can be expected at
194 Chapter 10: Conclusion
the other units located at different levels. In addition, due to site restrictions, we could
not install a local reference weather station at the case study site, therefore, the nearest
weather station was used as the reference weather station.
In terms of building shape and layout, this study used a case study with
rectangular floor layout with identical inlet and outlet for cross ventilation
configurations. The results and conclusions drawn from this study, therefore, may not
be applicable to the buildings with significantly different layout and/or opening
configurations. This also applies to the climates that greatly differ from Brisbane’s
climate.
The current study considers the occupants as active participants who have
control on the openings. The discomfort resulting from high air velocity, therefore,
was not taken into account.
According to the limitations stated above, as well as areas that were not covered,
the following future research directions are suggested:
In this thesis, the correlation between wind speed and air velocity at the
building was investigated based on the data gathered at one level of a high-rise
building. More comprehensive investigations including air velocity data from
different floors at different heights are suggested which can lead to an empirical
equation that can be used for ventilation prediction using meteorological data.
The effect of the surrounding obstructions was partially considered in the
analysis of this study, however, the combined effects of design related
parameters and surrounding constructions configurations on natural ventilation
performance are yet to be investigated.
This study investigates the effect of balcony provision on ventilation
performance of indoor spaces. Environmental conditions inside the balconies,
however, were not evaluated. Given that higher floors of high-rise buildings are
subject to high wind magnitude, a study on balconies from the draft perspective
can be beneficial to identify the effective height for the addition of balcony into
high-rise buildings.
Cross ventilation was proved to provide thermal comfort much more
effectively than single-sided ventilation. Considering that occupants would use
air-conditioners when they are uncomfortable and vice versa, cross ventilation
Chapter 10: Conclusion 195
offers a potential energy reduction. The current energy simulation software using
in Australia (e.g. BERS Pro and ACCURATE), however, does not consider the
effect of ventilation mode on energy conservation. The addition of a feature that
accounts for such an effect, therefore, can improve the outcomes of these
programs.
This thesis focused on methods for natural ventilation prediction, as well
as, some design related parameters. The relative effect of all the design related
parameters on natural ventilation, however, were not investigated. Ultimately, a
tool that can optimise the natural ventilation design based on all the design
related parameters can be developed as a future work.
Bibliography 197
Bibliography
Aflaki, A., Mahyuddin, N., Al-Cheikh Mahmoud, Z., & Baharum, M. R. (2015). A
review on natural ventilation applications through building façade components
and ventilation openings in tropical climates. Energy and Buildings, 101, 153-
162. doi:http://dx.doi.org/10.1016/j.enbuild.2015.04.033
Aflaki, A., Mahyuddin, N., & Baharum, M. R. (2016). The influence of single-sided
ventilation towards the indoor thermal performance of high-rise residential
building: A field study. Energy and Buildings, 126, 146-158.
doi:http://dx.doi.org/10.1016/j.enbuild.2016.05.017
Ai, Z., Mak, C., & Cui, D. (2013). On-site measurements of ventilation performance
and indoor air quality in naturally ventilated high-rise residential buildings in
Hong Kong. Indoor and Built Environment, 24(2), 214-224.
Ai, Z., Mak, C., Niu, J., & Li, Z. (2011). Effect of balconies on thermal comfort in
wind-induced, naturally ventilated low-rise buildings. Building Services
Engineering Research and Technology, 32(3), 277-292.
Ai, Z., Mak, C., Niu, J., Li, Z., & Zhou, Q. (2011). The effect of balconies on
ventilation performance of low-rise buildings. Indoor and Built Environment,
20(6), 649-660.
Ai, Z. T., & Mak, C. M. (2014). Determination of single-sided ventilation rates in
multistory buildings: Evaluation of methods. Energy and Buildings, 69(0),
292-300. doi:http://dx.doi.org/10.1016/j.enbuild.2013.11.014
Allocca, C., Chen, Q., & Glicksman, L. R. (2003). Design analysis of single-sided
natural ventilation. Energy and Buildings, 35(8), 785-795.
doi:http://dx.doi.org/10.1016/S0378-7788(02)00239-6
Anderson, J. D. (1995). Computational fluid dynamics (Vol. 206): Springer.
Arens, E. A., Gonzalez, R., & Berglund, L. (1986). Thermal comfort under an
extended range of environmental conditions. ASHRAE Transactions, 92.
Ary, D., Jacobs, L. C., Irvine, C. K. S., & Walker, D. (2013). Introduction to research
in education: Cengage Learning.
198 Bibliography
Asfour, O. S., & Gadi, M. B. (2007). A comparison between CFD and network models
for predicting wind-driven ventilation in buildings. Building and environment,
42(12), 4079-4085.
ASHRAE. (2013). Standard 55-2013, Thermal Environmental Conditions for Human
Occupancy. American Society of Heating, Refrigerating, and Air-Conditioning
Engineers, Inc., Atlanta, Ga.
ASHRAE, A. (2010). ASHRAE 62.1-2010 Ventilation for Acceptable Indoor Air
Quality. Atlanta, GA.
ASHRAE Fundamentals Handbook. (2009). American society of heating,
refrigerating and air-conditioning engineers.
Australian Building Codes. (2011). National Construction Code Series: ABCB.
Australian Bureau of Statistics. Canberra, Australia.
Australian Bureau of Statistics. (2004). Canberra, Australia.
Australian Government Bureau of Meteorology. (2016). Retrieved from
http://www.bom.gov.au/
Australian Standard. (2012). AS 1668.4-2012 : The use of ventilation and
airconditioning in buildings Part 4: Natural ventilation of buildings: SAI
Global.
Awbi, H. B. (1994). Design considerations for naturally ventilated buildings.
Renewable energy, 5(5–8), 1081-1090. doi:http://dx.doi.org/10.1016/0960-
1481(94)90135-X
Awbi, H. B. (1996). Air movement in naturally-ventilated buildings. Renewable
energy, 8(1), 241-247.
Awbi, H. B. (2003). Ventilation of buildings: Taylor & Francis.
Axley, J. (2007). Multizone airflow modeling in buildings: History and theory.
