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.


• 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/

https://doi.org/10.1016/j.enbuild.2017.09.061

http://dx.doi.org/10.1016/j.buildenv.2017.07.016



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.


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.


• 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.





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


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.


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:


• 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


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


(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:


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 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


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.


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


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


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


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


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.


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


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


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


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


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.


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.


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,


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


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


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


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.


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).


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.


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.


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.

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.


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


(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.


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).


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”


(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.


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.


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.


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


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 (%

)


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


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.


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.


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


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


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.


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


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.


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


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


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


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


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.


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


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.


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.


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


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.


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.


(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).


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.).


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.


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.


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


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


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,


& 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


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


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


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

•


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


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


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


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.


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


Feasibility

Concept Design

Detail Design

Final Design

/Documentation

Construction

Design Stages Activities

After Construction


• 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


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.


Reference Problem

focus

Building

Type

Ventilation

Type Methodology

Climate/


(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.


Reference Problem

focus

Building

Type

Ventilation

Type Methodology

Climate/


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


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


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,


Network airflow model: COMIS

Heat transfer model: HTB2


Reference Problem

focus

Building

Type

Ventilation

Type Methodology

Climate/


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.


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


scale measurements.



CFD simulations.


Reference Problem

focus

Building

Type

Ventilation

Type Methodology

Climate/


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


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.


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


Reference Problem

focus

Building

Type

Ventilation

Type Methodology

Climate/


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.


Reference Problem

focus

Building

Type

Ventilation

Type Methodology

Climate/



CFD simulations.


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


scale measurements.



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.


Omrani,

Drogemuller et al.

Natural

ventilation

heuristics

High-rise

residential

buildings

Wind-driven

ventilation

CFD simulations

Brisbane,

Australia

(subtropical)

Velocity

magnitude


CFD simulations.



Reference Problem

focus

Building

Type

Ventilation

Type Methodology

Climate/


(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


CFD simulations.


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


Reference Problem

focus

Building

Type

Ventilation

Type Methodology

Climate/


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.

















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:


Sara Omrani Collected the experimental data, analysed

the data, produced the graphics, developed

the study, and wrote the manuscript.



Robin Drogemuller Assisted in developing the study, proof-

read and reviewed the manuscript.

Bianca Capra Assisted in developing the study.




__Veronica Garcia Hansen___ QUT Verified Signature______28/04/2017_____

Name Signature Date



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,



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.



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



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.



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.



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.



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



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



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.



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



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



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


















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



Bianca Capra Assisted in developing the study and

analysis, proof-read and reviewed the

manuscript.






__Veronica Garcia Hansen___ ______28/04/2017_____

Name Signature Date

QUT Verified Signature


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)


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, &


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


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


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


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


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


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).


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


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


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


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


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.


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.


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.


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.


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.


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.


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.


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


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


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


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


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


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).


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


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.


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


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


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


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,


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


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.

















144 Chapter 7: Thermal comfort evaluation of natural ventilation mode: case study of a high-rise residential

building




the data, produced the graphics, developed

the study, and wrote the manuscript.





Bianca Capra Assisted in developing the study.




__Veronica Garcia Hansen___ ____________________________28/04/2017_____

Name Signature Date


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


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.


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


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)


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


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.


7.3.1 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


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


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


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.


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


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


















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


graphics, developed the study, and wrote

the manuscript.



Bianca Capra Assisted in developing the study and

analysis, proof-read and reviewed the

manuscript.

Robin Drogemuller Assisted in developing the study.




__Veronica Garcia Hansen___ _QUT Verified Signature_____28/04/2017_____

Name Signature Date



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



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



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.



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).



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



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



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)



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



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).



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



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



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



(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).


8.3.1 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



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.



Figure 8.7. Results summary for both ventilation modes (A), cross ventilation (B), and single-sided

ventilation(C)


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



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.



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



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.



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.



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.


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,



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



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



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



• 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.



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.


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


Feasibility

Concept Design

Detail Design

Final Design

/Documentation

Construction

Design Stages Activities

After Construction


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


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


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


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.