HVAC&R Research, 13(6), 907-928.
Aynsley, R. (2007). Natural ventilation in passive design. BEDP Environment Design
Guide Tec.
Bibliography 199
Bailey, B. (2000). Constraints, limitations and achievements in greenhouse natural
ventilation. Paper presented at the International Conference and British-Israeli
Workshop on Greenhouse Techniques towards the 3rd Millennium 534.
Bangalee, M. Z. I., Lin, S. Y., & Miau, J. J. (2012). Wind driven natural ventilation
through multiple windows of a building: A computational approach. Energy
and Buildings, 45(0), 317-325.
doi:http://dx.doi.org/10.1016/j.enbuild.2011.11.025
Blocken, B., Defraeye, T., Derome, D., & Carmeliet, J. (2009). High-resolution CFD
simulations for forced convective heat transfer coefficients at the facade of a
low-rise building. Building and environment, 44(12), 2396-2412.
doi:http://dx.doi.org/10.1016/j.buildenv.2009.04.004
Blocken, B., & Gualtieri, C. (2012). Ten iterative steps for model development and
evaluation applied to Computational Fluid Dynamics for Environmental Fluid
Mechanics. Environmental Modelling & Software, 33, 1-22.
doi:http://dx.doi.org/10.1016/j.envsoft.2012.02.001
Brager, G. S., & De Dear, R. J. (1998). Thermal adaptation in the built environment:
a literature review. Energy and Buildings, 27(1), 83-96.
BS 5925: Code of practice for ventilation principles and designing for natural
ventilation. (1991). British Standards Institute.
Buys, L., Summerville, J., Bell, L., & Kennedy, R. (2008). „Exploring the Social
Impacts of High-Density Living: Considering the Residents‟ Perspective.‟ In.
Proceedings of Subtropical Cities 2008: From fault-lines to sight-lines
Subtropical urbanism in 20-20, 3–6 September 2008.
Caciolo, M., Cui, S., Stabat, P., & Marchio, D. (2013). Development of a new
correlation for single-sided natural ventilation adapted to leeward conditions.
Energy and Buildings, 60, 372-382.
Cao, B., Ouyang, Q., Zhu, Y., Huang, L., Hu, H., & Deng, G. (2012). Development of
a multivariate regression model for overall satisfaction in public buildings
based on field studies in Beijing and Shanghai. Building and environment, 47,
394-399. doi:http://dx.doi.org/10.1016/j.buildenv.2011.06.022
Carrilho da Graça, G., Chen, Q., Glicksman, L. R., & Norford, L. K. (2002).
Simulation of wind-driven ventilative cooling systems for an apartment
building in Beijing and Shanghai. Energy and Buildings, 34(1), 1-11.
doi:http://dx.doi.org/10.1016/S0378-7788(01)00083-4
200 Bibliography
Chand, I., Bhargava, P., & Krishak, N. (1998). Effect of balconies on ventilation
inducing aeromotive force on low-rise buildings. Building and environment,
33(6), 385-396.
Chen, Q. (2004). Using computational tools to factor wind into architectural
environment design. Energy and Buildings, 36(12), 1197-1209.
doi:http://dx.doi.org/10.1016/j.enbuild.2003.10.013
Chen, Q. (2009). Ventilation performance prediction for buildings: A method
overview and recent applications. Building and environment, 44(4), 848-858.
doi:http://dx.doi.org/10.1016/j.buildenv.2008.05.025
Chen, Q., Lee, K., Mazumdar, S., Poussou, S., Wang, L., Wang, M., & Zhang, Z.
(2010). Ventilation performance prediction for buildings: Model assessment.
Building and environment, 45(2), 295-303.
doi:http://dx.doi.org/10.1016/j.buildenv.2009.06.008
Cheung, C. K., Fuller, R. J., & Luther, M. B. (2005). Energy-efficient envelope design
for high-rise apartments. Energy and Buildings, 37(1), 37-48.
doi:http://dx.doi.org/10.1016/j.enbuild.2004.05.002
Chiang, W.-H., & Anh, N. D. (2012). NATURAL VENTILATION INSIDE
COURTYARD-APARTMENT BUILDING IN TAIWAN. Paper presented at the
Fourth German-Austrian IBPSA Conference, Berlin.
Chu, C.-R., & Chiang, B.-F. (2013). Wind-driven cross ventilation with internal
obstacles. Energy and Buildings, 67(0), 201-209.
doi:http://dx.doi.org/10.1016/j.enbuild.2013.07.086
Chu, C.-R., & Chiang, B.-F. (2014). Wind-driven cross ventilation in long buildings.
Building and environment, 80, 150-158.
doi:http://dx.doi.org/10.1016/j.buildenv.2014.05.017
Chu, C.-R., Chiu, Y.-H., Tsai, Y.-T., & Wu, S.-L. (2015). Wind-driven natural
ventilation for buildings with two openings on the same external wall. Energy
and Buildings, 108, 365-372.
Chu, C.-R., Chiu, Y. H., Tsai, Y.-T., & Wu, S.-L. (2015). Wind-driven natural
ventilation for buildings with two openings on the same external wall. Energy
and Buildings, 108, 365-372.
doi:http://dx.doi.org/10.1016/j.enbuild.2015.09.041
Bibliography 201
Chu, C. R., Chen, R.-H., & Chen, J.-W. (2011). A laboratory experiment of shear-
induced natural ventilation. Energy and Buildings, 43(10), 2631-2637.
doi:http://dx.doi.org/10.1016/j.enbuild.2011.06.014
Chu, C. R., Chiu, Y.-H., & Wang, Y.-W. (2010). An experimental study of wind-
driven cross ventilation in partitioned buildings. Energy and Buildings, 42(5),
667-673.
CIE (Centre for International Economics). (2007). Capitalising on the buildings
sectors potential to lessen the costs of a broad based GHG emissions cut:
prepared for ASBEC Climate Change Task.
Cooper, P., & Linden, P. (1996). Natural ventilation of an enclosure containing two
buoyancy sources. Journal of Fluid Mechanics, 311, 153-176.
Croome, D., Gan, G., & Awbi, H. (1993). Thermal comfort and air quality in offices.