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


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


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,


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


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


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.

http://dx.doi.org/10.1016/j.enbuild.2015.04.033



http://dx.doi.org/10.1016/S0378-7788(02)00239-6

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.

http://www.bom.gov.au/

http://dx.doi.org/10.1016/0960-1481(94)90135-X

http://dx.doi.org/10.1016/0960-1481(94)90135-X

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.


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.




http://dx.doi.org/10.1016/j.envsoft.2012.02.001


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.


Chen, Q. (2009). Ventilation performance prediction for buildings: A method

overview and recent applications. Building and environment, 44(4), 848-858.


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.


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.


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.


Chu, C.-R., & Chiang, B.-F. (2014). Wind-driven cross ventilation in long buildings.

Building and environment, 80, 150-158.


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










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.


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.


http://dx.doi.org/10.1016/0167-6105(75)90006-9

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.


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.


Feustel, H. E., & Smith, B. V. (1997). COMIS 3.0 user’s guide. Berkeley, CA:

Lawrence Berkeley National Laboratory.

http://dx.doi.org/10.1080/14733315.2016.1252146


http://dx.doi.org/10.1016/0167-6105(92)90045-C


http://dx.doi.org/10.1016/S0378-7788(98)00043-7

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

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.


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.


http://dx.doi.org/10.1080/14733315.2011.11683951

http://dx.doi.org/10.1016/j.enpol.2005.11.010



http://dx.doi.org/10.1016/j.desal.2007.04.035

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.


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.

http://dx.doi.org/10.1016/0007-3628(67)90007-2

http://dx.doi.org/10.1016/j.compfluid.2010.02.004


http://cbe/

http://dx.doi.org/10.1016/S0360-1323(98)00053-5

http://dx.doi.org/10.1016/S0360-1323(98)00053-5

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

http://dx.doi.org/10.1016/S0167-6105(02)00380-X

http://dx.doi.org/10.1016/S0360-1323(01)00036-1

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.


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.


http://dx.doi.org/10.1016/S0306-2619(99)00031-8

https://www.corelogic.com.au/news/high-rise-approvals-are-booming

http://dx.doi.org/10.1016/j.apacoust.2012.11.005

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.


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.


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.


http://dx.doi.org/10.1016/0003-682X(77)90010-X

http://sine.ni.com/nips/cds/pages/image?imagepath=/images/products/us/usb-9162_ni-9234_l.jpg&title=Acceleration%20Data%20Logger&oracleLang=us

http://sine.ni.com/nips/cds/pages/image?imagepath=/images/products/us/usb-9162_ni-9234_l.jpg&title=Acceleration%20Data%20Logger&oracleLang=us

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.


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



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.


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


http://dx.doi.org/10.1016/j.biosystemseng.2009.02.007



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.


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.


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.



http://dx.doi.org/10.1016/S0378-7788(00)00082-7

https://doi.org/10.1016/j.apenergy.2014.05.035

http://dx.doi.org/10.1016/S0378-7788(00)00063-3

http://dx.doi.org/10.1016/S0378-7788(00)00063-3


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



Pérez-Lombard, L., Ortiz, J., & Pout, C. (2008). A review on buildings energy

consumption information. Energy and Buildings, 40(3), 394-398.


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

http://dx.doi.org/10.1016/j.applthermaleng.2011.11.028



http://dx.doi.org/10.1016/S0378-7788(02)00089-0

http://dx.doi.org/10.1016/S0378-7788(03)00076-8

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


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.


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.

http://dx.doi.org/10.1016/j.rser.2009.11.005



https://doi.org/10.1016/j.rser.2010.10.016

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.


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

http://dx.doi.org/10.1016/0360-1323(90)90010-O


http://dx.doi.org/10.1016/j.rser.2011.10.008


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.


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.


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.


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.


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.


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.

NATURAL VENTILATION IN HIGH RISE APARTMENTS IN HOT … · 2018-03-05 · ii Natural Ventilation in...

Documents

Transcript of NATURAL VENTILATION IN HIGH RISE APARTMENTS IN HOT … · 2018-03-05 · ii Natural Ventilation in...