Indoor air, 93, 37-42.
Crouch, C. (2012). Doing research in design (1 ed.). London: Berg Publishers.
Dalgliesh, W. A. (1975). Comparison of model/full-scale wind pressures on a high-
rise building. Journal of Wind Engineering and Industrial Aerodynamics, 1,
55-66. doi:http://dx.doi.org/10.1016/0167-6105(75)90006-9
Davenport, A. G. (1960). Rationale for determining design wind velocities. ASCE
Journal of the Structural Division, 86(5), 39-68.
De Dear, R. (2010). Thermal comfort in natural ventilation—a neurophysiological
hypothesis. Paper presented at the 2010 WINDSOR CONFERENCE:
ADAPTING TO CHANGE: NEW THINKING ON COMFORT.
De Dear, R., & Brager, G. S. (1998). Developing an adaptive model of thermal comfort
and preference.
De Dear, R., & Schiller Brager, G. (2001). The adaptive model of thermal comfort and
energy conservation in the built environment. International journal of
biometeorology, 45(2), 100-108.
De Dear, R. J. (1998). A global database of thermal comfort field experiments.
ASHRAE Transactions, 104, 1141.
202 Bibliography
De Gids, W., & Phaff, H. (1982). Ventilation rates and energy consumption due to
open windows: a brief overview of research in the Netherlands. Air infiltration
review, 4(1), 4-5.
Derakhshan, S., & Shaker, A. (2017). Numerical study of the cross-ventilation of an
isolated building with different opening aspect ratios and locations for various
wind directions. International Journal of Ventilation, 16(1), 42-60.
doi:http://dx.doi.org/10.1080/14733315.2016.1252146
Ding, W., Hasemi, Y., & Yamada, T. (2005). Natural ventilation performance of a
double-skin façade with a solar chimney. Energy and Buildings, 37(4), 411-
418. doi:http://dx.doi.org/10.1016/j.enbuild.2004.08.002
Ernest, D. R., Bauman, F. S., & Arens, E. A. (1992). The effects of external wind
pressure distributions on wind-induced air motion inside buildings. Journal of
Wind Engineering and Industrial Aerodynamics, 44(1), 2539-2550.
doi:http://dx.doi.org/10.1016/0167-6105(92)90045-C
Escombe, A. R., Oeser, C. C., Gilman, R. H., Navincopa, M., Ticona, E., Pan, W., . . .
Moore, D. A. (2007). Natural ventilation for the prevention of airborne
contagion. PLoS medicine, 4(2), e68.
Etheridge, D. (2011). Natural ventilation of buildings: theory, measurement and
design: John Wiley & Sons.
Evola, G., & Popov, V. (2006). Computational analysis of wind driven natural
ventilation in buildings. Energy and Buildings, 38(5), 491-501.
doi:http://dx.doi.org/10.1016/j.enbuild.2005.08.008
Fanger, P. O. (1970). Thermal comfort. Analysis and applications in environmental
engineering.
Fanger, P. O., & Toftum, J. (2002). Extension of the PMV model to non-air-
conditioned buildings in warm climates. Energy and Buildings, 34(6), 533-536.
Feustel, H. E. (1999). COMIS—an international multizone air-flow and contaminant
transport model. Energy and Buildings, 30(1), 3-18.
doi:http://dx.doi.org/10.1016/S0378-7788(98)00043-7
Feustel, H. E., & Smith, B. V. (1997). COMIS 3.0 user’s guide. Berkeley, CA:
Lawrence Berkeley National Laboratory.
Bibliography 203
Fisk, W. J., & Rosenfeld, A. H. (1997). Estimates of improved productivity and health
from better indoor environments. Indoor air, 7(3), 158-172.
Fluent, A. Release 15.0, Help System, Theory Guide, ANSYS: Inc.
Fluent, A. (2016). Release 17.0 (Version 17.0). PA 15317: ANSYS, Inc.
Fountain, M., Arens, E., De Dear, R., Bauman, F., & Miura, K. (1994). Locally
controlled air movement preferred in warm isothermal environments.
Fountain, M., & Arens, E. A. (1993). Air movement and thermal comfort. ASHRAE
journal, 35(8).
Fountain, M., & Huizenga, C. (1996). WinComf: A Windows 3.1 Thermal Sensation
Model—User’s Manual. Berkeley: Environmental Analytics.
Franke, J., Hellsten, A., Schlünzen, H., & Carissimo, B. (2007). Best Practice
Guideline for the CFD Simulation of Flows in the Urban Environment. COST
Action 732: Quality Assurance and Improvement of Microscale
Meteorological Models (2007). Hamburg, Germany.
Frontczak, M., & Wargocki, P. (2011). Literature survey on how different factors
influence human comfort in indoor environments. Building and environment,
46(4), 922-937.
Fung, Y., & Lee, W. (2014). Identifying the most influential parameter affecting
natural ventilation performance in high-rise high-density residential buildings.
Indoor and Built Environment, 1420326X14536189.
Gagge, A. P., Fobelets, A., & Berglund, L. (1986). A standard predictive index of
human response to the thermal environment. ASHRAE Trans.;(United States),
92.
Gan, G. (2000). Effective depth of fresh air distribution in rooms with single-sided
natural ventilation. Energy and Buildings, 31(1), 65-73.
Gao, C. (2011). The study of natural ventilation in residential buildings. The Hong
Kong Polytechnic University.
Gao, C., & Lee, W. (2011a). Evaluating the influence of openings configuration on
natural ventilation performance of residential units in Hong Kong. Building
and environment, 46(4), 961-969.
doi:https://doi.org/10.1016/j.buildenv.2010.10.029
204 Bibliography
Gao, C., & Lee, W. (2011b). Evaluating the influence of window types on the natural
ventilation performance of residential buildings in Hong Kong. International
Journal of Ventilation, 10(3), 227-238.
doi:http://dx.doi.org/10.1080/14733315.2011.11683951
Geller, H., Harrington, P., Rosenfeld, A. H., Tanishima, S., & Unander, F. (2006).
Polices for increasing energy efficiency: Thirty years of experience in OECD
countries. Energy Policy, 34(5), 556-573.
doi:http://dx.doi.org/10.1016/j.enpol.2005.11.010
Givoni, B. (1962). Basic study of Ventilation problems in housing in hot countries.
Final Report, Building Research Station, Technion-Israel Institute of
Technology, Haifa.
Givoni, B. (1968). Ventilation problems in hot countries: Israel Institute of
Technology.
Givoni, B. (1969). Man, Climate and Architecture (H. J. Cowan Ed.). London: Elsevier
Publishing Company.
Gratia, E., & De Herde, A. (2004a). Is day natural ventilation still possible in office
buildings with a double-skin façade? Building and environment, 39(4), 399-
409. doi:http://dx.doi.org/10.1016/j.buildenv.2003.10.006
Gratia, E., & De Herde, A. (2004b). Natural cooling strategies efficiency in an office
building with a double-skin façade. Energy and Buildings, 36(11), 1139-1152.
doi:http://dx.doi.org/10.1016/j.enbuild.2004.05.004
Groat, L. N., & Wang, D. (2013). Architectural research methods (Second;2.
Aufl.;2;2nd; ed.). Hoboken, New Jersey: Wiley.
Günel, M. H., & Ilgin, H. E. (2014). Tall Buildings: Structural Systems and
Aerodynamic Form: Routledge.
Hassan, M. A., Guirguis, N. M., Shaalan, M. R., & El-Shazly, K. M. (2007).
Investigation of effects of window combinations on ventilation characteristics
for thermal comfort in buildings. Desalination, 209(1), 251-260.
doi:http://dx.doi.org/10.1016/j.desal.2007.04.035
Heiselberg, P., Svidt, K., & Nielsen, P. V. (2001). Characteristics of airflow from open
windows. Building and environment, 36(7), 859-869.
doi:http://dx.doi.org/10.1016/S0360-1323(01)00012-9
Bibliography 205
Hitchin, E. R., & Wilson, C. B. (1967). A review of experimental techniques for the
investigation of natural ventilation in buildings. Building Science, 2(1), 59-82.
doi:http://dx.doi.org/10.1016/0007-3628(67)90007-2
Hooff, T. v., & Blocken, B. (2010). On the effect of wind direction and urban
surroundings on natural ventilation of a large semi-enclosed stadium.
Computers & Fluids, 39(7), 1146-1155.
doi:http://dx.doi.org/10.1016/j.compfluid.2010.02.004
Horan, J. M., & Finn, D. P. (2008). Sensitivity of air change rates in a naturally
ventilated atrium space subject to variations in external wind speed and
direction. Energy and Buildings, 40(8), 1577-1585.
doi:http://dx.doi.org/10.1016/j.enbuild.2008.02.013
Hoyt, T., Schiavon, S., Piccioli, A., Moon, D., & Steinfeld, K. (2013). CBE thermal
comfort tool. Center for the Built Environment, University of California
Berkeley, http://cbe. berkeley. edu/comforttool.
Hu, C.-H., Ohba, M., & Yoshie, R. (2008). CFD modelling of unsteady cross
ventilation flows using LES. Journal of Wind Engineering and Industrial
Aerodynamics, 96(10), 1692-1706.
Humphreys, M. (1978). Outdoor temperatures and comfort indoors. Batiment
International, Building Research and Practice, 6(2), 92-92.
Hunt, G. R., & Linden, P. P. (1999). The fluid mechanics of natural ventilation—
displacement ventilation by buoyancy-driven flows assisted by wind. Building
and environment, 34(6), 707-720. doi:http://dx.doi.org/10.1016/S0360-
1323(98)00053-5
IBC, I. (2006). International building code. International Code Council, Inc.(formerly
BOCA, ICBO and SBCCI), 4051, 60478-65795.
. ICEM CFD (Version 17.0). (2016). PA 15317: ANSYS, Inc.
IECC, A. (2012). International Energy Conservation Code 2012. C403, 2.
International Energy Agency. (2015). Key world energy statistics: International
Energy Agency (IEA).
Irwin, P., Kilpatrick, J., Robinson, J., & Frisque, A. (2008). Wind and tall buildings:
negatives and positives. The structural design of tall and special buildings,
17(5), 915-928.
206 Bibliography
Jiang, Y., Alexander, D., Jenkins, H., Arthur, R., & Chen, Q. (2003). Natural
ventilation in buildings: measurement in a wind tunnel and numerical
simulation with large-eddy simulation. Journal of Wind Engineering and
Industrial Aerodynamics, 91(3), 331-353.
doi:http://dx.doi.org/10.1016/S0167-6105(02)00380-X
Jiang, Y., & Chen, Q. (2001). Study of natural ventilation in buildings by large eddy
simulation. Journal of Wind Engineering and Industrial Aerodynamics, 89(13),
1155-1178.
Jiang, Y., & Chen, Q. (2002). Effect of fluctuating wind direction on cross natural
ventilation in buildings from large eddy simulation. Building and environment,
37(4), 379-386. doi:http://dx.doi.org/10.1016/S0360-1323(01)00036-1
Jiang, Y., & Chen, Q. (2003). Buoyancy-driven single-sided natural ventilation in
buildings with large openings. International Journal of Heat and Mass
Transfer, 46(6), 973-988.
Jin, R., Hang, J., Liu, S., Wei, J., Liu, Y., Xie, J., & Sandberg, M. (2015). Numerical
investigation of wind-driven natural ventilation performance in a multi-storey
hospital by coupling indoor and outdoor airflow. Indoor and Built
Environment, 25(8), 1226-1247. doi:10.1177/1420326X15595689
Jiru, T. E., & Bitsuamlak, G. T. (2010). Application of CFD in modelling wind-
induced natural ventilation of buildings-a review. International Journal of
Ventilation, 9(2), 131-147.
Johnson, M.-H., Zhai, Z., & Krarti, M. (2012). Performance evaluation of network
airflow models for natural ventilation. HVAC&R Research, 18(3), 349-365.
Karava, P., Stathopoulos, T., & Athienitis, A. (2011). Airflow assessment in cross-
ventilated buildings with operable façade elements. Building and environment,
46(1), 266-279.
Katayama, T., Tsutsumi, J., & Ishii, A. (1992). Full-scale measurements and wind
tunnel tests on cross-ventilation. Journal of Wind Engineering and Industrial
Aerodynamics, 44(1), 2553-2562.
Katz, J. (2010). Introductory fluid mechanics: Cambridge University Press.
Kawamura, S., Kimoto, E., Fukushima, T., & Taniike, Y. (1988). Environmental wind
characteristics around the base of a tall building-a comparison between model
Bibliography 207
test and full scale experiment. Journal of Wind Engineering and Industrial
Aerodynamics, 28(1), 149-158.
Kennedy, R., Buys, L., & Miller, E. (2015). Residents’ Experiences of Privacy and
Comfort in Multi-Storey Apartment Dwellings in Subtropical Brisbane.
Sustainability, 7(6), 7741-7761.
Kennedy, R. J., & Buys, L. (2010). Dimensions of liveability: a tool for sustainable
cities. Paper presented at the Proceedings of SB10mad Sustainable Building
Conference.
Kolokotroni, M., & Aronis, A. (1999). Cooling-energy reduction in air-conditioned
offices by using night ventilation. Applied Energy, 63(4), 241-253.
doi:http://dx.doi.org/10.1016/S0306-2619(99)00031-8
Kotani, H., Narasaki, M., Sato, R., & Yamanaka, T. (1996). Natural ventilation caused
by stack effect in large courtyard of high-rise building. Paper presented at the
Proceedings of ROOMVENT.
Kusher, C. (2016). High-rise approvals are booming. Retrieved from
https://www.corelogic.com.au/news/high-rise-approvals-are-booming
Kwon, B., & Park, Y. (2013). Interior noise control with an active window system.
Applied Acoustics, 74(5), 647-652.
doi:http://dx.doi.org/10.1016/j.apacoust.2012.11.005
Lakehal, D., & Rodi, W. (1997). Calculation of the flow past a surface-mounted cube
with two-layer turbulence models. Journal of Wind Engineering and Industrial
Aerodynamics, 67, 65-78.
Larsen, T. S., & Heiselberg, P. (2008). Single-sided natural ventilation driven by wind
pressure and temperature difference. Energy and Buildings, 40(6), 1031-1040.
Le, D., Li, N., & Su, L. STUDY ON WIND ENVIRONMENT CHARACTERISTIC
OF DIFFERENT HEIGHT HIGH-RISE RESIDENTIAL BUILDINGS IN
CHANGSHA, CHINA.
Li, Y., & Delsante, A. (2001). Natural ventilation induced by combined wind and
thermal forces. Building and environment, 36(1), 59-71.
doi:http://dx.doi.org/10.1016/S0360-1323(99)00070-0
208 Bibliography
Liddament, M., Axley, J., Heiselberg, P., Li, Y., & Stathopoulos, T. (2006). Achieving
natural and hybrid ventilation in practice. International Journal of Ventilation,
5(1), 115-130.
Liddament, M. W. (1996). A guide to energy efficient ventilation: Air Infiltration and
Ventilation Centre Conventry, UK.
Linden, P., Lane-Serff, G., & Smeed, D. (1990). Emptying filling boxes: the fluid
mechanics of natural ventilation. Journal of Fluid Mechanics, 212, 309-335.
Linden, P. F. (1999). The fluid mechanics of natural ventilation. Annual review of fluid
mechanics, 31(1), 201-238.
Liping, W., & Hien, W. N. (2007). Applying natural ventilation for thermal comfort
in residential buildings in Singapore. Architectural Science Review, 50(3), 224-
233.
Liu, P.-C., Ford, B., & Etheridge, D. (2012). A modelling study of segmentation of
naturally ventilated tall office buildings in a hot and humid climate.
International Journal of Ventilation, 11(1), 29-42.
Liu, X., Niu, J., Perino, M., & Heiselberg, P. (2008). Numerical simulation of inter-
flat air cross-contamination under the condition of single-sided natural
ventilation. Journal of Building Performance Simulation, 1(2), 133-147.
Lo, L. J., Banks, D., & Novoselac, A. (2012). Combined wind tunnel and CFD analysis
for indoor airflow prediction of wind-driven cross ventilation. Building and
environment, 60, 12-23.
Lo, L. J., & Novoselac, A. (2012). Cross ventilation with small openings:
Measurements in a multi-zone test building. Building and environment, 57,
377-386.
Lukkunaprasit, P., Ruangrassamee, A., & Thanasisathit, N. (2009). Tsunami loading
on buildings with openings. Science of Tsunami Hazards, 28(5), 303.
Luo, Z., Zhao, J., Gao, J., & He, L. (2007). Estimating natural-ventilation potential
considering both thermal comfort and IAQ issues. Building and environment,
42(6), 2289-2298.
Mak, C., Niu, J., Lee, C., & Chan, K. (2007). A numerical simulation of wing walls
using computational fluid dynamics. Energy and Buildings, 39(9), 995-1002.
Bibliography 209
Malkawi, A., & Augenbroe, G. (2004). Advanced building simulation: Routledge.
Manual, L. U. (1998). National Instruments. Austin, TX.
Marino, C., Nucara, A., & Pietrafesa, M. (2012). Proposal of comfort classification
indexes suitable for both single environments and whole buildings. Building
and environment, 57, 58-67.
doi:http://dx.doi.org/10.1016/j.buildenv.2012.04.012
MAXIM, O. U. s. G. (2009). Version 1.4. AN3358, sep.
Melikov, A. K., Popiolek, Z., Silva, M., Care, I., & Sefker, T. (2007). Accuracy
limitations for low-velocity measurements and draft assessment in rooms.
HVAC&R Research, 13(6), 971-986.
Mendell, M. J., Fisk, W. J., Deddens, J. A., Seavey, W. G., Smith, A. H., Smith, D. F.,
. . . Goldman, L. R. (1996). Elevated symptom prevalence associated with
ventilation type in office buildings. Epidemiology, 7(6), 583-589.
Miller, W., & Buys, L. (2012). Anatomy of a sub-tropical Positive Energy Home
(PEH). Solar Energy, 86(1), 231-241.
Miller, W. F., & Nazari, H. S. (2013). Occupant comfort, the housing industry and
electricity infrastructure: understanding the synergies. Paper presented at the
Book of Proceedings of the 5th International Urban Design Conference: Peer
Reviewed Papers.
Mohsen, E. A., & Oldham, D. J. (1977). Traffic noise reduction due to the screening
effect of balconies on a building façade. Applied Acoustics, 10(4), 243-257.
doi:http://dx.doi.org/10.1016/0003-682X(77)90010-X
Montazeri, H., & Blocken, B. (2013). CFD simulation of wind-induced pressure
coefficients on buildings with and without balconies: validation and sensitivity
analysis. Building and environment, 60, 137-149.
National Instruments. (2017). Acceleration Data Logger. Retrieved from
http://sine.ni.com/nips/cds/pages/image?imagepath=/images/products/us/usb-
9162_ni-9234_l.jpg&title=Acceleration%20Data%20Logger&oracleLang=us
Negrao, C. O. (1995). Conflation of computational fluid dynamics and building
thermal simulation. University of Strathclyde Glasgow.
210 Bibliography
Nikas, K. S., Nikolopoulos, N., & Nikolopoulos, A. (2010). Numerical study of a
naturally cross-ventilated building. Energy and Buildings, 42(4), 422-434.
Niu, J. (2004). Some significant environmental issues in high-rise residential building
design in urban areas. Energy and Buildings, 36(12), 1259-1263.
doi:http://dx.doi.org/10.1016/j.enbuild.2003.07.005
Norton, T., Grant, J., Fallon, R., & Sun, D.-W. (2009). Assessing the ventilation
effectiveness of naturally ventilated livestock buildings under wind dominated
conditions using computational fluid dynamics. Biosystems Engineering,
103(1), 78-99. doi:http://dx.doi.org/10.1016/j.biosystemseng.2009.02.007
O'Sullivan, P. D., & Kolokotroni, M. (2017). A field study of wind dominant single
sided ventilation through a narrow slotted architectural louvre system. Energy
and Buildings, 138, 733-747.
doi:http://dx.doi.org/10.1016/j.enbuild.2016.11.025
Ohba, M., & Lun, I. (2010). Overview of natural cross-ventilation studies and the latest
simulation design tools used in building ventilation-related research. Advances
in Building Energy Research, 4(1), 127-166.
Omrani, S., Capra, B., Garcia-Hansen, V., & Drogemuller, R. (2015). Investigation of
the effect of balconies on natural ventilation of dwellings in high-rise
residential buildings in subtropical climate. Paper presented at the Living and
Learning: Research for a Better Built Environment: 49th International
Conference of the Architectural Science Association 2015, Melbourne,
Australia.
Omrani, S., Drogemuller, R., Garcia-Hansen, V., & Capra, B. (2014). Natural
ventilation heuristics in high-rise residential buildings: evaluation and
prediction. Paper presented at the Across: Architectural Research through to
Practice: 48th International Conference of the Architectural Science
Association (ANZAScA) 2014.
Omrani, S., Garcia-Hansen, V., Capra, B., & Drogemuller, R. (2017a). Natural
ventilation in multi-storey buildings: Design process and review of evaluation
tools. Building and environment, 116, 182-194.
doi:http://dx.doi.org/10.1016/j.buildenv.2017.02.012
Omrani, S., Garcia-Hansen, V., Capra, B. R., & Drogemuller, R. (2017b). Effect of
natural ventilation mode on thermal comfort and ventilation performance: Full-
scale measurement. Energy and Buildings, 156(Supplement C), 1-16.
doi:https://doi.org/10.1016/j.enbuild.2017.09.061
Bibliography 211
Omrani, S., Garcia-Hansen, V., Capra, B. R., & Drogemuller, R. (2017c). On the effect
of provision of balconies on natural ventilation and thermal comfort in high-
rise residential buildings. Building and environment, 123, 504-516.
doi:http://dx.doi.org/10.1016/j.buildenv.2017.07.016
Omrani, S., Garcia-Hansen, V., Drogemuller, R., & Capra, B. (2016a). Predicting
environmental conditions at building site for Natural ventilation design:
Correlation of meteorological data to air speed at building openings. Paper
presented at the Fifty years later: Revisiting the role of architectural science in
design and practice: 50th International Conference of the Architectural Science
Association 2016, Adelaide, Australia.
Omrani, S., Garcia-Hansen, V., Drogemuller, R., & Capra, B. (2016b). Thermal
comfort evaluation of natural ventilation mode: case study of a high-rise
residential building. Paper presented at the Fifty years later: Revisiting the role
of architectural science in design and practice: 50th International Conference
of the Architectural Science Association 2016, Adelaide, Australia.
Orme, M. (2001). Estimates of the energy impact of ventilation and associated
financial expenditures. Energy and Buildings, 33(3), 199-205.
doi:http://dx.doi.org/10.1016/S0378-7788(00)00082-7
Oropeza-Perez, I., & Østergaard, P. A. (2014). Energy saving potential of utilizing
natural ventilation under warm conditions – A case study of Mexico. Applied
Energy, 130(Supplement C), 20-32.
doi:https://doi.org/10.1016/j.apenergy.2014.05.035
Orszag, S. A., Yakhot, V., Flannery, W. S., Boysan, F., Choudhury, D., Maruzewski,
J., & Patel, B. (1993). Renormalization group modeling and turbulence
simulations. Near-wall turbulent flows, 1031-1046.
Pachauri, R. K., Allen, M., Barros, V., Broome, J., Cramer, W., Christ, R., . . .
Dasgupta, P. (2014). Climate Change 2014: Synthesis Report. Contribution of
Working Groups I, II and III to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change.
Papakonstantinou, K. A., Kiranoudis, C. T., & Markatos, N. C. (2000). Numerical
simulation of air flow field in single-sided ventilated buildings. Energy and
Buildings, 33(1), 41-48. doi:http://dx.doi.org/10.1016/S0378-7788(00)00063-
3
Park, J. S. (2013). Long-term field measurement on effects of wind speed and
directional fluctuation on wind-driven cross ventilation in a mock-up building.
Building and environment, 62, 1-8.
doi:http://dx.doi.org/10.1016/j.buildenv.2012.12.013
212 Bibliography
Pasut, W., & De Carli, M. (2012). Evaluation of various CFD modelling strategies in
predicting airflow and temperature in a naturally ventilated double skin façade.
Applied thermal engineering, 37, 267-274.
doi:http://dx.doi.org/10.1016/j.applthermaleng.2011.11.028
Patton, M. Q. (2002). Qualitative research and evaluation methods.
Pears, A. (2007). Residential Sector Energy Efficiency Scenario—Background,
framework and rationales: Melbourne.
Perén, J., Van Hooff, T., Leite, B., & Blocken, B. (2015). CFD analysis of cross-
ventilation of a generic isolated building with asymmetric opening positions:
impact of roof angle and opening location. Building and environment, 85, 263-
276.
Perén, J. I., van Hooff, T., Leite, B. C. C., & Blocken, B. (2015). Impact of eaves on
cross-ventilation of a generic isolated leeward sawtooth roof building:
Windward eaves, leeward eaves and eaves inclination. Building and
environment, 92, 578-590.
doi:http://dx.doi.org/10.1016/j.buildenv.2015.05.011
Pérez-Lombard, L., Ortiz, J., & Pout, C. (2008). A review on buildings energy
consumption information. Energy and Buildings, 40(3), 394-398.
doi:http://dx.doi.org/10.1016/j.enbuild.2007.03.007
Prajongsan, P., & Sharples, S. (2012). Enhancing natural ventilation, thermal comfort
and energy savings in high-rise residential buildings in Bangkok through the
use of ventilation shafts. Building and environment, 50, 104-113.
Prianto, E., & Depecker, P. (2002). Characteristic of airflow as the effect of balcony,
opening design and internal division on indoor velocity: A case study of
traditional dwelling in urban living quarter in tropical humid region. Energy
and Buildings, 34(4), 401-409.
Prianto, E., & Depecker, P. (2003). Optimization of architectural design elements in
tropical humid region with thermal comfort approach. Energy and Buildings,
35(3), 273-280. doi:http://dx.doi.org/10.1016/S0378-7788(02)00089-0
Priyadarsini, R., Cheong, K. W., & Wong, N. H. (2004). Enhancement of natural
ventilation in high-rise residential buildings using stack system. Energy and
Buildings, 36(1), 61-71. doi:http://dx.doi.org/10.1016/S0378-7788(03)00076-
8
Bibliography 213
Ramponi, R., & Blocken, B. (2012). CFD simulation of cross-ventilation for a generic
isolated building: impact of computational parameters. Building and
environment, 53, 34-48.
Recast, E. (2010). Directive 2010/31/EU of the European Parliament and of the
Council of 19 May 2010 on the energy performance of buildings (recast).
Official Journal of the European Union, 18(06), 2010.
Roetzel, A., Tsangrassoulis, A., Dietrich, U., & Busching, S. (2010). A review of
occupant control on natural ventilation. Renewable and Sustainable Energy
Reviews, 14(3), 1001-1013. doi:http://dx.doi.org/10.1016/j.rser.2009.11.005
RP, A. (1997). Developing an Adaptive Model of Thermal Comfort and Preference.
Santamouris, M. (2016). Cooling the buildings – past, present and future. Energy and
Buildings, 128, 617-638. doi:http://dx.doi.org/10.1016/j.enbuild.2016.07.034
Santamouris, M., & Allard, F. (1998). Natural ventilation in buildings: a design
handbook: Earthscan.
Schaelin, A., Dorer, V., Maas, J. v. d., & Moser, A. (1993). Improvement of multizone
model predictions by detailed flow path values from CFD calculations.
ASHRAE Transactions-American Society of Heating Refrigerating
Airconditioning Engin, 99(2), 709-720.
Schulze, T., & Eicker, U. (2013). Controlled natural ventilation for energy efficient
buildings. Energy and Buildings, 56, 221-232.
doi:http://dx.doi.org/10.1016/j.enbuild.2012.07.044
Seifert, J., Li, Y., Axley, J., & Rösler, M. (2006). Calculation of wind-driven cross
ventilation in buildings with large openings. Journal of Wind Engineering and
Industrial Aerodynamics, 94(12), 925-947.
Shameri, M. A., Alghoul, M. A., Sopian, K., Zain, M. F. M., & Elayeb, O. (2011).
Perspectives of double skin façade systems in buildings and energy saving.
Renewable and Sustainable Energy Reviews, 15(3), 1468-1475.
doi:https://doi.org/10.1016/j.rser.2010.10.016
Shao, L., & Riffat, S. (1994). Tracer-gas mixing with air: effect of tracer species.
Applied Energy, 49(2), 197-211.
214 Bibliography
Shao, L., Sharples, S., & Ward, I. (1993). Tracer gas mixing with air. Building Services
Engineering Research and Technology, 14(2), 43-50.
Sherman, M. H. (1990). Tracer-gas techniques for measuring ventilation in a single
zone. Building and environment, 25(4), 365-374.
doi:http://dx.doi.org/10.1016/0360-1323(90)90010-O
Shetabivash, H. (2015). Investigation of opening position and shape on the natural
cross ventilation. Energy and Buildings, 93, 1-15.
doi:http://dx.doi.org/10.1016/j.enbuild.2014.12.053
Song, D., & Choi, Y.-J. (2012). Effect of building regulation on energy consumption
in residential buildings in Korea. Renewable and Sustainable Energy Reviews,
16(1), 1074-1081. doi:http://dx.doi.org/10.1016/j.rser.2011.10.008
Stake, R. E. (2008). Qualitative case studies.
Standard, I. (1994). 7730. Moderate thermal environments–Determination of the PMV
and PPD indices and specification of the conditions for thermal comfort.
Sun, Y., & Zhang, Y. (2007). An overview of room air motion measurement:
technology and application. HVAC&R Research, 13(6), 929-950.
Swanborn, P. (2010). Case study research: what, why and how (1 ed.): SAGE
Publications.
Szokolay, S. V. (2004). Introduction to architectural science: the basis of sustainable
design: Routledge.
Tan, G., & Glicksman, L. R. (2005). Application of integrating multi-zone model with
CFD simulation to natural ventilation prediction. Energy and Buildings,
37(10), 1049-1057. doi:http://dx.doi.org/10.1016/j.enbuild.2004.12.009
Tantasavasdi, C., Srebric, J., & Chen, Q. (2001). Natural ventilation design for houses
in Thailand. Energy and Buildings, 33(8), 815-824.
Team, R. C. (2014). The R project for statistical computing. Available at www. R-
project. org/. Accessed October, 31, 2014.
Tong, Z., Chen, Y., & Malkawi, A. (2016). Defining the Influence Region in
neighborhood-scale CFD simulations for natural ventilation design. Applied
Energy, 182, 625-633. doi:https://doi.org/10.1016/j.apenergy.2016.08.098
Bibliography 215
Tong, Z., Chen, Y., Malkawi, A., Adamkiewicz, G., & Spengler, J. D. (2016).
Quantifying the impact of traffic-related air pollution on the indoor air quality
of a naturally ventilated building. Environment International, 89, 138-146.
doi:http://dx.doi.org/10.1016/j.envint.2016.01.016
Truong, P. H. (2012). Recommendations for the analysis and design of naturally
ventilated buildings in urban areas. Massachusetts Institute of Technology.
Tsou, J.-Y. (2001). Strategy on applying computational fluid dynamic for building
performance evaluation. Automation in Construction, 10(3), 327-335.
doi:http://dx.doi.org/10.1016/S0926-5805(00)00049-2
van Hooff, T., Blocken, B., Aanen, L., & Bronsema, B. (2011). A venturi-shaped roof
for wind-induced natural ventilation of buildings: Wind tunnel and CFD
evaluation of different design configurations. Building and environment, 46(9),
1797-1807. doi:http://dx.doi.org/10.1016/j.buildenv.2011.02.009
Visagavel, K., & Srinivasan, P. S. S. (2009). Analysis of single side ventilated and
cross ventilated rooms by varying the width of the window opening using CFD.
Solar Energy, 83(1), 2-5. doi:http://dx.doi.org/10.1016/j.solener.2008.06.004
von Grabe, J., Svoboda, P., & Bäumler, A. (2014). Window ventilation efficiency in
the case of buoyancy ventilation. Energy and Buildings, 72, 203-211.
doi:http://dx.doi.org/10.1016/j.enbuild.2013.10.006
Walker, A. (2008). Natural ventilation. National Renewable Energy Laboratory, 6,
2640.
Walton, G. N., & Dols, W. S. (2005). CONTAM user guide and program
documentation.
Wang, H., Karava, P., & Chen, Q. (2015). Development of simple semiempirical
models for calculating airflow through hopper, awning, and casement windows
for single-sided natural ventilation. Energy and Buildings, 96, 373-384.
doi:http://dx.doi.org/10.1016/j.enbuild.2015.03.041
Wang, L., & Hien, W. N. (2007). The impacts of ventilation strategies and facade on
indoor thermal environment for naturally ventilated residential buildings in
Singapore. Building and environment, 42(12), 4006-4015.
Wang, L., Hien, W. N., & Li, S. (2007). Facade design optimization for naturally
ventilated residential buildings in Singapore. Energy and Buildings, 39(8),
954-961. doi:http://dx.doi.org/10.1016/j.enbuild.2006.10.011
216 Bibliography
Warren, P. (1977). Ventilation through openings on one wall only. Paper presented at
the Heat Transfer in Buildings, Proceedings of ICHMT seminar. Hemisphere.
New York, EE. UU.
Wong, N., Feriadi, H., Tham, K., Sekhar, C., & Cheong, K. (2000). Natural Ventilation
Characteristics of Courtyard Buildings in Singapore. Paper presented at the
RoomVent 2000-7th International Conference on Air Distribution in Rooms.
Wong, P., Prasad, D., & Behnia, M. (2008). A new type of double-skin façade
configuration for the hot and humid climate. Energy and Buildings, 40(10),
1941-1945.
Wood, A., & Salib, R. (2013). Guide to Natural Ventilation in High Rise Office
Buildings: Routledge.
Yik, F. W., & Lun, Y. F. (2010). Energy saving by utilizing natural ventilation in
public housing in Hong Kong. Indoor and Built Environment, 19(1), 73-87.
Yin, R. K. (2009). Case study research: design and methods (Fourth ed.). Los Angeles:
SAGE.
Yin, R. K. (2012). Applications of case study research (Third ed.): Sage.
Yin, W., Zhang, G., Yang, W., & Wang, X. (2010). Natural ventilation potential model
considering solution multiplicity, window opening percentage, air velocity and
humidity in China. Building and environment, 45(2), 338-344.
doi:http://dx.doi.org/10.1016/j.buildenv.2009.06.012
Zhai, Z., Chen, Q., Klems, J. H., & Haves, P. (2001). Strategies for coupling energy
simulation and computational fluid dynamics programs. Lawrence Berkeley
National Laboratory.
Zhai, Z. J., Zhang, Z., Zhang, W., & Chen, Q. Y. (2007). Evaluation of various
turbulence models in predicting airflow and turbulence in enclosed
environments by CFD: Part 1—Summary of prevalent turbulence models.
HVAC&R Research, 13(6), 853-870.
Zhou, C., Wang, Z., Chen, Q., Jiang, Y., & Pei, J. (2014). Design optimization and
field demonstration of natural ventilation for high-rise residential buildings.
Energy and Buildings, 82, 457-465.
Bibliography 217
Zhou, J., Ye, C., Hu, Y., Hemida, H., Zhang, G., & Yang, W. (2017). Development of
a model for single-sided, wind-driven natural ventilation in buildings. Building
Services Engineering Research and Technology, 0143624417699658.
Zhou, Y., Kijewski, T., & Kareem, A. (2003). Aerodynamic loads on tall buildings:
interactive database. Journal of structural engineering, 129(3), 394-404.